Compare commits

..

4 commits

Author SHA1 Message Date
Simon Willison
693f4ed601 Show insert message in correct place, 1s inactive delay
Since I want this to be usable by untrusted datasette-apps it is important that
they cannot show the dialog in a way that tricks the user into clicking insert.

So a 1s delay on making that button active.
2026-06-20 12:23:52 -07:00
Simon Willison
76e6a3bc38 New .insertDialog() JavaScript method
Anywhere in Datasette can now open an insert dialog
pre-populated with suggested data.

Using it in one place already, on the row page
to provide an insert button for related rows.
2026-06-19 23:40:15 -07:00
Simon Willison
f3fa6ecf61 Render error pages using datasette.render_template()
Avoids risk of important variables not being present.
2026-06-19 23:38:50 -07:00
Simon Willison
a76646e3fd GET /db/table/-/insert returns data needed for dialog
As part of making the insert dialog less dependent on the table page.
2026-06-19 22:51:21 -07:00
223 changed files with 6155 additions and 28340 deletions

View file

@ -1,39 +0,0 @@
name: "Setup SQLite version"
description: "Build and activate a specific SQLite version from its amalgamation archive"
inputs:
version:
description: "The SQLite version to install"
required: true
cflags:
description: "CFLAGS to use when compiling SQLite"
required: false
default: ""
skip-activate:
description: "Set to true to skip modifying the library path"
required: false
default: "false"
fallback-urls:
description: "Whitespace-separated fallback download URLs to try after sqlite.org"
required: false
default: ""
outputs:
sqlite-location:
description: "Directory containing the compiled SQLite library"
value: ${{ steps.build.outputs.sqlite-location }}
runs:
using: "composite"
steps:
- shell: bash
run: mkdir -p "$RUNNER_TEMP/sqlite-versions/downloads"
- uses: actions/cache@v6
with:
path: ${{ runner.temp }}/sqlite-versions/downloads
key: setup-sqlite-version-${{ inputs.version }}-amalgamation-v1
- id: build
shell: bash
run: bash "$GITHUB_ACTION_PATH/setup-sqlite-version.sh"
env:
SQLITE_VERSION: ${{ inputs.version }}
SQLITE_CFLAGS: ${{ inputs.cflags }}
SQLITE_SKIP_ACTIVATE: ${{ inputs.skip-activate }}
SQLITE_EXTRA_FALLBACK_URLS: ${{ inputs.fallback-urls }}

View file

@ -1,144 +0,0 @@
#!/usr/bin/env bash
set -euo pipefail
version_spec="${SQLITE_VERSION:?SQLITE_VERSION is required}"
cflags="${SQLITE_CFLAGS:-}"
skip_activate="${SQLITE_SKIP_ACTIVATE:-false}"
extra_fallback_urls="${SQLITE_EXTRA_FALLBACK_URLS:-}"
case "$version_spec" in
3.46 | 3.46.0)
sqlite_version="3.46.0"
sqlite_year="2024"
amalgamation_id="3460000"
builtin_fallback_urls="https://static.simonwillison.net/static/2026/sqlite-amalgamation-3460000.zip"
;;
3.25 | 3.25.0)
sqlite_version="3.25.0"
sqlite_year="2018"
amalgamation_id="3250000"
builtin_fallback_urls="https://static.simonwillison.net/static/2026/sqlite-amalgamation-3250000.zip?v=1"
;;
*)
echo "::error::Unsupported SQLite version '$version_spec'. Add its release year and amalgamation id to $GITHUB_ACTION_PATH/setup-sqlite-version.sh."
exit 1
;;
esac
case "$(uname -s)" in
Linux)
library_name="libsqlite3.so.0"
library_path_var="LD_LIBRARY_PATH"
;;
Darwin)
library_name="libsqlite3.dylib"
library_path_var="DYLD_LIBRARY_PATH"
;;
*)
echo "::error::Unsupported platform $(uname -s)"
exit 1
;;
esac
runner_temp="${RUNNER_TEMP:-}"
if [ -z "$runner_temp" ]; then
runner_temp="$(mktemp -d)"
fi
filename="sqlite-amalgamation-${amalgamation_id}"
official_url="https://www.sqlite.org/${sqlite_year}/${filename}.zip"
download_dir="${runner_temp}/sqlite-versions/downloads"
source_root="${runner_temp}/sqlite-versions/source"
source_dir="${source_root}/${filename}"
build_dir="${runner_temp}/sqlite-versions/build/${sqlite_version}"
archive_path="${download_dir}/${filename}.zip"
mkdir -p "$download_dir" "$source_root" "$build_dir"
download_archive() {
local url
local candidate_path="${archive_path}.tmp"
local urls=("$official_url")
for url in $builtin_fallback_urls $extra_fallback_urls; do
urls+=("$url")
done
rm -f "$candidate_path"
for url in "${urls[@]}"; do
echo "Downloading SQLite ${sqlite_version} amalgamation from ${url}"
if curl \
--fail \
--location \
--show-error \
--retry 5 \
--retry-delay 2 \
--retry-max-time 180 \
--retry-all-errors \
--connect-timeout 20 \
--max-time 240 \
--output "$candidate_path" \
"$url"; then
mv "$candidate_path" "$archive_path"
return 0
fi
echo "::warning::Download failed from ${url}"
rm -f "$candidate_path"
done
echo "::error::Could not download SQLite ${sqlite_version} amalgamation"
return 1
}
if [ ! -f "${source_dir}/sqlite3.c" ]; then
if [ ! -f "$archive_path" ]; then
download_archive
fi
rm -rf "$source_dir"
unzip -q "$archive_path" -d "$source_root"
fi
if [ ! -f "${source_dir}/sqlite3.c" ]; then
echo "::error::Expected ${source_dir}/sqlite3.c after extracting ${archive_path}"
exit 1
fi
read -r -a cflag_args <<< "$cflags"
echo "Compiling SQLite ${sqlite_version} to ${build_dir}/${library_name}"
gcc \
-fPIC \
-shared \
"${cflag_args[@]}" \
"${source_dir}/sqlite3.c" \
"-I${source_dir}" \
-o "${build_dir}/${library_name}"
if [ "$library_name" = "libsqlite3.so.0" ]; then
ln -sf "$library_name" "${build_dir}/libsqlite3.so"
fi
if [ -n "${GITHUB_OUTPUT:-}" ]; then
echo "sqlite-location=${build_dir}" >> "$GITHUB_OUTPUT"
else
echo "sqlite-location=${build_dir}"
fi
case "$(printf '%s' "$skip_activate" | tr '[:upper:]' '[:lower:]')" in
true | 1 | yes)
echo "Skipping ${library_path_var} activation"
;;
*)
existing_value="${!library_path_var:-}"
if [ -n "${GITHUB_ENV:-}" ]; then
if [ -n "$existing_value" ]; then
echo "${library_path_var}=${build_dir}:${existing_value}" >> "$GITHUB_ENV"
else
echo "${library_path_var}=${build_dir}" >> "$GITHUB_ENV"
fi
fi
echo "Added ${build_dir} to ${library_path_var}"
;;
esac

View file

@ -14,46 +14,24 @@ jobs:
deploy: deploy:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Check deployment prerequisites
id: deployment-prerequisites
env:
GCP_SA_KEY: ${{ secrets.GCP_SA_KEY }}
LATEST_DATASETTE_SECRET: ${{ secrets.LATEST_DATASETTE_SECRET }}
run: |
missing=()
for variable in GCP_SA_KEY LATEST_DATASETTE_SECRET; do
if [[ -z "${!variable:-}" ]]; then
missing+=("$variable")
fi
done
if (( ${#missing[@]} )); then
echo "::notice::Skipping deployment because required environment variables are missing: ${missing[*]}"
echo "available=false" >> "$GITHUB_OUTPUT"
else
echo "available=true" >> "$GITHUB_OUTPUT"
fi
- name: Check out datasette - name: Check out datasette
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }} uses: actions/checkout@v6
uses: actions/checkout@v7
- name: Set up Python - name: Set up Python
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
python-version: "3.13" python-version: "3.13"
cache: pip cache: pip
- name: Install Python dependencies - name: Install Python dependencies
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
run: | run: |
python -m pip install --upgrade pip python -m pip install --upgrade pip
python -m pip install . --group dev python -m pip install . --group dev
python -m pip install sphinx-to-sqlite==0.1a1 "s3-credentials>=0.17" python -m pip install sphinx-to-sqlite==0.1a1
- name: Run tests - name: Run tests
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' && github.ref == 'refs/heads/main' }} if: ${{ github.ref == 'refs/heads/main' }}
run: | run: |
pytest -n auto -m "not serial" pytest -n auto -m "not serial"
pytest -m "serial" pytest -m "serial"
- name: Build fixtures.db and other files needed to deploy the demo - name: Build fixtures.db and other files needed to deploy the demo
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
run: |- run: |-
python tests/fixtures.py \ python tests/fixtures.py \
fixtures.db \ fixtures.db \
@ -61,18 +39,14 @@ jobs:
fixtures-metadata.json \ fixtures-metadata.json \
plugins \ plugins \
--extra-db-filename extra_database.db --extra-db-filename extra_database.db
# Package the config with the plugins, excluding test-only plugin secrets
# that reference temporary files outside the deployed container.
jq 'del(.plugins)' fixtures-config.json > plugins/fixtures-config.json
- name: Build docs.db - name: Build docs.db
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' && github.ref == 'refs/heads/main' }} if: ${{ github.ref == 'refs/heads/main' }}
run: |- run: |-
cd docs cd docs
DISABLE_SPHINX_INLINE_TABS=1 sphinx-build -b xml . _build DISABLE_SPHINX_INLINE_TABS=1 sphinx-build -b xml . _build
sphinx-to-sqlite ../docs.db _build sphinx-to-sqlite ../docs.db _build
cd .. cd ..
- name: Set up the alternate-route demo - name: Set up the alternate-route demo
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
run: | run: |
echo ' echo '
from datasette import hookimpl from datasette import hookimpl
@ -84,7 +58,6 @@ jobs:
' > plugins/alternative_route.py ' > plugins/alternative_route.py
cp fixtures.db fixtures2.db cp fixtures.db fixtures2.db
- name: And the counters writable stored query demo - name: And the counters writable stored query demo
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
run: | run: |
cat > plugins/counters.py <<EOF cat > plugins/counters.py <<EOF
from datasette import hookimpl from datasette import hookimpl
@ -124,15 +97,12 @@ jobs:
# cat metadata.json # cat metadata.json
- id: auth - id: auth
name: Authenticate to Google Cloud name: Authenticate to Google Cloud
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
uses: google-github-actions/auth@v3 uses: google-github-actions/auth@v3
with: with:
credentials_json: ${{ secrets.GCP_SA_KEY }} credentials_json: ${{ secrets.GCP_SA_KEY }}
- name: Set up Cloud SDK - name: Set up Cloud SDK
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
uses: google-github-actions/setup-gcloud@v3 uses: google-github-actions/setup-gcloud@v3
- name: Deploy to Cloud Run - name: Deploy to Cloud Run
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' }}
env: env:
LATEST_DATASETTE_SECRET: ${{ secrets.LATEST_DATASETTE_SECRET }} LATEST_DATASETTE_SECRET: ${{ secrets.LATEST_DATASETTE_SECRET }}
run: |- run: |-
@ -147,16 +117,16 @@ jobs:
--plugins-dir=plugins \ --plugins-dir=plugins \
--branch=$GITHUB_SHA \ --branch=$GITHUB_SHA \
--version-note=$GITHUB_SHA \ --version-note=$GITHUB_SHA \
--extra-options="--config plugins/fixtures-config.json --setting template_debug 1 --setting trace_debug 1 --crossdb --root" \ --extra-options="--setting template_debug 1 --setting trace_debug 1 --crossdb --root" \
--install 'datasette-ephemeral-tables>=0.2.2' \ --install 'datasette-ephemeral-tables>=0.2.2' \
--service "datasette-latest$SUFFIX" \ --service "datasette-latest$SUFFIX" \
--secret $LATEST_DATASETTE_SECRET --secret $LATEST_DATASETTE_SECRET
- name: Upload latest documentation database to S3 (only for main) - name: Deploy to docs as well (only for main)
if: ${{ steps.deployment-prerequisites.outputs.available == 'true' && github.ref == 'refs/heads/main' }} if: ${{ github.ref == 'refs/heads/main' }}
env:
AWS_ACCESS_KEY_ID: ${{ secrets.S3_DATASETTE_DOCS_ACCESS_KEY }}
AWS_SECRET_ACCESS_KEY: ${{ secrets.S3_DATASETTE_DOCS_SECRET_KEY }}
run: |- run: |-
# Keep development documentation separate from the stable release database. # Deploy docs.db to a different service
s3-credentials put-object datasette-docs latest/docs.db docs.db \ datasette publish cloudrun docs.db \
--content-type application/octet-stream --branch=$GITHUB_SHA \
--version-note=$GITHUB_SHA \
--extra-options="--setting template_debug 1" \
--service=datasette-docs-latest

View file

@ -0,0 +1,16 @@
name: Read the Docs Pull Request Preview
on:
pull_request:
types:
- opened
permissions:
pull-requests: write
jobs:
documentation-links:
runs-on: ubuntu-latest
steps:
- uses: readthedocs/actions/preview@v1
with:
project-slug: "datasette"

View file

@ -2,15 +2,9 @@ name: Playwright
on: on:
push: push:
branches:
- main
pull_request: pull_request:
workflow_dispatch: workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -22,7 +16,7 @@ jobs:
matrix: matrix:
browser: [chromium, firefox, webkit] browser: [chromium, firefox, webkit]
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python 3.14 - name: Set up Python 3.14
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -31,14 +25,14 @@ jobs:
cache: pip cache: pip
cache-dependency-path: pyproject.toml cache-dependency-path: pyproject.toml
- name: Cache uv - name: Cache uv
uses: actions/cache@v6 uses: actions/cache@v5
with: with:
path: ~/.cache/uv path: ~/.cache/uv
key: ${{ runner.os }}-py3.14-uv-${{ hashFiles('pyproject.toml') }} key: ${{ runner.os }}-py3.14-uv-${{ hashFiles('pyproject.toml') }}
restore-keys: | restore-keys: |
${{ runner.os }}-py3.14-uv- ${{ runner.os }}-py3.14-uv-
- name: Cache Playwright browsers - name: Cache Playwright browsers
uses: actions/cache@v6 uses: actions/cache@v5
with: with:
path: ~/.cache/ms-playwright/ path: ~/.cache/ms-playwright/
key: ${{ runner.os }}-playwright-${{ matrix.browser }}-${{ hashFiles('pyproject.toml') }} key: ${{ runner.os }}-playwright-${{ matrix.browser }}-${{ hashFiles('pyproject.toml') }}

View file

@ -1,15 +1,6 @@
name: Check JavaScript for conformance with Prettier name: Check JavaScript for conformance with Prettier
on: on: [push]
push:
branches:
- main
pull_request:
workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -19,8 +10,8 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Check out repo - name: Check out repo
uses: actions/checkout@v7 uses: actions/checkout@v6
- uses: actions/cache@v6 - uses: actions/cache@v5
name: Configure npm caching name: Configure npm caching
with: with:
path: ~/.npm path: ~/.npm

View file

@ -2,7 +2,7 @@ name: Publish Python Package
on: on:
release: release:
types: [published] types: [created]
permissions: permissions:
contents: read contents: read
@ -14,7 +14,7 @@ jobs:
matrix: matrix:
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"] python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python ${{ matrix.python-version }} - name: Set up Python ${{ matrix.python-version }}
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -35,7 +35,7 @@ jobs:
permissions: permissions:
id-token: write id-token: write
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python - name: Set up Python
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -51,14 +51,12 @@ jobs:
- name: Publish - name: Publish
uses: pypa/gh-action-pypi-publish@release/v1 uses: pypa/gh-action-pypi-publish@release/v1
# After the first non-prerelease 1.0 release, disable this job on 0.65.x,
# even for later 0.65 releases, so they cannot overwrite the 1.0 stable docs.
deploy_static_docs: deploy_static_docs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
needs: [deploy] needs: [deploy]
if: "!github.event.release.prerelease" if: "!github.event.release.prerelease"
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python - name: Set up Python
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -68,27 +66,33 @@ jobs:
- name: Install dependencies - name: Install dependencies
run: | run: |
python -m pip install . --group dev python -m pip install . --group dev
python -m pip install sphinx-to-sqlite==0.1a1 "s3-credentials>=0.17" python -m pip install sphinx-to-sqlite==0.1a1
- name: Build docs.db - name: Build docs.db
run: |- run: |-
cd docs cd docs
DISABLE_SPHINX_INLINE_TABS=1 sphinx-build -b xml . _build DISABLE_SPHINX_INLINE_TABS=1 sphinx-build -b xml . _build
sphinx-to-sqlite ../docs.db _build sphinx-to-sqlite ../docs.db _build
cd .. cd ..
- name: Upload stable documentation database to S3 - id: auth
env: name: Authenticate to Google Cloud
AWS_ACCESS_KEY_ID: ${{ secrets.S3_DATASETTE_DOCS_ACCESS_KEY }} uses: google-github-actions/auth@v2
AWS_SECRET_ACCESS_KEY: ${{ secrets.S3_DATASETTE_DOCS_SECRET_KEY }} with:
credentials_json: ${{ secrets.GCP_SA_KEY }}
- name: Set up Cloud SDK
uses: google-github-actions/setup-gcloud@v3
- name: Deploy stable-docs.datasette.io to Cloud Run
run: |- run: |-
s3-credentials put-object datasette-docs docs.db docs.db \ gcloud config set run/region us-central1
--content-type application/octet-stream gcloud config set project datasette-222320
datasette publish cloudrun docs.db \
--service=datasette-docs-stable
deploy_docker: deploy_docker:
runs-on: ubuntu-latest runs-on: ubuntu-latest
needs: [deploy] needs: [deploy]
if: "!github.event.release.prerelease" if: "!github.event.release.prerelease"
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Build and push to Docker Hub - name: Build and push to Docker Hub
env: env:
DOCKER_USER: ${{ secrets.DOCKER_USER }} DOCKER_USER: ${{ secrets.DOCKER_USER }}

View file

@ -13,7 +13,7 @@ jobs:
deploy_docker: deploy_docker:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Build and push to Docker Hub - name: Build and push to Docker Hub
env: env:
DOCKER_USER: ${{ secrets.DOCKER_USER }} DOCKER_USER: ${{ secrets.DOCKER_USER }}

View file

@ -1,15 +1,6 @@
name: Check spelling in documentation name: Check spelling in documentation
on: on: [push, pull_request]
push:
branches:
- main
pull_request:
workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -18,7 +9,7 @@ jobs:
spellcheck: spellcheck:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python - name: Set up Python
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:

View file

@ -15,7 +15,7 @@ jobs:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- name: Checkout repository - name: Checkout repository
uses: actions/checkout@v7 uses: actions/checkout@v6
with: with:
fetch-depth: 0 # We need all commits to find docs/ changes fetch-depth: 0 # We need all commits to find docs/ changes
- name: Set up Git user - name: Set up Git user

40
.github/workflows/test-coverage.yml vendored Normal file
View file

@ -0,0 +1,40 @@
name: Calculate test coverage
on:
push:
branches:
- main
pull_request:
branches:
- main
permissions:
contents: read
jobs:
test:
runs-on: ubuntu-latest
steps:
- name: Check out datasette
uses: actions/checkout@v6
- name: Set up Python
uses: actions/setup-python@v6
with:
python-version: '3.12'
cache: 'pip'
cache-dependency-path: '**/pyproject.toml'
- name: Install Python dependencies
run: |
python -m pip install --upgrade pip
python -m pip install . --group dev
python -m pip install pytest-cov
- name: Run tests
run: |-
ls -lah
cat .coveragerc
pytest -m "not serial" --cov=datasette --cov-config=.coveragerc --cov-report xml:coverage.xml --cov-report term -x
ls -lah
- name: Upload coverage report
uses: codecov/codecov-action@v1
with:
token: ${{ secrets.CODECOV_TOKEN }}
file: coverage.xml

View file

@ -2,15 +2,9 @@ name: Test in Pyodide with shot-scraper
on: on:
push: push:
branches:
- main
pull_request: pull_request:
workflow_dispatch: workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -18,7 +12,7 @@ jobs:
test: test:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python 3.10 - name: Set up Python 3.10
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -26,7 +20,7 @@ jobs:
cache: 'pip' cache: 'pip'
cache-dependency-path: '**/pyproject.toml' cache-dependency-path: '**/pyproject.toml'
- name: Cache Playwright browsers - name: Cache Playwright browsers
uses: actions/cache@v6 uses: actions/cache@v5
with: with:
path: ~/.cache/ms-playwright/ path: ~/.cache/ms-playwright/
key: ${{ runner.os }}-browsers key: ${{ runner.os }}-browsers

View file

@ -1,15 +1,6 @@
name: Test SQLite versions name: Test SQLite versions
on: on: [push, pull_request]
push:
branches:
- main
pull_request:
workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -21,10 +12,10 @@ jobs:
strategy: strategy:
matrix: matrix:
platform: [ubuntu-latest] platform: [ubuntu-latest]
python-version: ["3.13"] python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
sqlite-version: [ sqlite-version: [
#"3", # latest version #"3", # latest version
#"3.46", "3.46",
#"3.45", #"3.45",
#"3.27", #"3.27",
#"3.26", #"3.26",
@ -34,7 +25,7 @@ jobs:
#"3.23.1" # 2018-04-10, before UPSERT #"3.23.1" # 2018-04-10, before UPSERT
] ]
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python ${{ matrix.python-version }} - name: Set up Python ${{ matrix.python-version }}
uses: actions/setup-python@v6 uses: actions/setup-python@v6
with: with:
@ -43,7 +34,7 @@ jobs:
cache: pip cache: pip
cache-dependency-path: pyproject.toml cache-dependency-path: pyproject.toml
- name: Set up SQLite ${{ matrix.sqlite-version }} - name: Set up SQLite ${{ matrix.sqlite-version }}
uses: ./.github/actions/setup-sqlite-version uses: asg017/sqlite-versions@71ea0de37ae739c33e447af91ba71dda8fcf22e6
with: with:
version: ${{ matrix.sqlite-version }} version: ${{ matrix.sqlite-version }}
cflags: "-DSQLITE_ENABLE_DESERIALIZE -DSQLITE_ENABLE_FTS5 -DSQLITE_ENABLE_FTS4 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_RTREE -DSQLITE_ENABLE_JSON1" cflags: "-DSQLITE_ENABLE_DESERIALIZE -DSQLITE_ENABLE_FTS5 -DSQLITE_ENABLE_FTS4 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_RTREE -DSQLITE_ENABLE_JSON1"

View file

@ -1,15 +1,6 @@
name: Test name: Test
on: on: [push, pull_request]
push:
branches:
- main
pull_request:
workflow_dispatch:
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions: permissions:
contents: read contents: read
@ -18,22 +9,17 @@ jobs:
test: test:
runs-on: ubuntu-latest runs-on: ubuntu-latest
strategy: strategy:
fail-fast: false
matrix: matrix:
python-version: ["3.10", "3.11", "3.12", "3.13", "3.14", "3.15"] python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
include:
- python-version: "3.14"
coverage: true
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Set up Python ${{ matrix.python-version }} - name: Set up Python ${{ matrix.python-version }}
uses: actions/setup-python@v7 uses: actions/setup-python@v6
with: with:
python-version: ${{ matrix.python-version }} python-version: ${{ matrix.python-version }}
allow-prereleases: true allow-prereleases: true
cache: pip cache: pip
cache-dependency-path: pyproject.toml cache-dependency-path: pyproject.toml
check-latest: true
- name: Build extension for --load-extension test - name: Build extension for --load-extension test
run: |- run: |-
(cd tests && gcc ext.c -fPIC -shared -o ext.so) (cd tests && gcc ext.c -fPIC -shared -o ext.so)
@ -41,27 +27,12 @@ jobs:
run: | run: |
pip install . --group dev pip install . --group dev
pip freeze pip freeze
- name: Install pytest-cov
if: ${{ matrix.coverage }}
run: pip install pytest-cov
- name: Run tests - name: Run tests
run: | run: |
if [ "${{ matrix.coverage }}" = "true" ]; then
COV="--cov=datasette --cov-config=.coveragerc"
pytest -n auto -m "not serial" $COV --cov-report=
pytest -m "serial" $COV --cov-append --cov-report xml:coverage.xml --cov-report term
else
pytest -n auto -m "not serial" pytest -n auto -m "not serial"
pytest -m "serial" pytest -m "serial"
fi
# And the test that exceeds a localhost HTTPS server # And the test that exceeds a localhost HTTPS server
tests/test_datasette_https_server.sh tests/test_datasette_https_server.sh
- name: Upload coverage report
if: ${{ matrix.coverage }}
uses: codecov/codecov-action@v5
with:
token: ${{ secrets.CODECOV_TOKEN }}
files: coverage.xml
- name: Black - name: Black
run: | run: |
black --version black --version

View file

@ -10,6 +10,6 @@ jobs:
build: build:
runs-on: macos-latest runs-on: macos-latest
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Setup tmate session - name: Setup tmate session
uses: mxschmitt/action-tmate@v3 uses: mxschmitt/action-tmate@v3

View file

@ -11,7 +11,7 @@ jobs:
build: build:
runs-on: ubuntu-latest runs-on: ubuntu-latest
steps: steps:
- uses: actions/checkout@v7 - uses: actions/checkout@v6
- name: Setup tmate session - name: Setup tmate session
uses: mxschmitt/action-tmate@v3 uses: mxschmitt/action-tmate@v3
env: env:

2
.gitignore vendored
View file

@ -5,8 +5,6 @@ datasets.json
scratchpad scratchpad
ignored/
.vscode .vscode
uv.lock uv.lock

View file

@ -1,4 +1,4 @@
FROM python:3.11-slim-bookworm AS build FROM python:3.11.0-slim-bullseye as build
# Version of Datasette to install, e.g. 0.55 # Version of Datasette to install, e.g. 0.55
# docker build . -t datasette --build-arg VERSION=0.55 # docker build . -t datasette --build-arg VERSION=0.55

View file

@ -33,11 +33,10 @@ export DATASETTE_SECRET := "not_a_secret"
uv run codespell datasette -S datasette/static --ignore-words docs/codespell-ignore-words.txt uv run codespell datasette -S datasette/static --ignore-words docs/codespell-ignore-words.txt
uv run codespell tests --ignore-words docs/codespell-ignore-words.txt uv run codespell tests --ignore-words docs/codespell-ignore-words.txt
# Run linters: black, ruff, prettier, cog # Run linters: black, ruff, cog
@lint: codespell @lint: codespell
uv run black datasette tests --check uv run black datasette tests --check
uv run ruff check datasette tests uv run ruff check datasette tests
npm run prettier -- --check
uv run cog --check README.md docs/*.rst uv run cog --check README.md docs/*.rst
# Apply ruff fixes # Apply ruff fixes

View file

@ -36,7 +36,7 @@ You can also install it using `pip` or `pipx`:
pip install datasette pip install datasette
Datasette requires Python 3.10 or higher. We also have [detailed installation instructions](https://docs.datasette.io/en/stable/installation.html) covering other options such as Docker. Datasette requires Python 3.8 or higher. We also have [detailed installation instructions](https://docs.datasette.io/en/stable/installation.html) covering other options such as Docker.
## Basic usage ## Basic usage

View file

@ -1,15 +1,8 @@
from datasette.permissions import Permission # noqa from datasette.permissions import Permission # noqa
from datasette.version import __version_info__, __version__ # noqa from datasette.version import __version_info__, __version__ # noqa
from datasette.events import Event # noqa from datasette.events import Event # noqa
from datasette.background_tasks import BackgroundTask, BackgroundTaskSupervisor # noqa from datasette.tokens import TokenHandler, TokenRestrictions # noqa
from datasette.tokens import TokenHandler, TokenInvalid, TokenRestrictions # noqa from datasette.utils.asgi import Forbidden, NotFound, Request, Response # noqa
from datasette.utils.asgi import ( # noqa
Forbidden,
NotFound,
PayloadTooLarge,
Request,
Response,
)
from datasette.utils import actor_matches_allow # noqa from datasette.utils import actor_matches_allow # noqa
from datasette.views import Context # noqa from datasette.views import Context # noqa
from .hookspecs import hookimpl # noqa from .hookspecs import hookimpl # noqa

View file

@ -89,8 +89,7 @@ def pytest_runtest_protocol(item, nextitem):
continue continue
try: try:
ds.close() ds.close()
except Exception as e: # noqa: BLE001 except Exception as e:
# Surfaced as a pytest warning; teardown must not fail the run
item.warn( item.warn(
pytest.PytestUnraisableExceptionWarning( pytest.PytestUnraisableExceptionWarning(
f"Error closing Datasette instance: {e!r}" f"Error closing Datasette instance: {e!r}"

View file

@ -1,9 +1,7 @@
import time
from itsdangerous import BadSignature
from datasette import hookimpl from datasette import hookimpl
from itsdangerous import BadSignature
from datasette.utils import baseconv from datasette.utils import baseconv
import time
@hookimpl @hookimpl

File diff suppressed because it is too large Load diff

View file

@ -1,227 +0,0 @@
"""
Supervised background-task registration for Datasette core.
Plugins that need long-lived background work (a polling loop, a queue
consumer, a scheduled job runner) register it with
``datasette.add_background_task(func, name=None)`` - typically from a
``startup`` plugin hook - instead of fire-and-forgetting their own
``asyncio.create_task()``. Core owns:
- **references**: every launched ``asyncio.Task`` is kept alive on a
:class:`BackgroundTaskSupervisor`, so it can never be silently garbage
collected the way an unreferenced ``create_task()`` call can be;
- **launch timing**: registered work is buffered until
:meth:`BackgroundTaskSupervisor.launch_all` runs, which core arranges to
happen only after *every* plugin's ``startup`` hook has finished - so
a task that depends on another plugin having registered something first
doesn't need ``tryfirst=True`` ordering tricks;
- **crash surfacing**: an unhandled exception in a background task is
logged with its full traceback to the ``datasette.background_tasks``
logger and recorded on the handle, instead of becoming an "Task
exception was never retrieved" warning nobody sees;
- **cancellation**: :meth:`BackgroundTaskSupervisor.cancel_all` cancels
every task still running and waits (with a grace period) for them to
actually stop.
"""
from __future__ import annotations
import asyncio
import datetime
import functools
import logging
from collections.abc import Awaitable, Callable
logger = logging.getLogger("datasette.background_tasks")
def _utcnow_iso() -> str:
return datetime.datetime.now(datetime.timezone.utc).isoformat()
def _function_path(func: Callable) -> str:
"""Describe the callable without guessing which plugin registered it."""
while isinstance(func, functools.partial):
func = func.func
if not hasattr(func, "__qualname__"):
func = type(func).__call__
return f"{func.__module__}.{func.__qualname__}"
class BackgroundTask:
"""A handle to a single piece of supervised background work.
States: ``registered`` (added but not yet launched) -> ``running`` ->
one of ``completed`` (returned cleanly), ``crashed`` (raised an
exception other than ``CancelledError`` - see ``.exception``), or
``cancelled`` (``.cancel()`` was called, or it was still running at
shutdown).
"""
def __init__(
self,
name: str,
func: Callable[[object], Awaitable[None]],
):
self.name = name
self.state = "registered"
self.task: asyncio.Task | None = None
self.exception: BaseException | None = None
self.started_at: str | None = None
self.function = _function_path(func)
self._func = func
self._supervisor: BackgroundTaskSupervisor | None = None
def cancel(self) -> None:
"""Cancel this task.
If it has already been launched, cancels the underlying
``asyncio.Task`` - its state becomes ``cancelled`` once the
cancellation is observed (asynchronously, via the task's done
callback). If it has not been launched yet, this is a no-op as
far as asyncio is concerned (there's no task to cancel) but it
deregisters the handle from its supervisor so it never runs.
"""
if self.task is not None:
self.task.cancel()
elif self._supervisor is not None:
self._supervisor._deregister(self)
def __repr__(self) -> str:
return f"<BackgroundTask name={self.name!r} state={self.state!r}>"
class BackgroundTaskSupervisor:
"""Owns registration and launch of every :class:`BackgroundTask` for a
single ``Datasette`` instance.
Registration (:meth:`add`) is separate from launch
(:meth:`launch_all`): plugins register work whenever convenient
(typically from a ``startup`` hook, but request handlers can register
dynamic per-job work too), and it either sits buffered until
:meth:`launch_all` runs, or - if :meth:`launch_all` has already run -
starts immediately.
Strong references to every :class:`BackgroundTask` (and its
``asyncio.Task``) are kept for the life of the instance, by design -
that's what makes the enrichments-style "fire-and-forget task gets
garbage collected mid-flight" bug impossible here. There is currently
no pruning of completed/crashed/cancelled tasks, so a plugin that
dynamically registers many short-lived tasks over a long process
lifetime (a per-job registration pattern, e.g. one task per queued
job) will grow this list without bound. That's an accepted v1
trade-off in favour of full introspection (``/-/tasks``); revisit
with a pruning or capping policy if unbounded growth is reported in
practice.
"""
def __init__(self, datasette):
self._datasette = datasette
self._tasks: list[BackgroundTask] = []
self._names = set()
self._launched = False
self._lock = asyncio.Lock()
def add(self, func, name=None) -> BackgroundTask:
base_name = name or getattr(func, "__qualname__", None) or repr(func)
actual_name = self._unique_name(base_name)
handle = BackgroundTask(actual_name, func)
handle._supervisor = self
self._tasks.append(handle)
self._names.add(actual_name)
if self._launched:
self._launch_one(handle)
return handle
def _unique_name(self, base_name: str) -> str:
if base_name not in self._names:
return base_name
n = 2
while f"{base_name}-{n}" in self._names:
n += 1
return f"{base_name}-{n}"
def _deregister(self, handle: BackgroundTask) -> None:
try:
self._tasks.remove(handle)
except ValueError:
pass
self._names.discard(handle.name)
def _launch_one(self, handle: BackgroundTask) -> None:
handle.state = "running"
handle.started_at = _utcnow_iso()
handle.task = asyncio.create_task(
handle._func(self._datasette), name=handle.name
)
handle.task.add_done_callback(functools.partial(_on_task_done, handle))
async def launch_all(self) -> None:
"""Launch every currently-registered task that hasn't launched
yet. Idempotent and safe to call concurrently: subsequent (or
racing) calls are no-ops once the first has set ``self._launched``.
"""
if self._launched:
return
async with self._lock:
if self._launched:
return
self._launched = True
for handle in list(self._tasks):
if handle.task is None:
self._launch_one(handle)
async def cancel_all(self, grace: float = 5.0) -> None:
"""Cancel every task that isn't already done, then wait up to
``grace`` seconds for them to actually finish. Stragglers still
running after that are logged by name (but left to finish or not
on their own - this does not forcibly kill them, asyncio has no
mechanism for that).
"""
handles_by_task = {
handle.task: handle for handle in self._tasks if handle.task is not None
}
pending = [task for task in handles_by_task if not task.done()]
for task in pending:
task.cancel()
if not pending:
return
_done, not_done = await asyncio.wait(pending, timeout=grace)
if not_done:
names = sorted(handles_by_task[task].name for task in not_done)
logger.warning(
"%d background task(s) did not finish within the %.1fs grace "
"period after cancellation: %s",
len(names),
grace,
", ".join(names),
)
def tasks(self) -> list[BackgroundTask]:
"""Return every registered :class:`BackgroundTask`, launched or
not, in registration order. Used by the ``/-/tasks`` debug
endpoint.
"""
return list(self._tasks)
@property
def launched(self) -> bool:
"""Whether :meth:`launch_all` has run yet - lets ``/-/tasks``
distinguish "no tasks registered" from "tasks registered but
nothing has armed the launch yet" without reaching for the
private ``_launched`` attribute.
"""
return self._launched
def _on_task_done(handle: BackgroundTask, task: asyncio.Task) -> None:
if task.cancelled():
handle.state = "cancelled"
return
exc = task.exception()
if exc is not None:
handle.state = "crashed"
handle.exception = exc
logger.error("Background task %r crashed", handle.name, exc_info=exc)
return
handle.state = "completed"

View file

@ -1,8 +1,7 @@
import hashlib
from datasette import hookimpl from datasette import hookimpl
from datasette.utils.asgi import Response, BadRequest
from datasette.utils import to_css_class from datasette.utils import to_css_class
from datasette.utils.asgi import BadRequest, Response import hashlib
_BLOB_COLUMN = "_blob_column" _BLOB_COLUMN = "_blob_column"
_BLOB_HASH = "_blob_hash" _BLOB_HASH = "_blob_hash"

View file

@ -1,45 +1,43 @@
import asyncio import asyncio
import uvicorn
import click
from click import formatting
from click.types import CompositeParamType
from click_default_group import DefaultGroup
import functools import functools
import json import json
import os import os
import pathlib import pathlib
from runpy import run_module
import shutil import shutil
from subprocess import call
import sys import sys
import textwrap import textwrap
import webbrowser import webbrowser
from runpy import run_module
from subprocess import call
import click
import uvicorn
from click import formatting
from click.types import CompositeParamType
from click_default_group import DefaultGroup
from .app import ( from .app import (
Datasette,
DEFAULT_SETTINGS, DEFAULT_SETTINGS,
SETTINGS, SETTINGS,
SQLITE_LIMIT_ATTACHED, SQLITE_LIMIT_ATTACHED,
Datasette,
pm, pm,
) )
from .inspect import inspect_tables from .inspect import inspect_tables
from .utils import ( from .utils import (
ConnectionProblem,
LoadExtension, LoadExtension,
SpatialiteConnectionProblem,
SpatialiteNotFound,
StartupError, StartupError,
StaticMount,
ValueAsBooleanError,
check_connection, check_connection,
deep_dict_update, deep_dict_update,
find_spatialite, find_spatialite,
parse_metadata,
ConnectionProblem,
SpatialiteConnectionProblem,
initial_path_for_datasette, initial_path_for_datasette,
pairs_to_nested_config, pairs_to_nested_config,
parse_metadata,
temporary_docker_directory, temporary_docker_directory,
value_as_boolean, value_as_boolean,
SpatialiteNotFound,
StaticMount,
ValueAsBooleanError,
) )
from .utils.sqlite import sqlite3 from .utils.sqlite import sqlite3
from .utils.testing import TestClient from .utils.testing import TestClient
@ -77,7 +75,7 @@ class Setting(CompositeParamType):
# Datasette 1.0, we turn bare setting names into setting.name # Datasette 1.0, we turn bare setting names into setting.name
# Type checking for those older settings # Type checking for those older settings
default = DEFAULT_SETTINGS[name] default = DEFAULT_SETTINGS[name]
name = f"settings.{name}" name = "settings.{}".format(name)
if isinstance(default, bool): if isinstance(default, bool):
try: try:
return name, "true" if value_as_boolean(value) else "false" return name, "true" if value_as_boolean(value) else "false"
@ -173,6 +171,7 @@ async def inspect_(files, sqlite_extensions):
@cli.group() @cli.group()
def publish(): def publish():
"""Publish specified SQLite database files to the internet along with a Datasette-powered interface and API""" """Publish specified SQLite database files to the internet along with a Datasette-powered interface and API"""
pass
# Register publish plugins # Register publish plugins
@ -497,7 +496,6 @@ def uninstall(packages, yes):
"--internal", "--internal",
type=click.Path(), type=click.Path(),
help="Path to a persistent Datasette internal SQLite database", help="Path to a persistent Datasette internal SQLite database",
envvar="DATASETTE_INTERNAL",
) )
def serve( def serve(
files, files,
@ -580,27 +578,27 @@ def serve(
# https://github.com/simonw/datasette/issues/2389 # https://github.com/simonw/datasette/issues/2389
deep_dict_update(config_data, settings_updates) deep_dict_update(config_data, settings_updates)
kwargs = { kwargs = dict(
"immutables": immutable, immutables=immutable,
"cache_headers": not reload, cache_headers=not reload,
"cors": cors, cors=cors,
"inspect_data": inspect_data, inspect_data=inspect_data,
"config": config_data, config=config_data,
"metadata": metadata_data, metadata=metadata_data,
"sqlite_extensions": sqlite_extensions, sqlite_extensions=sqlite_extensions,
"template_dir": template_dir, template_dir=template_dir,
"plugins_dir": plugins_dir, plugins_dir=plugins_dir,
"static_mounts": static, static_mounts=static,
"settings": None, # These are passed in config= now settings=None, # These are passed in config= now
"memory": memory, memory=memory,
"secret": secret, secret=secret,
"version_note": version_note, version_note=version_note,
"pdb": pdb, pdb=pdb,
"crossdb": crossdb, crossdb=crossdb,
"nolock": nolock, nolock=nolock,
"internal": internal, internal=internal,
"default_deny": default_deny, default_deny=default_deny,
} )
# Separate directories from files # Separate directories from files
directories = [f for f in files if os.path.isdir(f)] directories = [f for f in files if os.path.isdir(f)]
@ -623,7 +621,9 @@ def serve(
conn.close() conn.close()
else: else:
raise click.ClickException( raise click.ClickException(
f"Invalid value for '[FILES]...': Path '{file}' does not exist." "Invalid value for '[FILES]...': Path '{}' does not exist.".format(
file
)
) )
# Check for duplicate files by resolving all paths to their absolute forms # Check for duplicate files by resolving all paths to their absolute forms
@ -664,6 +664,16 @@ def serve(
# Private utility mechanism for writing unit tests # Private utility mechanism for writing unit tests
return ds return ds
# Run async soundness checks before startup hooks, since invoke_startup
# now populates internal tables which requires querying each database
run_sync(lambda: check_databases(ds))
# Run the "startup" plugin hooks
try:
run_sync(ds.invoke_startup)
except StartupError as e:
raise click.ClickException(e.args[0])
if headers and not get: if headers and not get:
raise click.ClickException("--headers can only be used with --get") raise click.ClickException("--headers can only be used with --get")
@ -671,23 +681,10 @@ def serve(
raise click.ClickException("--token can only be used with --get") raise click.ClickException("--token can only be used with --get")
if get: if get:
# --get means we don't run Uvicorn at all
run_sync(lambda: check_databases(ds))
try:
run_sync(ds.invoke_startup)
except StartupError as e:
raise click.ClickException(e.args[0])
# --get never launches background tasks: TestClient's request below
# flows through the full ASGI stack, including the
# AsgiRunOnFirstRequest fallback, which would otherwise launch them.
ds._suppress_background_tasks = True
client = TestClient(ds) client = TestClient(ds)
request_headers = {} request_headers = {}
if token: if token:
request_headers["Authorization"] = f"Bearer {token}" request_headers["Authorization"] = "Bearer {}".format(token)
cookies = {} cookies = {}
if actor: if actor:
cookies["ds_actor"] = client.actor_cookie(json.loads(actor)) cookies["ds_actor"] = client.actor_cookie(json.loads(actor))
@ -708,23 +705,6 @@ def serve(
sys.exit(exit_code) sys.exit(exit_code)
return return
# check_databases, invoke_startup() and the uvicorn server all run on a
# single event loop, so that anything a plugin's "startup" hook schedules
# on the loop (asyncio.create_task, Lock/Queue/Event objects, ...) is
# still alive when the server starts handling requests.
async def _serve_async():
# Populate internal catalog tables before invoke_startup
await check_databases(ds)
# Run the full startup sequence (immutable-database table-count
# precompute + the "startup" plugin hooks) via the same entry point
# AsgiLifespan/AsgiRunOnFirstRequest use, so it's not skipped when
# uvicorn's lifespan.startup fires moments later.
try:
await ds._startup_sequence()
except StartupError as e:
raise click.ClickException(e.args[0])
# Start the server # Start the server
url = None url = None
if root: if root:
@ -736,26 +716,19 @@ def serve(
if open_browser: if open_browser:
if url is None: if url is None:
# Figure out most convenient URL - to table, database or homepage # Figure out most convenient URL - to table, database or homepage
path = await initial_path_for_datasette(ds) path = run_sync(lambda: initial_path_for_datasette(ds))
url = f"http://{host}:{port}{path}" url = f"http://{host}:{port}{path}"
webbrowser.open(url) webbrowser.open(url)
uvicorn_kwargs = { uvicorn_kwargs = dict(
"host": host, host=host, port=port, log_level="info", lifespan="on", workers=1
"port": port, )
"log_level": "info",
"lifespan": "on",
"workers": 1,
}
if uds: if uds:
uvicorn_kwargs["uds"] = uds uvicorn_kwargs["uds"] = uds
if ssl_keyfile: if ssl_keyfile:
uvicorn_kwargs["ssl_keyfile"] = ssl_keyfile uvicorn_kwargs["ssl_keyfile"] = ssl_keyfile
if ssl_certfile: if ssl_certfile:
uvicorn_kwargs["ssl_certfile"] = ssl_certfile uvicorn_kwargs["ssl_certfile"] = ssl_certfile
server = uvicorn.Server(uvicorn.Config(ds.app(), **uvicorn_kwargs)) uvicorn.run(ds.app(), **uvicorn_kwargs)
await server.serve()
asyncio.run(_serve_async())
@cli.command() @cli.command()
@ -912,7 +885,7 @@ async def check_databases(ds):
) )
except ConnectionProblem as e: except ConnectionProblem as e:
raise click.UsageError( raise click.UsageError(
f"Connection to {database.path} failed check: {e.args[0]!s}" f"Connection to {database.path} failed check: {str(e.args[0])}"
) )
# If --crossdb and more than SQLITE_LIMIT_ATTACHED show warning # If --crossdb and more than SQLITE_LIMIT_ATTACHED show warning
if ( if (
@ -920,5 +893,9 @@ async def check_databases(ds):
and len([db for db in ds.databases.values() if not db.is_memory]) and len([db for db in ds.databases.values() if not db.is_memory])
> SQLITE_LIMIT_ATTACHED > SQLITE_LIMIT_ATTACHED
): ):
msg = f"Warning: --crossdb only works with the first {SQLITE_LIMIT_ATTACHED} attached databases" msg = (
"Warning: --crossdb only works with the first {} attached databases".format(
SQLITE_LIMIT_ATTACHED
)
)
click.echo(click.style(msg, bold=True, fg="yellow"), err=True) click.echo(click.style(msg, bold=True, fg="yellow"), err=True)

View file

@ -64,14 +64,14 @@ class ColumnType:
Return an HTML string to render this cell value, or None to Return an HTML string to render this cell value, or None to
fall through to the default render_cell plugin hook chain. fall through to the default render_cell plugin hook chain.
""" """
return return None
async def validate(self, value, datasette): async def validate(self, value, datasette):
""" """
Validate a value before it is written. Return None if valid, Validate a value before it is written. Return None if valid,
or a string error message if invalid. or a string error message if invalid.
""" """
return return None
async def transform_value(self, value, datasette): async def transform_value(self, value, datasette):
""" """

View file

@ -40,12 +40,12 @@ def _origin_tuple(value):
scheme = (parsed.scheme or "").lower() scheme = (parsed.scheme or "").lower()
host = (parsed.hostname or "").lower() host = (parsed.hostname or "").lower()
if not scheme or not host: if not scheme or not host:
raise ValueError(f"missing scheme or host in {value!r}") raise ValueError("missing scheme or host in {!r}".format(value))
port = parsed.port # may raise ValueError on bad ports port = parsed.port # may raise ValueError on bad ports
if port is None: if port is None:
port = DEFAULT_PORTS.get(scheme) port = DEFAULT_PORTS.get(scheme)
if port is None: if port is None:
raise ValueError(f"unknown default port for scheme {scheme!r}") raise ValueError("unknown default port for scheme {!r}".format(scheme))
return scheme, host, port return scheme, host, port
@ -125,7 +125,9 @@ class CrossOriginProtectionMiddleware:
return return
await self._forbid( await self._forbid(
send, send,
f"Sec-Fetch-Site was {sec_fetch_site!r}, expected 'same-origin' or 'none'", "Sec-Fetch-Site was {!r}, expected 'same-origin' or 'none'".format(
sec_fetch_site
),
) )
return return
@ -139,11 +141,11 @@ class CrossOriginProtectionMiddleware:
request_scheme = self._request_scheme(scope) request_scheme = self._request_scheme(scope)
try: try:
origin_tuple = _origin_tuple(origin) origin_tuple = _origin_tuple(origin)
expected_tuple = _origin_tuple(f"{request_scheme}://{host}") expected_tuple = _origin_tuple("{}://{}".format(request_scheme, host))
except ValueError: except ValueError:
await self._forbid( await self._forbid(
send, send,
f"Malformed Origin {origin!r} or Host {host!r}", "Malformed Origin {!r} or Host {!r}".format(origin, host),
) )
return return
@ -153,7 +155,7 @@ class CrossOriginProtectionMiddleware:
await self._forbid( await self._forbid(
send, send,
f"Origin {origin!r} does not match Host {host!r}", "Origin {!r} does not match Host {!r}".format(origin, host),
) )
def _request_scheme(self, scope): def _request_scheme(self, scope):
@ -161,8 +163,7 @@ class CrossOriginProtectionMiddleware:
try: try:
if self.datasette.setting("force_https_urls"): if self.datasette.setting("force_https_urls"):
return "https" return "https"
except Exception: # noqa: BLE001, S110 except Exception:
# Settings may not be readable this early; fall back to the ASGI scheme
pass pass
return scope.get("scheme") or "http" return scope.get("scheme") or "http"

View file

@ -1,35 +1,33 @@
import asyncio import asyncio
import atexit import atexit
from collections import namedtuple
import inspect import inspect
import os import os
from pathlib import Path
import queue import queue
import sqlite_utils
import sys import sys
import tempfile import tempfile
import threading import threading
import uuid import uuid
from collections import namedtuple
from pathlib import Path
import sqlite_utils
from .inspect import inspect_hash
from .tracer import trace from .tracer import trace
from .utils import ( from .utils import (
call_with_supported_arguments, call_with_supported_arguments,
detect_fts, detect_fts,
detect_primary_keys, detect_primary_keys,
detect_spatialite, detect_spatialite,
escape_sqlite,
get_all_foreign_keys, get_all_foreign_keys,
get_outbound_foreign_keys, get_outbound_foreign_keys,
md5_not_usedforsecurity, md5_not_usedforsecurity,
sqlite3,
sqlite_timelimit, sqlite_timelimit,
table_column_details, sqlite3,
table_columns, table_columns,
table_column_details,
) )
from .utils.sql_analysis import SQLAnalysis, analyze_sql_tables from .utils.sql_analysis import SQLAnalysis, analyze_sql_tables
from .utils.sqlite import sqlite_derived_table_dependencies, sqlite_hidden_table_names from .utils.sqlite import sqlite_hidden_table_names
from .inspect import inspect_hash
connections = threading.local() connections = threading.local()
@ -85,7 +83,6 @@ class Database:
self.cached_hash = None self.cached_hash = None
self.cached_size = None self.cached_size = None
self._cached_table_counts = None self._cached_table_counts = None
self._cached_derived_table_dependencies = None
self._write_thread = None self._write_thread = None
self._write_queue = None self._write_queue = None
self._closed = False self._closed = False
@ -94,15 +91,16 @@ class Database:
# These are used when in non-threaded mode: # These are used when in non-threaded mode:
self._read_connection = None self._read_connection = None
self._write_connection = None self._write_connection = None
# Track file and memory connections, including reads on worker threads, # This is used to track all file connections so they can be closed
# so close() can release all of them from the calling thread. self._all_file_connections = []
self._all_connections = []
if not is_temp_disk: if not is_temp_disk:
self.mode = mode self.mode = mode
def _check_not_closed(self): def _check_not_closed(self):
if self._closed: if self._closed:
raise DatasetteClosedError(f"Database {self.name!r} has been closed") raise DatasetteClosedError(
"Database {!r} has been closed".format(self.name)
)
def _remove_pending_execute_future(self, future): def _remove_pending_execute_future(self, future):
with self._pending_execute_futures_lock: with self._pending_execute_futures_lock:
@ -141,18 +139,15 @@ class Database:
if write: if write:
extra_kwargs["isolation_level"] = "IMMEDIATE" extra_kwargs["isolation_level"] = "IMMEDIATE"
if self.memory_name: if self.memory_name:
uri = f"file:{self.memory_name}?mode=memory&cache=shared" uri = "file:{}?mode=memory&cache=shared".format(self.memory_name)
conn = sqlite3.connect( conn = sqlite3.connect(
uri, uri=True, check_same_thread=False, **extra_kwargs uri, uri=True, check_same_thread=False, **extra_kwargs
) )
if not write: if not write:
conn.execute("PRAGMA query_only=1") conn.execute("PRAGMA query_only=1")
self._all_connections.append(conn)
return conn return conn
if self.is_memory: if self.is_memory:
conn = sqlite3.connect(":memory:", uri=True, check_same_thread=False) return sqlite3.connect(":memory:", uri=True)
self._all_connections.append(conn)
return conn
# mode=ro or immutable=1? # mode=ro or immutable=1?
if self.is_mutable: if self.is_mutable:
@ -169,7 +164,7 @@ class Database:
conn = sqlite3.connect( conn = sqlite3.connect(
f"file:{self.path}{qs}", uri=True, check_same_thread=False, **extra_kwargs f"file:{self.path}{qs}", uri=True, check_same_thread=False, **extra_kwargs
) )
self._all_connections.append(conn) self._all_file_connections.append(conn)
if self.is_temp_disk and not self._wal_enabled: if self.is_temp_disk and not self._wal_enabled:
conn.execute("PRAGMA journal_mode=WAL") conn.execute("PRAGMA journal_mode=WAL")
self._wal_enabled = True self._wal_enabled = True
@ -197,22 +192,23 @@ class Database:
write_thread.join(timeout=10) write_thread.join(timeout=10)
if write_thread.is_alive(): if write_thread.is_alive():
sys.stderr.write( sys.stderr.write(
f"Datasette: write thread for {self.name!r} did not exit within 10s\n" "Datasette: write thread for {!r} did not exit within 10s\n".format(
self.name
)
) )
sys.stderr.flush() sys.stderr.flush()
for future in pending_execute_futures: for future in pending_execute_futures:
try: try:
future.result() future.result()
except Exception: # noqa: BLE001, S110 except Exception:
# Shutdown teardown - a failed pending write must not block close()
pass pass
# Close anything still tracked in _all_connections # Close anything still tracked in _all_file_connections
for connection in self._all_connections: for connection in self._all_file_connections:
try: try:
connection.close() connection.close()
except Exception: # noqa: BLE001, S110 except Exception:
pass pass
self._all_connections = [] self._all_file_connections = []
# Drop per-thread cached read connections we can reach # Drop per-thread cached read connections we can reach
try: try:
delattr(connections, self._thread_local_id) delattr(connections, self._thread_local_id)
@ -222,13 +218,13 @@ class Database:
if self._read_connection is not None: if self._read_connection is not None:
try: try:
self._read_connection.close() self._read_connection.close()
except Exception: # noqa: BLE001, S110 except Exception:
pass pass
self._read_connection = None self._read_connection = None
if self._write_connection is not None: if self._write_connection is not None:
try: try:
self._write_connection.close() self._write_connection.close()
except Exception: # noqa: BLE001, S110 except Exception:
pass pass
self._write_connection = None self._write_connection = None
if self.is_temp_disk: if self.is_temp_disk:
@ -250,34 +246,19 @@ class Database:
request=None, request=None,
return_all=False, return_all=False,
returning_limit=EXECUTE_WRITE_RETURNING_LIMIT, returning_limit=EXECUTE_WRITE_RETURNING_LIMIT,
transaction=True,
time_limit_ms=2000,
): ):
self._check_not_closed() self._check_not_closed()
if returning_limit < 0: if returning_limit < 0:
raise ValueError("returning_limit must be >= 0") raise ValueError("returning_limit must be >= 0")
def execute_sql(conn): def _inner(conn):
cursor = conn.execute(sql, params or []) cursor = conn.execute(sql, params or [])
return ExecuteWriteResult.from_cursor( return ExecuteWriteResult.from_cursor(
cursor, return_all=return_all, returning_limit=returning_limit cursor, return_all=return_all, returning_limit=returning_limit
) )
def _inner(conn):
try:
if time_limit_ms is None:
return execute_sql(conn)
with sqlite_timelimit(conn, time_limit_ms):
return execute_sql(conn)
except (sqlite3.OperationalError, sqlite3.DatabaseError) as e:
if e.args == ("interrupted",):
raise QueryInterrupted(e, sql, params)
raise
with trace("sql", database=self.name, sql=sql.strip(), params=params): with trace("sql", database=self.name, sql=sql.strip(), params=params):
results = await self.execute_write_fn( results = await self.execute_write_fn(_inner, block=block, request=request)
_inner, block=block, request=request, transaction=transaction
)
return results return results
async def execute_write_script(self, sql, block=True, request=None): async def execute_write_script(self, sql, block=True, request=None):
@ -328,9 +309,9 @@ class Database:
finally: finally:
isolated_connection.close() isolated_connection.close()
try: try:
self._all_connections.remove(isolated_connection) self._all_file_connections.remove(isolated_connection)
except ValueError: except ValueError:
# May already have been cleared by close(). # Was probably a memory connection
pass pass
if self.ds.executor is None: if self.ds.executor is None:
@ -367,19 +348,9 @@ class Database:
self.ds._prepare_connection(self._write_connection, self.name) self.ds._prepare_connection(self._write_connection, self.name)
if transaction: if transaction:
with self._write_connection: with self._write_connection:
self._write_connection.execute("BEGIN IMMEDIATE")
result = fn(self._write_connection) result = fn(self._write_connection)
else: else:
result = fn(self._write_connection) result = fn(self._write_connection)
if not block:
# There is no write thread here, so the write has already
# finished. Hand back the same (task_id, reply_future) shape
# _send_to_write_thread() returns, with the future already
# resolved, so the block=False path below is identical in
# both modes.
reply_future = asyncio.get_running_loop().create_future()
reply_future.set_result(result)
result = (uuid.uuid4(), reply_future)
else: else:
result = await self._send_to_write_thread( result = await self._send_to_write_thread(
fn, block=block, transaction=transaction fn, block=block, transaction=transaction
@ -395,8 +366,7 @@ class Database:
async def _dispatch_events_after_write(): async def _dispatch_events_after_write():
try: try:
await reply_future await reply_future
except Exception: # noqa: BLE001 except Exception:
# The write failed; skip success events regardless of why
# if the write failed, don't emit success events # if the write failed, don't emit success events
return return
for event in pending_events: for event in pending_events:
@ -449,9 +419,11 @@ class Database:
self._write_thread = threading.Thread( self._write_thread = threading.Thread(
target=self._execute_writes, daemon=True target=self._execute_writes, daemon=True
) )
self._write_thread.name = f"_execute_writes for database {self.name}" self._write_thread.name = "_execute_writes for database {}".format(
self.name
)
self._write_thread.start() self._write_thread.start()
task_id = uuid.uuid4() task_id = uuid.uuid5(uuid.NAMESPACE_DNS, "datasette.io")
loop = asyncio.get_running_loop() loop = asyncio.get_running_loop()
reply_future = loop.create_future() reply_future = loop.create_future()
self._write_queue.put( self._write_queue.put(
@ -470,8 +442,7 @@ class Database:
try: try:
conn = self.connect(write=True) conn = self.connect(write=True)
self.ds._prepare_connection(conn, self.name) self.ds._prepare_connection(conn, self.name)
except Exception as e: # noqa: BLE001 except Exception as e:
# Stored and re-raised to whoever queues the next write
conn_exception = e conn_exception = e
while True: while True:
task = self._write_queue.get() task = self._write_queue.get()
@ -479,8 +450,7 @@ class Database:
if conn is not None: if conn is not None:
try: try:
conn.close() conn.close()
except Exception: # noqa: BLE001, S110 except Exception:
# Best-effort close as the write thread exits
pass pass
return return
exception = None exception = None
@ -495,25 +465,23 @@ class Database:
finally: finally:
isolated_connection.close() isolated_connection.close()
try: try:
self._all_connections.remove(isolated_connection) self._all_file_connections.remove(isolated_connection)
except ValueError: except ValueError:
# May already have been cleared by close(). # Was probably a memory connection
pass pass
except Exception as e: # noqa: BLE001 except Exception as e:
# Write thread must survive any task failure or the database wedges sys.stderr.write("{}\n".format(e))
sys.stderr.write(f"{e}\n")
sys.stderr.flush() sys.stderr.flush()
exception = e exception = e
else: else:
try: try:
if task.transaction: if task.transaction:
with conn: with conn:
conn.execute("BEGIN IMMEDIATE")
result = task.fn(conn) result = task.fn(conn)
else: else:
result = task.fn(conn) result = task.fn(conn)
except Exception as e: # noqa: BLE001 except Exception as e:
sys.stderr.write(f"{e}\n") sys.stderr.write("{}\n".format(e))
sys.stderr.flush() sys.stderr.flush()
exception = e exception = e
_deliver_write_result(task, result, exception) _deliver_write_result(task, result, exception)
@ -580,7 +548,9 @@ class Database:
raise QueryInterrupted(e, sql, params) raise QueryInterrupted(e, sql, params)
if log_sql_errors: if log_sql_errors:
sys.stderr.write( sys.stderr.write(
f"ERROR: conn={conn}, sql = {sql!r}, params = {params}: {e}\n" "ERROR: conn={}, sql = {}, params = {}: {}\n".format(
conn, repr(sql), params, e
)
) )
sys.stderr.flush() sys.stderr.flush()
raise raise
@ -633,7 +603,7 @@ class Database:
try: try:
table_count = ( table_count = (
await self.execute( await self.execute(
f"select count(*) from (select * from {escape_sqlite(table)} limit {self.count_limit + 1})", f"select count(*) from (select * from [{table}] limit {self.count_limit + 1})",
custom_time_limit=limit, custom_time_limit=limit,
) )
).rows[0][0] ).rows[0][0]
@ -737,9 +707,9 @@ class Database:
column_names column_names
and len(column_names) == 2 and len(column_names) == 2
and ("id" in column_names or "pk" in column_names) and ("id" in column_names or "pk" in column_names)
and set(column_names) != {"id", "pk"} and not set(column_names) == {"id", "pk"}
): ):
return next(c for c in column_names if c not in ("id", "pk")) return [c for c in column_names if c not in ("id", "pk")][0]
# Couldn't find a label: # Couldn't find a label:
return None return None
@ -785,17 +755,6 @@ class Database:
return hidden_tables return hidden_tables
async def derived_table_dependencies(self):
"""Return implementation tables and the tables they derive from."""
schema_version = (await self.execute("PRAGMA schema_version")).first()[0]
if (
self._cached_derived_table_dependencies is None
or self._cached_derived_table_dependencies[0] != schema_version
):
dependencies = await self.execute_fn(sqlite_derived_table_dependencies)
self._cached_derived_table_dependencies = (schema_version, dependencies)
return self._cached_derived_table_dependencies[1]
async def view_names(self): async def view_names(self):
results = await self.execute("select name from sqlite_master where type='view'") results = await self.execute("select name from sqlite_master where type='view'")
return [r[0] for r in results.rows] return [r[0] for r in results.rows]
@ -892,10 +851,10 @@ def _apply_write_wrapper(fn, wrapper_factory, track_event):
class WriteTask: class WriteTask:
__slots__ = ( __slots__ = (
"fn", "fn",
"isolated_connection", "task_id",
"loop", "loop",
"reply_future", "reply_future",
"task_id", "isolated_connection",
"transaction", "transaction",
) )
@ -936,7 +895,7 @@ class QueryInterrupted(Exception):
self.params = params self.params = params
def __str__(self): def __str__(self):
return f"QueryInterrupted: {self.e}" return "QueryInterrupted: {}".format(self.e)
class MultipleValues(Exception): class MultipleValues(Exception):

View file

@ -2,8 +2,8 @@ from datasette import hookimpl
from datasette.permissions import Action from datasette.permissions import Action
from datasette.resources import ( from datasette.resources import (
DatabaseResource, DatabaseResource,
QueryResource,
TableResource, TableResource,
QueryResource,
) )
@ -61,12 +61,6 @@ def register_actions():
description="Create tables", description="Create tables",
resource_class=DatabaseResource, resource_class=DatabaseResource,
), ),
Action(
name="create-view",
abbr="cv",
description="Create views",
resource_class=DatabaseResource,
),
Action( Action(
name="store-query", name="store-query",
abbr="sq", abbr="sq",
@ -117,12 +111,6 @@ def register_actions():
description="Drop tables", description="Drop tables",
resource_class=TableResource, resource_class=TableResource,
), ),
Action(
name="drop-view",
abbr="dv",
description="Drop views",
resource_class=TableResource,
),
# Query-level actions (child-level) # Query-level actions (child-level)
Action( Action(
name="view-query", name="view-query",

View file

@ -6,17 +6,6 @@ import markupsafe
from datasette import hookimpl from datasette import hookimpl
from datasette.column_types import ColumnType, SQLiteType from datasette.column_types import ColumnType, SQLiteType
_HTTP_URL_RE = re.compile(r"https?://\S+", re.IGNORECASE)
def _normalize_http_url(value):
if not isinstance(value, str):
return None
normalized = value.strip()
if not _HTTP_URL_RE.fullmatch(normalized):
return None
return normalized
class UrlColumnType(ColumnType): class UrlColumnType(ColumnType):
name = "url" name = "url"
@ -26,10 +15,7 @@ class UrlColumnType(ColumnType):
async def render_cell(self, value, column, table, database, datasette, request): async def render_cell(self, value, column, table, database, datasette, request):
if not value or not isinstance(value, str): if not value or not isinstance(value, str):
return None return None
normalized = _normalize_http_url(value) escaped = markupsafe.escape(value.strip())
if normalized is None:
return markupsafe.escape(value.strip())
escaped = markupsafe.escape(normalized)
return markupsafe.Markup(f'<a href="{escaped}">{escaped}</a>') return markupsafe.Markup(f'<a href="{escaped}">{escaped}</a>')
async def validate(self, value, datasette): async def validate(self, value, datasette):
@ -37,7 +23,7 @@ class UrlColumnType(ColumnType):
return None return None
if not isinstance(value, str): if not isinstance(value, str):
return "URL must be a string" return "URL must be a string"
if _normalize_http_url(value) is None: if not re.match(r"^https?://\S+$", value.strip()):
return "Invalid URL" return "Invalid URL"
return None return None

View file

@ -1,9 +1,8 @@
from datasette import hookimpl
import datetime import datetime
import os import os
import time import time
from datasette import hookimpl
def header(key, request): def header(key, request):
key = key.replace("_", "-").encode("utf-8") key = key.replace("_", "-").encode("utf-8")

View file

@ -17,29 +17,18 @@ UNION/INTERSECT operations. The order of evaluation is:
from __future__ import annotations from __future__ import annotations
from .config import config_permissions_sql as config_permissions_sql
from .defaults import (
DEFAULT_ALLOW_ACTIONS as DEFAULT_ALLOW_ACTIONS,
)
from .defaults import (
default_action_permissions_sql as default_action_permissions_sql,
)
from .defaults import (
# Avoid "datasette.default_permissions" does not explicitly export attribute
default_allow_sql_check as default_allow_sql_check,
)
from .defaults import (
default_query_permissions_sql as default_query_permissions_sql,
)
from .restrictions import (
ActorRestrictions as ActorRestrictions,
)
# Re-export all hooks and public utilities # Re-export all hooks and public utilities
from .restrictions import ( from .restrictions import (
actor_restrictions_sql as actor_restrictions_sql, actor_restrictions_sql as actor_restrictions_sql,
)
from .restrictions import (
restrictions_allow_action as restrictions_allow_action, restrictions_allow_action as restrictions_allow_action,
ActorRestrictions as ActorRestrictions,
) )
from .root import root_user_permissions_sql as root_user_permissions_sql from .root import root_user_permissions_sql as root_user_permissions_sql
from .config import config_permissions_sql as config_permissions_sql
from .defaults import (
# Avoid "datasette.default_permissions" does not explicitly export attribute
default_allow_sql_check as default_allow_sql_check,
default_action_permissions_sql as default_action_permissions_sql,
default_query_permissions_sql as default_query_permissions_sql,
DEFAULT_ALLOW_ACTIONS as DEFAULT_ALLOW_ACTIONS,
)

View file

@ -6,7 +6,7 @@ Applies permission rules from datasette.yaml configuration.
from __future__ import annotations from __future__ import annotations
from typing import TYPE_CHECKING, Any from typing import TYPE_CHECKING, Any, List, Optional, Set, Tuple
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -55,8 +55,8 @@ class ConfigPermissionProcessor:
def __init__( def __init__(
self, self,
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
): ):
self.datasette = datasette self.datasette = datasette
@ -74,8 +74,8 @@ class ConfigPermissionProcessor:
self.restrictions = actor.get("_r", {}) if actor else {} self.restrictions = actor.get("_r", {}) if actor else {}
# Pre-compute restriction info for efficiency # Pre-compute restriction info for efficiency
self.restricted_databases: set[str] = set() self.restricted_databases: Set[str] = set()
self.restricted_tables: set[tuple[str, str]] = set() self.restricted_tables: Set[Tuple[str, str]] = set()
if self.has_restrictions: if self.has_restrictions:
self.restricted_databases = { self.restricted_databases = {
@ -92,27 +92,16 @@ class ConfigPermissionProcessor:
# Tables implicitly reference their parent databases # Tables implicitly reference their parent databases
self.restricted_databases.update(db for db, _ in self.restricted_tables) self.restricted_databases.update(db for db, _ in self.restricted_tables)
# Resolve identity keys once per action, rather than scanning the def evaluate_allow_block(self, allow_block: Any) -> Optional[bool]:
# restriction allowlist for every configured table's allow block.
self.restricted_table_keys = {
(db, self.action_obj.normalize_child(table) if self.action_obj else table)
for db, table in self.restricted_tables
}
def evaluate_allow_block(self, allow_block: Any) -> bool | None:
"""Evaluate an allow block against the current actor.""" """Evaluate an allow block against the current actor."""
if allow_block is None: if allow_block is None:
return None return None
# Values passed using ``-s permissions.* 1`` or ``0`` are parsed as
# integers, but should retain the CLI's boolean 1/0 behavior.
if isinstance(allow_block, int) and allow_block in (0, 1):
return bool(allow_block)
return actor_matches_allow(self.actor, allow_block) return actor_matches_allow(self.actor, allow_block)
def is_in_restriction_allowlist( def is_in_restriction_allowlist(
self, self,
parent: str | None, parent: Optional[str],
child: str | None, child: Optional[str],
) -> bool: ) -> bool:
"""Check if resource is allowed by actor restrictions.""" """Check if resource is allowed by actor restrictions."""
if not self.has_restrictions: if not self.has_restrictions:
@ -132,10 +121,8 @@ class ConfigPermissionProcessor:
if parent: if parent:
table_restrictions = (self.restrictions.get("r", {}) or {}).get(parent, {}) table_restrictions = (self.restrictions.get("r", {}) or {}).get(parent, {})
if child: if child:
child_key = ( table_actions = table_restrictions.get(child, [])
self.action_obj.normalize_child(child) if self.action_obj else child if self.action_checks.intersection(table_actions):
)
if (parent, child_key) in self.restricted_table_keys:
return True return True
else: else:
# Parent query should proceed if any child in this database is allowlisted # Parent query should proceed if any child in this database is allowlisted
@ -156,9 +143,9 @@ class ConfigPermissionProcessor:
def add_permissions_rule( def add_permissions_rule(
self, self,
parent: str | None, parent: Optional[str],
child: str | None, child: Optional[str],
permissions_block: dict | None, permissions_block: Optional[dict],
scope_desc: str, scope_desc: str,
) -> None: ) -> None:
"""Add a rule from a permissions:{action} block.""" """Add a rule from a permissions:{action} block."""
@ -178,8 +165,8 @@ class ConfigPermissionProcessor:
def add_allow_block_rule( def add_allow_block_rule(
self, self,
parent: str | None, parent: Optional[str],
child: str | None, child: Optional[str],
allow_block: Any, allow_block: Any,
scope_desc: str, scope_desc: str,
) -> None: ) -> None:
@ -211,8 +198,8 @@ class ConfigPermissionProcessor:
def _add_restriction_gate_denies( def _add_restriction_gate_denies(
self, self,
parent: str | None, parent: Optional[str],
child: str | None, child: Optional[str],
is_allowed: bool, is_allowed: bool,
scope_desc: str, scope_desc: str,
) -> None: ) -> None:
@ -244,7 +231,7 @@ class ConfigPermissionProcessor:
if db_name == parent: if db_name == parent:
self.collector.add(db_name, table_name, False, reason) self.collector.add(db_name, table_name, False, reason)
def process(self) -> PermissionSQL | None: def process(self) -> Optional[PermissionSQL]:
"""Process all config rules and return combined PermissionSQL.""" """Process all config rules and return combined PermissionSQL."""
self._process_root_permissions() self._process_root_permissions()
self._process_databases() self._process_databases()
@ -434,10 +421,10 @@ class ConfigPermissionProcessor:
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def config_permissions_sql( async def config_permissions_sql(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
) -> list[PermissionSQL] | None: ) -> Optional[List[PermissionSQL]]:
""" """
Apply permission rules from datasette.yaml configuration. Apply permission rules from datasette.yaml configuration.

View file

@ -6,7 +6,7 @@ Provides default allow rules for standard view/execute actions.
from __future__ import annotations from __future__ import annotations
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, Optional
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -29,17 +29,18 @@ DEFAULT_ALLOW_ACTIONS = frozenset(
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def default_allow_sql_check( async def default_allow_sql_check(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
) -> PermissionSQL | None: ) -> Optional[PermissionSQL]:
""" """
Enforce the default_allow_sql setting. Enforce the default_allow_sql setting.
When default_allow_sql is false (the default), execute-sql is denied When default_allow_sql is false (the default), execute-sql is denied
unless explicitly allowed by config or other rules. unless explicitly allowed by config or other rules.
""" """
if action == "execute-sql" and not datasette.setting("default_allow_sql"): if action == "execute-sql":
if not datasette.setting("default_allow_sql"):
return PermissionSQL.deny(reason="default_allow_sql is false") return PermissionSQL.deny(reason="default_allow_sql is false")
return None return None
@ -47,10 +48,10 @@ async def default_allow_sql_check(
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def default_action_permissions_sql( async def default_action_permissions_sql(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
) -> PermissionSQL | None: ) -> Optional[PermissionSQL]:
""" """
Provide default allow rules for standard view/execute actions. Provide default allow rules for standard view/execute actions.
@ -70,10 +71,10 @@ async def default_action_permissions_sql(
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def default_query_permissions_sql( async def default_query_permissions_sql(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
) -> PermissionSQL | None: ) -> Optional[PermissionSQL]:
actor_id = actor.get("id") if isinstance(actor, dict) else None actor_id = actor.get("id") if isinstance(actor, dict) else None
if action not in {"view-query", "update-query", "delete-query"}: if action not in {"view-query", "update-query", "delete-query"}:

View file

@ -5,7 +5,7 @@ Shared helper utilities for default permission implementations.
from __future__ import annotations from __future__ import annotations
from dataclasses import dataclass from dataclasses import dataclass
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, List, Optional, Set
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -13,7 +13,7 @@ if TYPE_CHECKING:
from datasette.permissions import PermissionSQL from datasette.permissions import PermissionSQL
def get_action_name_variants(datasette: Datasette, action: str) -> set[str]: def get_action_name_variants(datasette: "Datasette", action: str) -> Set[str]:
""" """
Get all name variants for an action (full name and abbreviation). Get all name variants for an action (full name and abbreviation).
@ -27,7 +27,7 @@ def get_action_name_variants(datasette: Datasette, action: str) -> set[str]:
return variants return variants
def action_in_list(datasette: Datasette, action: str, action_list: list) -> bool: def action_in_list(datasette: "Datasette", action: str, action_list: list) -> bool:
"""Check if an action (or its abbreviation) is in a list.""" """Check if an action (or its abbreviation) is in a list."""
return bool(get_action_name_variants(datasette, action).intersection(action_list)) return bool(get_action_name_variants(datasette, action).intersection(action_list))
@ -36,8 +36,8 @@ def action_in_list(datasette: Datasette, action: str, action_list: list) -> bool
class PermissionRow: class PermissionRow:
"""A single permission rule row.""" """A single permission rule row."""
parent: str | None parent: Optional[str]
child: str | None child: Optional[str]
allow: bool allow: bool
reason: str reason: str
@ -46,14 +46,14 @@ class PermissionRowCollector:
"""Collects permission rows and converts them to PermissionSQL.""" """Collects permission rows and converts them to PermissionSQL."""
def __init__(self, prefix: str = "row"): def __init__(self, prefix: str = "row"):
self.rows: list[PermissionRow] = [] self.rows: List[PermissionRow] = []
self.prefix = prefix self.prefix = prefix
def add( def add(
self, self,
parent: str | None, parent: Optional[str],
child: str | None, child: Optional[str],
allow: bool | None, allow: Optional[bool],
reason: str, reason: str,
if_not_none: bool = False, if_not_none: bool = False,
) -> None: ) -> None:
@ -62,7 +62,7 @@ class PermissionRowCollector:
return return
self.rows.append(PermissionRow(parent, child, allow, reason)) self.rows.append(PermissionRow(parent, child, allow, reason))
def to_permission_sql(self) -> PermissionSQL | None: def to_permission_sql(self) -> Optional[PermissionSQL]:
"""Convert collected rows to a PermissionSQL object.""" """Convert collected rows to a PermissionSQL object."""
if not self.rows: if not self.rows:
return None return None

View file

@ -8,7 +8,7 @@ contains allowlists of resources the actor can access.
from __future__ import annotations from __future__ import annotations
from dataclasses import dataclass from dataclasses import dataclass
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, List, Optional, Set, Tuple
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -23,12 +23,12 @@ from .helpers import action_in_list, get_action_name_variants
class ActorRestrictions: class ActorRestrictions:
"""Parsed actor restrictions from the _r key.""" """Parsed actor restrictions from the _r key."""
global_actions: list[str] # _r.a - globally allowed actions global_actions: List[str] # _r.a - globally allowed actions
database_actions: dict # _r.d - {db_name: [actions]} database_actions: dict # _r.d - {db_name: [actions]}
table_actions: dict # _r.r - {db_name: {table: [actions]}} table_actions: dict # _r.r - {db_name: {table: [actions]}}
@classmethod @classmethod
def from_actor(cls, actor: dict | None) -> ActorRestrictions | None: def from_actor(cls, actor: Optional[dict]) -> Optional["ActorRestrictions"]:
"""Parse restrictions from actor dict. Returns None if no restrictions.""" """Parse restrictions from actor dict. Returns None if no restrictions."""
if not actor: if not actor:
return None return None
@ -44,11 +44,11 @@ class ActorRestrictions:
table_actions=restrictions.get("r", {}), table_actions=restrictions.get("r", {}),
) )
def is_action_globally_allowed(self, datasette: Datasette, action: str) -> bool: def is_action_globally_allowed(self, datasette: "Datasette", action: str) -> bool:
"""Check if action is in the global allowlist.""" """Check if action is in the global allowlist."""
return action_in_list(datasette, action, self.global_actions) return action_in_list(datasette, action, self.global_actions)
def get_allowed_databases(self, datasette: Datasette, action: str) -> set[str]: def get_allowed_databases(self, datasette: "Datasette", action: str) -> Set[str]:
"""Get database names where this action is allowed.""" """Get database names where this action is allowed."""
allowed = set() allowed = set()
for db_name, db_actions in self.database_actions.items(): for db_name, db_actions in self.database_actions.items():
@ -57,8 +57,8 @@ class ActorRestrictions:
return allowed return allowed
def get_allowed_tables( def get_allowed_tables(
self, datasette: Datasette, action: str self, datasette: "Datasette", action: str
) -> set[tuple[str, str]]: ) -> Set[Tuple[str, str]]:
"""Get (database, table) pairs where this action is allowed.""" """Get (database, table) pairs where this action is allowed."""
allowed = set() allowed = set()
for db_name, tables in self.table_actions.items(): for db_name, tables in self.table_actions.items():
@ -70,10 +70,10 @@ class ActorRestrictions:
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def actor_restrictions_sql( async def actor_restrictions_sql(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
action: str, action: str,
) -> list[PermissionSQL] | None: ) -> Optional[List[PermissionSQL]]:
""" """
Handle actor restriction-based permission rules. Handle actor restriction-based permission rules.
@ -140,10 +140,10 @@ async def actor_restrictions_sql(
def restrictions_allow_action( def restrictions_allow_action(
datasette: Datasette, datasette: "Datasette",
restrictions: dict, restrictions: dict,
action: str, action: str,
resource: str | tuple[str, str] | None, resource: Optional[str | Tuple[str, str]],
) -> bool: ) -> bool:
""" """
Check if restrictions allow the requested action on the requested resource. Check if restrictions allow the requested action on the requested resource.
@ -185,12 +185,8 @@ def restrictions_allow_action(
# Check table/resource level # Check table/resource level
if resource is not None and not isinstance(resource, str) and len(resource) == 2: if resource is not None and not isinstance(resource, str) and len(resource) == 2:
database, table = resource database, table = resource
action_obj = datasette.actions.get(action) table_allowed = restrictions.get("r", {}).get(database, {}).get(table)
normalize = action_obj.normalize_child if action_obj else lambda name: name if table_allowed is not None:
for table_name, table_allowed in (
restrictions.get("r", {}).get(database, {}).items()
):
if normalize(table_name) == normalize(table):
assert isinstance(table_allowed, list) assert isinstance(table_allowed, list)
if to_check.intersection(table_allowed): if to_check.intersection(table_allowed):
return True return True

View file

@ -6,7 +6,7 @@ Grants full permissions to the root user when --root flag is used.
from __future__ import annotations from __future__ import annotations
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, Optional
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -17,9 +17,9 @@ from datasette.permissions import PermissionSQL
@hookimpl(specname="permission_resources_sql") @hookimpl(specname="permission_resources_sql")
async def root_user_permissions_sql( async def root_user_permissions_sql(
datasette: Datasette, datasette: "Datasette",
actor: dict | None, actor: Optional[dict],
) -> PermissionSQL | None: ) -> Optional[PermissionSQL]:
""" """
Grant root user full permissions when --root flag is used. Grant root user full permissions when --root flag is used.
""" """

View file

@ -1,25 +0,0 @@
"""Default table-access policy for SQLite optimizer statistics."""
import json
from datasette import hookimpl
from datasette.permissions import PermissionSQL
@hookimpl
def permission_resources_sql(action):
if action != "view-table":
return None
return PermissionSQL(
sql="""
SELECT database_name AS parent, value AS child, 0 AS allow,
'SQLite statistics tables are denied by default' AS reason
FROM catalog_databases
CROSS JOIN json_each(:sqlite_statistics_names)
""",
params={
"sqlite_statistics_names": json.dumps(
["sqlite_stat1", "sqlite_stat2", "sqlite_stat3", "sqlite_stat4"]
)
},
)

View file

@ -7,7 +7,7 @@ to datasette.verify_token() so all registered handlers are tried.
from __future__ import annotations from __future__ import annotations
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, Optional
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
@ -17,13 +17,15 @@ from datasette.tokens import SignedTokenHandler
@hookimpl @hookimpl
def register_token_handler(datasette: Datasette): def register_token_handler(datasette: "Datasette"):
"""Register the default signed token handler.""" """Register the default signed token handler."""
return SignedTokenHandler() return SignedTokenHandler()
@hookimpl(specname="actor_from_request") @hookimpl(specname="actor_from_request")
async def actor_from_signed_api_token(datasette: Datasette, request) -> dict | None: async def actor_from_signed_api_token(
datasette: "Datasette", request
) -> Optional[dict]:
""" """
Authenticate requests using API tokens by delegating to all registered Authenticate requests using API tokens by delegating to all registered
token handlers via datasette.verify_token(). token handlers via datasette.verify_token().

View file

@ -1,29 +0,0 @@
from datasette import hookimpl
from datasette.resources import TableResource
@hookimpl
def table_actions(datasette, actor, database, table, request):
async def inner():
db = datasette.get_database(database)
if not db.is_mutable:
return []
if not await datasette.allowed(
action="alter-table",
resource=TableResource(database=database, table=table),
actor=actor,
):
return []
return [
{
"type": "button",
"label": "Alter table",
"description": "Change columns and primary key for this table.",
"attrs": {
"aria-label": f"Alter table {table}",
"data-table-action": "alter-table",
},
}
]
return inner

View file

@ -1,8 +1,7 @@
from abc import ABC, abstractproperty from abc import ABC, abstractproperty
from dataclasses import asdict, dataclass, field from dataclasses import asdict, dataclass, field
from datetime import datetime, timezone
from datasette.hookspecs import hookimpl from datasette.hookspecs import hookimpl
from datetime import datetime, timezone
@dataclass @dataclass

View file

@ -5,8 +5,6 @@ from typing import ClassVar
from asyncinject import Registry from asyncinject import Registry
from datasette.utils.asgi import BadRequest
def extra_names_from_request(request): def extra_names_from_request(request):
extra_bits = request.args.getlist("_extra") extra_bits = request.args.getlist("_extra")
@ -83,16 +81,6 @@ class ExtraRegistry:
def public_classes_for_scope(self, scope): def public_classes_for_scope(self, scope):
return self.classes_for_scope(scope, include_internal=False) return self.classes_for_scope(scope, include_internal=False)
def internal_classes_for_scope(self, scope):
# Extras that are available to HTML templates but excluded from
# JSON responses - plain Providers are dependency plumbing and
# never surface as keys, so they are not included
return [
cls
for cls in self.classes_for_scope(scope)
if issubclass(cls, Extra) and not cls.public
]
def _registry_for_scope(self, scope): def _registry_for_scope(self, scope):
registry = self._scope_registries.get(scope) registry = self._scope_registries.get(scope)
if registry is None: if registry is None:
@ -115,17 +103,6 @@ class ExtraRegistry:
self._allowed_names[key] = names self._allowed_names[key] = names
return names return names
def validate_requested(self, requested, scope):
"""
Raise BadRequest if any requested extra name is not a public extra
for this scope. Used by data formats such as .json - HTML pages
silently ignore unknown names instead.
"""
allowed = self._allowed_names_for_scope(scope, include_internal=False)
unknown = sorted(name for name in requested if name not in allowed)
if unknown:
raise BadRequest("Unknown _extra: {}".format(", ".join(unknown)))
async def resolve(self, requested, context, scope, include_internal=False): async def resolve(self, requested, context, scope, include_internal=False):
allowed_names = self._allowed_names_for_scope(scope, include_internal) allowed_names = self._allowed_names_for_scope(scope, include_internal)
requested_names = [name for name in requested if name in allowed_names] requested_names = [name for name in requested if name in allowed_names]

View file

@ -1,13 +1,12 @@
import json import json
import urllib import urllib
from datasette import hookimpl from datasette import hookimpl
from datasette.database import QueryInterrupted from datasette.database import QueryInterrupted
from datasette.utils import ( from datasette.utils import (
detect_json1,
escape_sqlite, escape_sqlite,
path_with_added_args, path_with_added_args,
path_with_removed_args, path_with_removed_args,
detect_json1,
sqlite3, sqlite3,
) )
@ -31,7 +30,7 @@ def load_facet_configs(request, table_config):
assert ( assert (
len(facet_config.values()) == 1 len(facet_config.values()) == 1
), "Metadata config dicts should be {type: config}" ), "Metadata config dicts should be {type: config}"
type, facet_config = next(iter(facet_config.items())) type, facet_config = list(facet_config.items())[0]
if isinstance(facet_config, str): if isinstance(facet_config, str):
facet_config = {"simple": facet_config} facet_config = {"simple": facet_config}
facet_configs.setdefault(type, []).append( facet_configs.setdefault(type, []).append(
@ -86,7 +85,7 @@ class Facet:
self.database = database self.database = database
# For foreign key expansion. Can be None for e.g. stored SQL queries: # For foreign key expansion. Can be None for e.g. stored SQL queries:
self.table = table self.table = table
self.sql = sql or f"select * from {escape_sqlite(table)}" self.sql = sql or f"select * from [{table}]"
self.params = params or [] self.params = params or []
self.table_config = table_config self.table_config = table_config
# row_count can be None, in which case we calculate it ourselves: # row_count can be None, in which case we calculate it ourselves:
@ -161,13 +160,18 @@ class ColumnFacet(Facet):
for column in columns: for column in columns:
if column in already_enabled: if column in already_enabled:
continue continue
suggested_facet_sql = f""" suggested_facet_sql = """
with limited as (select * from ({self.sql}) limit {self.suggest_consider}) with limited as (select * from ({sql}) limit {suggest_consider})
select {escape_sqlite(column)} as value, count(*) as n from limited select {column} as value, count(*) as n from limited
where value is not null where value is not null
group by value group by value
limit {facet_size + 1} limit {limit}
""" """.format(
column=escape_sqlite(column),
sql=self.sql,
limit=facet_size + 1,
suggest_consider=self.suggest_consider,
)
distinct_values = None distinct_values = None
try: try:
distinct_values = await self.ds.execute( distinct_values = await self.ds.execute(
@ -263,16 +267,11 @@ class ColumnFacet(Facet):
for row in facet_rows: for row in facet_rows:
column_qs = column column_qs = column
if column.startswith("_"): if column.startswith("_"):
column_qs = f"{column}__exact" column_qs = "{}__exact".format(column)
selected_args = { selected = (column_qs, str(row["value"])) in qs_pairs
key: str(row["value"])
for key in (column_qs, f"{column}__exact")
if (key, str(row["value"])) in qs_pairs
}
selected = bool(selected_args)
if selected: if selected:
toggle_path = path_with_removed_args( toggle_path = path_with_removed_args(
self.request, selected_args self.request, {column_qs: str(row["value"])}
) )
else: else:
toggle_path = path_with_added_args( toggle_path = path_with_added_args(
@ -343,12 +342,12 @@ class ArrayFacet(Facet):
for v in await self.ds.execute( for v in await self.ds.execute(
self.database, self.database,
( (
f"select {escape_sqlite(column)} from ({self.sql}) " "select {column} from ({sql}) "
f"where {escape_sqlite(column)} is not null " "where {column} is not null "
f"and {escape_sqlite(column)} != '' " "and {column} != '' "
f"and json_array_length({escape_sqlite(column)}) > 0 " "and json_array_length({column}) > 0 "
"limit 100" "limit 100"
), ).format(column=escape_sqlite(column), sql=self.sql),
self.params, self.params,
truncate=False, truncate=False,
custom_time_limit=self.ds.setting( custom_time_limit=self.ds.setting(
@ -389,14 +388,14 @@ class ArrayFacet(Facet):
source = source_and_config["source"] source = source_and_config["source"]
column = config.get("column") or config["simple"] column = config.get("column") or config["simple"]
# https://github.com/simonw/datasette/issues/448 # https://github.com/simonw/datasette/issues/448
facet_sql = f""" facet_sql = """
with inner as ({self.sql}), with inner as ({sql}),
deduped_array_items as ( deduped_array_items as (
select select
distinct j.value, distinct j.value,
inner.* inner.*
from from
json_each([inner].{escape_sqlite(column)}) j json_each([inner].{col}) j
join inner join inner
) )
select select
@ -407,8 +406,12 @@ class ArrayFacet(Facet):
group by group by
value value
order by order by
count(*) desc, value limit {facet_size + 1} count(*) desc, value limit {limit}
""" """.format(
col=escape_sqlite(column),
sql=self.sql,
limit=facet_size + 1,
)
try: try:
facet_rows_results = await self.ds.execute( facet_rows_results = await self.ds.execute(
self.database, self.database,

View file

@ -1,12 +1,8 @@
import json
import math
from typing import ClassVar
from datasette import hookimpl from datasette import hookimpl
from datasette.resources import DatabaseResource, TableResource from datasette.resources import DatabaseResource
from datasette.utils.asgi import BadRequest
from datasette.views.base import DatasetteError from datasette.views.base import DatasetteError
from datasette.utils.asgi import BadRequest
import json
from .utils import detect_json1, escape_sqlite, path_with_removed_args from .utils import detect_json1, escape_sqlite, path_with_removed_args
@ -52,20 +48,13 @@ def search_filters(request, database, table, datasette):
human_descriptions = [] human_descriptions = []
extra_context = {} extra_context = {}
# Figure out which trusted fts_table to use. Query string parameters can # Figure out which fts_table to use
# repeat this mapping (for backwards compatibility), but must not select
# a different table or primary key.
table_metadata = await datasette.table_config(database, table) table_metadata = await datasette.table_config(database, table)
db = datasette.get_database(database) db = datasette.get_database(database)
fts_table = table_metadata.get("fts_table") fts_table = request.args.get("_fts_table")
fts_table = fts_table or table_metadata.get("fts_table")
fts_table = fts_table or await db.fts_table(table) fts_table = fts_table or await db.fts_table(table)
fts_pk = table_metadata.get("fts_pk", "rowid") fts_pk = request.args.get("_fts_pk", table_metadata.get("fts_pk", "rowid"))
requested_fts_table = request.args.get("_fts_table")
requested_fts_pk = request.args.get("_fts_pk")
if (requested_fts_table and requested_fts_table != fts_table) or (
requested_fts_pk and requested_fts_pk != fts_pk
):
raise BadRequest("Invalid _fts_table or _fts_pk")
search_args = { search_args = {
key: request.args[key] key: request.args[key]
for key in request.args for key in request.args
@ -83,11 +72,6 @@ def search_filters(request, database, table, datasette):
extra_context["supports_search"] = bool(fts_table) extra_context["supports_search"] = bool(fts_table)
if fts_table and search_args: if fts_table and search_args:
await datasette.ensure_permission(
action="view-table",
resource=TableResource(database=database, table=fts_table),
actor=request.actor,
)
if "_search" in search_args: if "_search" in search_args:
# Simple ?_search=xxx # Simple ?_search=xxx
search = search_args["_search"] search = search_args["_search"]
@ -115,9 +99,9 @@ def search_filters(request, database, table, datasette):
fts_table=escape_sqlite(fts_table), fts_table=escape_sqlite(fts_table),
search_col=escape_sqlite(search_col), search_col=escape_sqlite(search_col),
match_clause=( match_clause=(
f":search_{i}" ":search_{}".format(i)
if search_mode_raw if search_mode_raw
else f"escape_fts(:search_{i})" else "escape_fts(:search_{})".format(i)
), ),
) )
) )
@ -148,18 +132,13 @@ def through_filters(request, database, table, datasette):
through_table = through_data["table"] through_table = through_data["table"]
other_column = through_data["column"] other_column = through_data["column"]
value = through_data["value"] value = through_data["value"]
await datasette.ensure_permission(
action="view-table",
resource=TableResource(database=database, table=through_table),
actor=request.actor,
)
db = datasette.get_database(database) db = datasette.get_database(database)
outgoing_foreign_keys = await db.foreign_keys_for_table(through_table) outgoing_foreign_keys = await db.foreign_keys_for_table(through_table)
fk_to_us = next( try:
(fk for fk in outgoing_foreign_keys if fk["other_table"] == table), fk_to_us = [
None, fk for fk in outgoing_foreign_keys if fk["other_table"] == table
) ][0]
if fk_to_us is None: except IndexError:
raise DatasetteError( raise DatasetteError(
"Invalid _through - could not find corresponding foreign key" "Invalid _through - could not find corresponding foreign key"
) )
@ -203,17 +182,6 @@ class Filter:
raise NotImplementedError raise NotImplementedError
def _coerce_numeric_filter_value(value):
try:
return int(value)
except ValueError:
try:
converted = float(value)
except ValueError:
return value
return converted if math.isfinite(converted) else value
class TemplatedFilter(Filter): class TemplatedFilter(Filter):
def __init__( def __init__(
self, self,
@ -235,17 +203,13 @@ class TemplatedFilter(Filter):
def where_clause(self, table, column, value, param_counter): def where_clause(self, table, column, value, param_counter):
converted = self.format.format(value) converted = self.format.format(value)
if self.numeric: if self.numeric and converted.isdigit():
converted = _coerce_numeric_filter_value(converted) converted = int(converted)
if self.no_argument: if self.no_argument:
kwargs = {"c": _quote_sqlite_identifier(column)} kwargs = {"c": column}
converted = None converted = None
else: else:
kwargs = { kwargs = {"c": column, "p": f"p{param_counter}", "t": table}
"c": _quote_sqlite_identifier(column),
"p": f"p{param_counter}",
"t": _quote_sqlite_identifier(table),
}
return self.sql_template.format(**kwargs), converted return self.sql_template.format(**kwargs), converted
def human_clause(self, column, value): def human_clause(self, column, value):
@ -259,14 +223,6 @@ class TemplatedFilter(Filter):
return template.format(c=column, v=value) return template.format(c=column, v=value)
def _quote_sqlite_identifier(identifier):
# Preserve the historic always-quoted SQL generated by TemplatedFilter.
escaped = escape_sqlite(identifier)
if escaped == identifier:
return f'"{identifier}"'
return escaped
class InFilter(Filter): class InFilter(Filter):
key = "in" key = "in"
display = "in" display = "in"
@ -308,56 +264,56 @@ class Filters:
TemplatedFilter( TemplatedFilter(
"exact", "exact",
"=", "=",
"{c} = :{p}", '"{c}" = :{p}',
lambda c, v: "{c} = {v}" if v.isdigit() else '{c} = "{v}"', lambda c, v: "{c} = {v}" if v.isdigit() else '{c} = "{v}"',
), ),
TemplatedFilter( TemplatedFilter(
"not", "not",
"!=", "!=",
"{c} != :{p}", '"{c}" != :{p}',
lambda c, v: "{c} != {v}" if v.isdigit() else '{c} != "{v}"', lambda c, v: "{c} != {v}" if v.isdigit() else '{c} != "{v}"',
), ),
TemplatedFilter( TemplatedFilter(
"contains", "contains",
"contains", "contains",
"{c} like :{p}", '"{c}" like :{p}',
'{c} contains "{v}"', '{c} contains "{v}"',
format="%{}%", format="%{}%",
), ),
TemplatedFilter( TemplatedFilter(
"notcontains", "notcontains",
"does not contain", "does not contain",
"{c} not like :{p}", '"{c}" not like :{p}',
'{c} does not contain "{v}"', '{c} does not contain "{v}"',
format="%{}%", format="%{}%",
), ),
TemplatedFilter( TemplatedFilter(
"endswith", "endswith",
"ends with", "ends with",
"{c} like :{p}", '"{c}" like :{p}',
'{c} ends with "{v}"', '{c} ends with "{v}"',
format="%{}", format="%{}",
), ),
TemplatedFilter( TemplatedFilter(
"startswith", "startswith",
"starts with", "starts with",
"{c} like :{p}", '"{c}" like :{p}',
'{c} starts with "{v}"', '{c} starts with "{v}"',
format="{}%", format="{}%",
), ),
TemplatedFilter("gt", ">", "{c} > :{p}", "{c} > {v}", numeric=True), TemplatedFilter("gt", ">", '"{c}" > :{p}', "{c} > {v}", numeric=True),
TemplatedFilter( TemplatedFilter(
"gte", "\u2265", "{c} >= :{p}", "{c} \u2265 {v}", numeric=True "gte", "\u2265", '"{c}" >= :{p}', "{c} \u2265 {v}", numeric=True
), ),
TemplatedFilter("lt", "<", "{c} < :{p}", "{c} < {v}", numeric=True), TemplatedFilter("lt", "<", '"{c}" < :{p}', "{c} < {v}", numeric=True),
TemplatedFilter( TemplatedFilter(
"lte", "\u2264", "{c} <= :{p}", "{c} \u2264 {v}", numeric=True "lte", "\u2264", '"{c}" <= :{p}', "{c} \u2264 {v}", numeric=True
), ),
TemplatedFilter("like", "like", "{c} like :{p}", '{c} like "{v}"'), TemplatedFilter("like", "like", '"{c}" like :{p}', '{c} like "{v}"'),
TemplatedFilter( TemplatedFilter(
"notlike", "not like", "{c} not like :{p}", '{c} not like "{v}"' "notlike", "not like", '"{c}" not like :{p}', '{c} not like "{v}"'
), ),
TemplatedFilter("glob", "glob", "{c} glob :{p}", '{c} glob "{v}"'), TemplatedFilter("glob", "glob", '"{c}" glob :{p}', '{c} glob "{v}"'),
InFilter(), InFilter(),
NotInFilter(), NotInFilter(),
] ]
@ -366,13 +322,13 @@ class Filters:
TemplatedFilter( TemplatedFilter(
"arraycontains", "arraycontains",
"array contains", "array contains",
""":{p} in (select value from json_each({t}.{c}))""", """:{p} in (select value from json_each([{t}].[{c}]))""",
'{c} contains "{v}"', '{c} contains "{v}"',
), ),
TemplatedFilter( TemplatedFilter(
"arraynotcontains", "arraynotcontains",
"array does not contain", "array does not contain",
""":{p} not in (select value from json_each({t}.{c}))""", """:{p} not in (select value from json_each([{t}].[{c}]))""",
'{c} does not contain "{v}"', '{c} does not contain "{v}"',
), ),
] ]
@ -380,34 +336,36 @@ class Filters:
else [] else []
) )
+ [ + [
TemplatedFilter("date", "date", "date({c}) = :{p}", '"{c}" is on date {v}'),
TemplatedFilter( TemplatedFilter(
"isnull", "is null", "{c} is null", "{c} is null", no_argument=True "date", "date", 'date("{c}") = :{p}', '"{c}" is on date {v}'
),
TemplatedFilter(
"isnull", "is null", '"{c}" is null', "{c} is null", no_argument=True
), ),
TemplatedFilter( TemplatedFilter(
"notnull", "notnull",
"is not null", "is not null",
"{c} is not null", '"{c}" is not null',
"{c} is not null", "{c} is not null",
no_argument=True, no_argument=True,
), ),
TemplatedFilter( TemplatedFilter(
"isblank", "isblank",
"is blank", "is blank",
"({c} is null or {c} = '')", '("{c}" is null or "{c}" = "")',
"{c} is blank", "{c} is blank",
no_argument=True, no_argument=True,
), ),
TemplatedFilter( TemplatedFilter(
"notblank", "notblank",
"is not blank", "is not blank",
"({c} is not null and {c} != '')", '("{c}" is not null and "{c}" != "")',
"{c} is not blank", "{c} is not blank",
no_argument=True, no_argument=True,
), ),
] ]
) )
_filters_by_key: ClassVar[dict[str, Filter]] = {f.key: f for f in _filters} _filters_by_key = {f.key: f for f in _filters}
def __init__(self, pairs): def __init__(self, pairs):
self.pairs = pairs self.pairs = pairs

View file

@ -1,10 +1,9 @@
from datasette.utils.sqlite import sqlite3
from datasette.utils import documented
import itertools import itertools
import random import random
import string import string
from datasette.utils import documented
from datasette.utils.sqlite import sqlite3
__all__ = [ __all__ = [
"EXTRA_DATABASE_SQL", "EXTRA_DATABASE_SQL",
"TABLES", "TABLES",
@ -347,7 +346,9 @@ CREATE VIEW searchable_view_configured_by_metadata AS
+ '\nINSERT INTO no_primary_key VALUES ("RENDER_CELL_DEMO", "a202", "b202", "c202");\n' + '\nINSERT INTO no_primary_key VALUES ("RENDER_CELL_DEMO", "a202", "b202", "c202");\n'
+ "\n".join( + "\n".join(
[ [
f'INSERT INTO compound_three_primary_keys VALUES ("{a}", "{b}", "{c}", "{content}");' 'INSERT INTO compound_three_primary_keys VALUES ("{a}", "{b}", "{c}", "{content}");'.format(
a=a, b=b, c=c, content=content
)
for a, b, c, content in generate_compound_rows(1001) for a, b, c, content in generate_compound_rows(1001)
] ]
) )

View file

@ -1,20 +1,9 @@
from datasette import Response, hookimpl from datasette import hookimpl, Response
from .utils import add_cors_headers
@hookimpl(trylast=True) @hookimpl(trylast=True)
def forbidden(datasette, request, message): def forbidden(datasette, request, message):
async def inner(): async def inner():
if (
request.path.split("?")[0].endswith(".json")
or "application/json" in (request.headers.get("accept") or "")
or request.headers.get("content-type") == "application/json"
):
headers = {}
if datasette.cors:
add_cors_headers(headers)
return Response.error(message, 403, headers=headers)
return Response.html( return Response.html(
await datasette.render_template( await datasette.render_template(
"error.html", "error.html",

View file

@ -1,21 +1,16 @@
import traceback from datasette import hookimpl, Response
from .utils import add_cors_headers
from markupsafe import Markup
from datasette import Response, hookimpl
from .utils import add_cors_headers, error_body
from .utils.asgi import ( from .utils.asgi import (
Base400, Base400,
) )
from .views.base import DatasetteError from .views.base import DatasetteError
from markupsafe import Markup
import traceback
# Debugger imports are deliberate - they back the "pdb" setting, which drops
# into a debugger on unhandled exceptions
try: try:
import ipdb as pdb # noqa: T100 import ipdb as pdb
except ImportError: except ImportError:
import pdb # noqa: T100 import pdb
try: try:
import rich import rich
@ -33,7 +28,6 @@ def handle_exception(datasette, request, exception):
rich.get_console().print_exception(show_locals=True) rich.get_console().print_exception(show_locals=True)
title = None title = None
plain_message = None
if isinstance(exception, Base400): if isinstance(exception, Base400):
status = exception.status status = exception.status
info = {} info = {}
@ -42,7 +36,6 @@ def handle_exception(datasette, request, exception):
status = exception.status status = exception.status
info = exception.error_dict info = exception.error_dict
message = exception.message message = exception.message
plain_message = exception.plain_message
if exception.message_is_html: if exception.message_is_html:
message = Markup(message) message = Markup(message)
title = exception.title title = exception.title
@ -52,17 +45,6 @@ def handle_exception(datasette, request, exception):
message = str(exception) message = str(exception)
traceback.print_exc() traceback.print_exc()
templates = [f"{status}.html", "error.html"] templates = [f"{status}.html", "error.html"]
headers = {}
if datasette.cors:
add_cors_headers(headers)
if request.path.split("?")[0].endswith(".json"):
body = dict(info)
body.update(error_body(plain_message or message, status))
return Response.json(body, status=status, headers=headers)
if request.path.split("?")[0].endswith(".csv"):
return Response.text(
plain_message or message, status=status, headers=headers
)
info.update( info.update(
{ {
"ok": False, "ok": False,
@ -71,15 +53,18 @@ def handle_exception(datasette, request, exception):
"title": title, "title": title,
} }
) )
environment = datasette.get_jinja_environment(request) headers = {}
template = environment.select_template(templates) if datasette.cors:
add_cors_headers(headers)
if request.path.split("?")[0].endswith(".json"):
return Response.json(info, status=status, headers=headers)
else:
return Response.html( return Response.html(
await template.render_async( await datasette.render_template(
dict( templates,
info, info,
urls=datasette.urls, request=request,
menu_links=list, view_name="error",
)
), ),
status=status, status=status,
headers=headers, headers=headers,

View file

@ -1,4 +1,5 @@
from pluggy import HookimplMarker, HookspecMarker from pluggy import HookimplMarker
from pluggy import HookspecMarker
hookspec = HookspecMarker("datasette") hookspec = HookspecMarker("datasette")
hookimpl = HookimplMarker("datasette") hookimpl = HookimplMarker("datasette")
@ -9,11 +10,6 @@ def startup(datasette):
"""Fires directly after Datasette first starts running""" """Fires directly after Datasette first starts running"""
@hookspec
def shutdown(datasette):
"""Called once when the Datasette server is shutting down"""
@hookspec @hookspec
def asgi_wrapper(datasette): def asgi_wrapper(datasette):
"""Returns an ASGI middleware callable to wrap our ASGI application with""" """Returns an ASGI middleware callable to wrap our ASGI application with"""
@ -50,7 +46,7 @@ def extra_body_script(
def extra_template_vars( def extra_template_vars(
template, database, table, columns, view_name, request, datasette template, database, table, columns, view_name, request, datasette
): ):
"""Extra template variables to be made available to the template - can return dict, None, callable or awaitable""" """Extra template variables to be made available to the template - can return dict or callable or awaitable"""
@hookspec @hookspec

View file

@ -1,13 +1,13 @@
import hashlib import hashlib
from .utils import ( from .utils import (
detect_spatialite,
detect_fts, detect_fts,
detect_primary_keys, detect_primary_keys,
detect_spatialite,
escape_sqlite, escape_sqlite,
get_all_foreign_keys, get_all_foreign_keys,
sqlite3,
table_columns, table_columns,
sqlite3,
) )
HASH_BLOCK_SIZE = 1024 * 1024 HASH_BLOCK_SIZE = 1024 * 1024
@ -95,10 +95,10 @@ def inspect_tables(conn, database_metadata):
""") """)
] ]
for t, table_info in tables.items(): for t in tables.keys():
for hidden_table in hidden_tables: for hidden_table in hidden_tables:
if t == hidden_table or t.startswith(hidden_table): if t == hidden_table or t.startswith(hidden_table):
table_info["hidden"] = True tables[t]["hidden"] = True
continue continue
return tables return tables

View file

@ -21,7 +21,7 @@ class JumpSQL:
search_text: str | None = None, search_text: str | None = None,
display_name: str | None = None, display_name: str | None = None,
item_type: str = "menu", item_type: str = "menu",
) -> JumpSQL: ) -> "JumpSQL":
if search_text is None: if search_text is None:
search_text = " ".join( search_text = " ".join(
text for text in (label, display_name, description) if text is not None text for text in (label, display_name, description) if text is not None

View file

@ -1,11 +1,7 @@
import contextvars
from abc import ABC, abstractmethod from abc import ABC, abstractmethod
from dataclasses import dataclass from dataclasses import dataclass
from typing import Any, NamedTuple from typing import Any, NamedTuple
import contextvars
_SQLITE_IDENTIFIER_CASE = str.maketrans(
"ABCDEFGHIJKLMNOPQRSTUVWXYZ", "abcdefghijklmnopqrstuvwxyz"
)
# Context variable to track when permission checks should be skipped # Context variable to track when permission checks should be skipped
_skip_permission_checks = contextvars.ContextVar( _skip_permission_checks = contextvars.ContextVar(
@ -53,15 +49,6 @@ class Resource(ABC):
# Class-level metadata (subclasses must define these) # Class-level metadata (subclasses must define these)
name: str = None # e.g., "table", "database", "model" name: str = None # e.g., "table", "database", "model"
parent_class: type["Resource"] | None = None # e.g., DatabaseResource for tables parent_class: type["Resource"] | None = None # e.g., DatabaseResource for tables
case_insensitive_child: bool = False
@classmethod
def normalize_child(cls, child: str | None) -> str | None:
"""Return a comparison key without changing the resource's display name."""
if cls.case_insensitive_child and child is not None:
# Match SQLite NOCASE: fold ASCII only, not Unicode lower/casefold.
return child.translate(_SQLITE_IDENTIFIER_CASE)
return child
# Instance-level optional extra attributes # Instance-level optional extra attributes
reasons: list[str] | None = None reasons: list[str] | None = None
@ -85,8 +72,8 @@ class Resource(ABC):
) )
def __repr__(self) -> str: def __repr__(self) -> str:
return ( return "{}(parent={!r}, child={!r})".format(
f"{self.__class__.__name__}(parent={self.parent!r}, child={self.child!r})" self.__class__.__name__, self.parent, self.child
) )
@property @property
@ -142,6 +129,7 @@ class Resource(ABC):
Must return two columns: parent, child Must return two columns: parent, child
""" """
pass
class AllowedResource(NamedTuple): class AllowedResource(NamedTuple):
@ -159,11 +147,6 @@ class Action:
resource_class: type[Resource] | None = None resource_class: type[Resource] | None = None
also_requires: str | None = None # Optional action name that must also be allowed also_requires: str | None = None # Optional action name that must also be allowed
def normalize_child(self, child: str | None) -> str | None:
if self.resource_class is None:
return child
return self.resource_class.normalize_child(child)
@property @property
def takes_parent(self) -> bool: def takes_parent(self) -> bool:
""" """

View file

@ -1,14 +1,20 @@
import importlib import importlib
import importlib.metadata as importlib_metadata
import importlib.resources as importlib_resources
import os import os
import sys
from pprint import pprint
import pluggy import pluggy
from pprint import pprint
import sys
from . import hookspecs from . import hookspecs
if sys.version_info >= (3, 9):
import importlib.resources as importlib_resources
else:
import importlib_resources
if sys.version_info >= (3, 10):
import importlib.metadata as importlib_metadata
else:
import importlib_metadata
DEFAULT_PLUGINS = ( DEFAULT_PLUGINS = (
"datasette.publish.heroku", "datasette.publish.heroku",
"datasette.publish.cloudrun", "datasette.publish.cloudrun",
@ -18,7 +24,6 @@ DEFAULT_PLUGINS = (
"datasette.actor_auth_cookie", "datasette.actor_auth_cookie",
"datasette.default_permissions", "datasette.default_permissions",
"datasette.default_permissions.tokens", "datasette.default_permissions.tokens",
"datasette.default_permissions.sqlite_statistics",
"datasette.default_actions", "datasette.default_actions",
"datasette.default_column_types", "datasette.default_column_types",
"datasette.default_magic_parameters", "datasette.default_magic_parameters",
@ -26,7 +31,6 @@ DEFAULT_PLUGINS = (
"datasette.default_debug_menu", "datasette.default_debug_menu",
"datasette.default_jump_items", "datasette.default_jump_items",
"datasette.default_database_actions", "datasette.default_database_actions",
"datasette.default_table_actions",
"datasette.default_query_actions", "datasette.default_query_actions",
"datasette.handle_exception", "datasette.handle_exception",
"datasette.forbidden", "datasette.forbidden",
@ -80,7 +84,7 @@ if DATASETTE_LOAD_PLUGINS is not None:
# Ensure name can be found in plugin_to_distinfo later: # Ensure name can be found in plugin_to_distinfo later:
pm._plugin_distinfo.append((mod, distribution)) pm._plugin_distinfo.append((mod, distribution))
except importlib_metadata.PackageNotFoundError: except importlib_metadata.PackageNotFoundError:
sys.stderr.write(f"Plugin {package_name} could not be found\n") sys.stderr.write("Plugin {} could not be found\n".format(package_name))
# Load default plugins # Load default plugins

View file

@ -1,17 +1,15 @@
from datasette import hookimpl
import click
import json import json
import os import os
import re import re
from subprocess import CalledProcessError, check_call, check_output from subprocess import CalledProcessError, check_call, check_output
import click
from datasette import hookimpl
from ..utils import temporary_docker_directory
from .common import ( from .common import (
add_common_publish_arguments_and_options, add_common_publish_arguments_and_options,
fail_if_publish_binary_not_installed, fail_if_publish_binary_not_installed,
) )
from ..utils import temporary_docker_directory
@hookimpl @hookimpl
@ -221,7 +219,7 @@ def publish_subcommand(publish):
check_call( check_call(
"gcloud builds submit --tag {}{}".format( "gcloud builds submit --tag {}{}".format(
image_id, f" --timeout {timeout}" if timeout else "" image_id, " --timeout {}".format(timeout) if timeout else ""
), ),
shell=True, shell=True,
) )
@ -233,7 +231,7 @@ def publish_subcommand(publish):
("--min-instances", min_instances), ("--min-instances", min_instances),
): ):
if value is not None: if value is not None:
extra_deploy_options.append(f"{option} {value}") extra_deploy_options.append("{} {}".format(option, value))
check_call( check_call(
"gcloud run deploy --allow-unauthenticated --platform=managed --image {} {}{}".format( "gcloud run deploy --allow-unauthenticated --platform=managed --image {} {}{}".format(
image_id, image_id,
@ -260,16 +258,24 @@ def _ensure_artifact_registry(artifact_project, artifact_region, artifact_reposi
) from exc ) from exc
describe_cmd = ( describe_cmd = (
f"gcloud artifacts repositories describe {artifact_repository} --project {artifact_project} " "gcloud artifacts repositories describe {repo} --project {project} "
f"--location {artifact_region} --quiet" "--location {location} --quiet"
).format(
repo=artifact_repository,
project=artifact_project,
location=artifact_region,
) )
try: try:
check_call(describe_cmd, shell=True) check_call(describe_cmd, shell=True)
return return
except CalledProcessError: except CalledProcessError:
create_cmd = ( create_cmd = (
f"gcloud artifacts repositories create {artifact_repository} --repository-format=docker " "gcloud artifacts repositories create {repo} --repository-format=docker "
f'--location {artifact_region} --project {artifact_project} --description "Datasette Cloud Run images" --quiet' '--location {location} --project {project} --description "Datasette Cloud Run images" --quiet'
).format(
repo=artifact_repository,
location=artifact_region,
project=artifact_project,
) )
try: try:
check_call(create_cmd, shell=True) check_call(create_cmd, shell=True)

View file

@ -1,11 +1,9 @@
from ..utils import StaticMount
import click
import os import os
import shutil import shutil
import sys import sys
import click
from ..utils import StaticMount
def add_common_publish_arguments_and_options(subcommand): def add_common_publish_arguments_and_options(subcommand):
for decorator in reversed( for decorator in reversed(
@ -78,7 +76,9 @@ def fail_if_publish_binary_not_installed(binary, publish_target, install_link):
"""Exit (with error message) if ``binary` isn't installed""" """Exit (with error message) if ``binary` isn't installed"""
if not shutil.which(binary): if not shutil.which(binary):
click.secho( click.secho(
f"Publishing to {publish_target} requires {binary} to be installed and configured", "Publishing to {publish_target} requires {binary} to be installed and configured".format(
publish_target=publish_target, binary=binary
),
bg="red", bg="red",
fg="white", fg="white",
bold=True, bold=True,

View file

@ -1,21 +1,19 @@
from contextlib import contextmanager
from datasette import hookimpl
import click
import json import json
import os import os
import pathlib import pathlib
import shlex import shlex
import shutil import shutil
import tempfile
from contextlib import contextmanager
from subprocess import call, check_output from subprocess import call, check_output
import tempfile
import click
from datasette import hookimpl
from datasette.utils import link_or_copy, link_or_copy_directory, parse_metadata
from .common import ( from .common import (
add_common_publish_arguments_and_options, add_common_publish_arguments_and_options,
fail_if_publish_binary_not_installed, fail_if_publish_binary_not_installed,
) )
from datasette.utils import link_or_copy, link_or_copy_directory, parse_metadata
@hookimpl @hookimpl
@ -236,7 +234,7 @@ def temporary_heroku_directory(
extras.extend(["--static", f"{mount_point}:{mount_point}"]) extras.extend(["--static", f"{mount_point}:{mount_point}"])
quoted_files = " ".join( quoted_files = " ".join(
[f"-i {shlex.quote(file_name)}" for file_name in file_names] ["-i {}".format(shlex.quote(file_name)) for file_name in file_names]
) )
procfile_cmd = "web: datasette serve --host 0.0.0.0 {quoted_files} --cors --port $PORT --inspect-file inspect-data.json {extras}".format( procfile_cmd = "web: datasette serve --host 0.0.0.0 {quoted_files} --cors --port $PORT --inspect-file inspect-data.json {extras}".format(
quoted_files=quoted_files, extras=" ".join(extras) quoted_files=quoted_files, extras=" ".join(extras)

View file

@ -1,13 +1,11 @@
import json import json
from datasette.extras import extra_names_from_request from datasette.extras import extra_names_from_request
from datasette.utils import ( from datasette.utils import (
CustomJSONEncoder,
error_body,
path_from_row_pks,
remove_infinites,
sqlite3,
value_as_boolean, value_as_boolean,
remove_infinites,
CustomJSONEncoder,
path_from_row_pks,
sqlite3,
) )
from datasette.utils.asgi import Response from datasette.utils.asgi import Response
@ -54,7 +52,8 @@ def json_renderer(request, args, data, error, truncated=None):
if error: if error:
shape = "objects" shape = "objects"
status_code = 400 status_code = 400
data.update(error_body(error, status_code)) data["error"] = error
data["ok"] = False
if truncated is not None: if truncated is not None:
data["truncated"] = truncated data["truncated"] = truncated
@ -88,8 +87,7 @@ def json_renderer(request, args, data, error, truncated=None):
object_rows[pk_string] = row object_rows[pk_string] = row
data = object_rows data = object_rows
if shape_error: if shape_error:
status_code = 400 data = {"ok": False, "error": shape_error}
data = error_body(shape_error, status_code)
elif shape == "array": elif shape == "array":
data = data["rows"] data = data["rows"]
@ -102,7 +100,12 @@ def json_renderer(request, args, data, error, truncated=None):
data["rows"] = [list(row.values()) for row in data["rows"]] data["rows"] = [list(row.values()) for row in data["rows"]]
else: else:
status_code = 400 status_code = 400
data = error_body(f"Invalid _shape: {shape}", status_code) data = {
"ok": False,
"error": f"Invalid _shape: {shape}",
"status": 400,
"title": None,
}
# Don't include "columns" in output # Don't include "columns" in output
# https://github.com/simonw/datasette/issues/2136 # https://github.com/simonw/datasette/issues/2136

View file

@ -25,7 +25,6 @@ class TableResource(Resource):
name = "table" name = "table"
parent_class = DatabaseResource parent_class = DatabaseResource
case_insensitive_child = True
def __init__(self, database: str, table: str): def __init__(self, database: str, table: str):
super().__init__(parent=database, child=table) super().__init__(parent=database, child=table)

File diff suppressed because it is too large Load diff

View file

@ -472,13 +472,11 @@ class ColumnChooser extends HTMLElement {
<span class="drag-item-check"> <span class="drag-item-check">
<input type="checkbox" ${this._checked.has(col) ? "checked" : ""}> <input type="checkbox" ${this._checked.has(col) ? "checked" : ""}>
</span> </span>
<span class="drag-item-label"></span> <span class="drag-item-label">${col}</span>
</label> </label>
<div class="drop-indicator"></div> <div class="drop-indicator"></div>
`; `;
li.querySelector(".drag-item-label").textContent = col;
li.querySelector("input").addEventListener("change", (e) => { li.querySelector("input").addEventListener("change", (e) => {
e.target.checked ? this._checked.add(col) : this._checked.delete(col); e.target.checked ? this._checked.add(col) : this._checked.delete(col);
this._updateCounts(); this._updateCounts();

View file

@ -25,6 +25,29 @@ const DOM_SELECTORS = {
facetResults: ".facet-results [data-column]", facetResults: ".facet-results [data-column]",
}; };
let editToolsPromise = null;
let autocompletePromise = null;
function datasettePath(path) {
const baseUrl = window.datasetteBaseUrl || "/";
return baseUrl.replace(/\/?$/, "/") + String(path || "").replace(/^\/+/, "");
}
function parseInsertDialogRow(value) {
if (!value) {
return {};
}
try {
const row = JSON.parse(value);
if (!row || typeof row !== "object" || Array.isArray(row)) {
throw new Error("row must be an object");
}
return row;
} catch (error) {
throw new Error("Invalid insert dialog row: " + error.message);
}
}
/** /**
* Monolith class for interacting with Datasette JS API * Monolith class for interacting with Datasette JS API
* Imported with DEFER, runs after main document parsed * Imported with DEFER, runs after main document parsed
@ -39,6 +62,62 @@ const datasetteManager = {
// Does pluginMetadata need to be serializable, or can we let it be stateful / have functions? // Does pluginMetadata need to be serializable, or can we let it be stateful / have functions?
plugins: new Map(), plugins: new Map(),
path: datasettePath,
loadAutocomplete: () => {
if (!window.customElements) {
return Promise.resolve(null);
}
if (customElements.get("datasette-autocomplete")) {
return Promise.resolve(customElements.get("datasette-autocomplete"));
}
if (!autocompletePromise) {
const url =
window.datasetteAutocompleteUrl ||
datasettePath("/-/static/autocomplete.js");
autocompletePromise = import(url).then(() =>
customElements.get("datasette-autocomplete"),
);
}
return autocompletePromise;
},
loadEditTools: () => {
if (window.__DATASETTE_EDIT_TOOLS__) {
if (window.__DATASETTE_EDIT_TOOLS__.install) {
window.__DATASETTE_EDIT_TOOLS__.install(datasetteManager);
}
return Promise.resolve(window.__DATASETTE_EDIT_TOOLS__);
}
if (!editToolsPromise) {
const url =
window.datasetteEditToolsUrl ||
datasettePath("/-/static/edit-tools.js");
editToolsPromise = import(url).then(() => {
if (!window.__DATASETTE_EDIT_TOOLS__) {
throw new Error("edit-tools.js did not register its API");
}
if (window.__DATASETTE_EDIT_TOOLS__.install) {
window.__DATASETTE_EDIT_TOOLS__.install(datasetteManager);
}
return window.__DATASETTE_EDIT_TOOLS__;
});
}
return editToolsPromise;
},
insertDialog: async (database, table, row, message, options) => {
const editTools = await datasetteManager.loadEditTools();
return editTools.insertDialog(
datasetteManager,
database,
table,
row,
message,
options,
);
},
registerPlugin: (name, pluginMetadata) => { registerPlugin: (name, pluginMetadata) => {
if (datasetteManager.plugins.has(name)) { if (datasetteManager.plugins.has(name)) {
console.warn(`Warning -> plugin ${name} was redefined`); console.warn(`Warning -> plugin ${name} was redefined`);
@ -230,9 +309,7 @@ const datasetteManager = {
}; };
const initializeDatasette = () => { const initializeDatasette = () => {
// Hide the global behind __ prefix. Ideally they should be listening for the window.datasette = datasetteManager;
// DATASETTE_EVENTS.INIT event to avoid the habit of reading from the window.
window.__DATASETTE__ = datasetteManager; window.__DATASETTE__ = datasetteManager;
const initDatasetteEvent = new CustomEvent(DATASETTE_EVENTS.INIT, { const initDatasetteEvent = new CustomEvent(DATASETTE_EVENTS.INIT, {
@ -240,6 +317,43 @@ const initializeDatasette = () => {
}); });
document.dispatchEvent(initDatasetteEvent); document.dispatchEvent(initDatasetteEvent);
document.addEventListener("click", function (event) {
const button = event.target.closest("[data-insert-dialog]");
if (!button) {
return;
}
event.preventDefault();
let row;
try {
row = parseInsertDialogRow(button.dataset.row || "{}");
} catch (error) {
console.error(error);
return;
}
const reloadOnInsert = button.hasAttribute("data-insert-dialog-reload");
datasetteManager
.insertDialog(
button.dataset.database,
button.dataset.table,
row,
button.dataset.message || "",
{ flashMessage: reloadOnInsert },
)
.then((result) => {
if (
reloadOnInsert &&
result &&
result.ok &&
result.status === "inserted"
) {
window.location.reload();
}
})
.catch((error) => {
console.error(error);
});
});
}; };
/** /**

File diff suppressed because it is too large Load diff

View file

@ -0,0 +1,56 @@
/*
https://github.com/luyilin/json-format-highlight
From https://unpkg.com/json-format-highlight@1.0.1/dist/json-format-highlight.js
MIT Licensed
*/
(function (global, factory) {
typeof exports === "object" && typeof module !== "undefined"
? (module.exports = factory())
: typeof define === "function" && define.amd
? define(factory)
: (global.jsonFormatHighlight = factory());
})(this, function () {
"use strict";
var defaultColors = {
keyColor: "dimgray",
numberColor: "lightskyblue",
stringColor: "lightcoral",
trueColor: "lightseagreen",
falseColor: "#f66578",
nullColor: "cornflowerblue",
};
function index(json, colorOptions) {
if (colorOptions === void 0) colorOptions = {};
if (!json) {
return;
}
if (typeof json !== "string") {
json = JSON.stringify(json, null, 2);
}
var colors = Object.assign({}, defaultColors, colorOptions);
json = json.replace(/&/g, "&").replace(/</g, "<").replace(/>/g, ">");
return json.replace(
/("(\\u[a-zA-Z0-9]{4}|\\[^u]|[^\\"])*"(\s*:)?|\b(true|false|null)\b|-?\d+(?:\.\d*)?(?:[eE][+]?\d+)?)/g,
function (match) {
var color = colors.numberColor;
if (/^"/.test(match)) {
color = /:$/.test(match) ? colors.keyColor : colors.stringColor;
} else {
color = /true/.test(match)
? colors.trueColor
: /false/.test(match)
? colors.falseColor
: /null/.test(match)
? colors.nullColor
: color;
}
return '<span style="color: ' + color + '">' + match + "</span>";
},
);
}
return index;
});

View file

@ -633,151 +633,32 @@ const initDatasetteTable = function (manager) {
}); });
}; };
function filterRowSelector(manager) { /* Add x buttons to the filter rows */
return manager.selectors.filterRows || manager.selectors.filterRow; function addButtonsToFilterRows(manager) {
} var x = "✖";
var rows = Array.from(
function filterRowsWithControls(manager) { document.querySelectorAll(manager.selectors.filterRow),
return Array.from(
document.querySelectorAll(filterRowSelector(manager)),
).filter((el) => el.querySelector(".filter-op")); ).filter((el) => el.querySelector(".filter-op"));
} rows.forEach((row) => {
var a = document.createElement("a");
function filterRowNumberFromName(name) { a.setAttribute("href", "#");
var match = name && name.match(/^_filter_column_(\d+)$/); a.setAttribute("aria-label", "Remove this filter");
return match ? parseInt(match[1], 10) : 0; a.style.textDecoration = "none";
} a.innerText = x;
a.addEventListener("click", (ev) => {
function nextFilterRowNumber(manager) { ev.preventDefault();
return filterRowsWithControls(manager).reduce((max, row) => { let row = ev.target.closest("div");
var column = row.querySelector("select");
return Math.max(max, filterRowNumberFromName(column && column.name));
}, 0) + 1;
}
function setFilterRowNumber(row, number) {
row.querySelector("select").name = `_filter_column_${number}`;
row.querySelector(".filter-op select").name = `_filter_op_${number}`;
row.querySelector("input.filter-value").name = `_filter_value_${number}`;
}
function resetFilterRow(row) {
row.querySelector("select").value = ""; row.querySelector("select").value = "";
row.querySelector(".filter-op select").value = "exact"; row.querySelector(".filter-op select").value = "exact";
row.querySelector("input.filter-value").value = ""; row.querySelector("input.filter-value").value = "";
} ev.target.closest("a").style.display = "none";
});
function updateFilterRowButtons(manager) { row.appendChild(a);
var rows = filterRowsWithControls(manager);
rows.forEach((row, index) => {
var removeButton = row.querySelector(".filter-row-remove");
var addButton = row.querySelector(".filter-row-add");
var column = row.querySelector("select"); var column = row.querySelector("select");
if (removeButton) { if (!column.value) {
removeButton.hidden = index === 0; a.style.display = "none";
}
if (addButton) {
addButton.hidden = index !== rows.length - 1 || !column.value;
}
var visibleButtonCount = [removeButton, addButton].filter(function (button) {
return button && !button.hidden;
}).length;
row.classList.toggle(
"filter-controls-row-has-buttons",
visibleButtonCount > 0,
);
row.classList.toggle(
"filter-controls-row-one-button",
visibleButtonCount === 1,
);
row.classList.toggle(
"filter-controls-row-two-buttons",
visibleButtonCount === 2,
);
});
}
function cloneFilterRow(row) {
var clone = row.cloneNode(true);
clone.querySelector("select").name = "_filter_column";
clone.querySelector(".filter-op select").name = "_filter_op";
clone.querySelector("input.filter-value").name = "_filter_value";
resetFilterRow(clone);
clone.querySelectorAll(".filter-row-icon").forEach((button) => button.remove());
return clone;
}
var FILTER_REMOVE_ICON_SVG = `<svg class="filter-row-remove-icon" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" width="14" height="14" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2.1" stroke-linecap="round" stroke-linejoin="round">
<path d="M3 6h18"></path>
<path d="M8 6V4a2 2 0 0 1 2-2h4a2 2 0 0 1 2 2v2"></path>
<path d="M19 6l-1 14a2 2 0 0 1-2 2H8a2 2 0 0 1-2-2L5 6"></path>
<path d="M10 11v6"></path>
<path d="M14 11v6"></path>
</svg>`;
var FILTER_ADD_ICON_SVG = `<svg class="filter-row-add-icon" aria-hidden="true" xmlns="http://www.w3.org/2000/svg" width="16" height="16" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="3" stroke-linecap="round" stroke-linejoin="round">
<path d="M5 12h14"></path>
<path d="M12 5v14"></path>
</svg>`;
function addFilterRowButtons(row, manager) {
var removeButton = document.createElement("button");
removeButton.type = "button";
removeButton.className = "filter-row-icon filter-row-remove";
removeButton.setAttribute("aria-label", "Remove this filter");
removeButton.title = "Remove this filter";
removeButton.tabIndex = 0;
removeButton.innerHTML = FILTER_REMOVE_ICON_SVG;
removeButton.addEventListener("click", (ev) => {
var row = ev.currentTarget.closest(filterRowSelector(manager));
var rows = filterRowsWithControls(manager);
var rowIndex = rows.indexOf(row);
var focusRow = rows[rowIndex + 1] || rows[rowIndex - 1] || null;
row.remove();
updateFilterRowButtons(manager);
if (focusRow) {
var focusTarget =
focusRow.querySelector(".filter-row-add:not([hidden])") ||
focusRow.querySelector("select");
if (focusTarget) {
focusTarget.focus();
}
} }
}); });
row.appendChild(removeButton);
var addButton = document.createElement("button");
addButton.type = "button";
addButton.className = "filter-row-icon filter-row-add";
addButton.setAttribute("aria-label", "Add another filter");
addButton.title = "Add another filter";
addButton.tabIndex = 0;
addButton.innerHTML = FILTER_ADD_ICON_SVG;
addButton.addEventListener("click", (ev) => {
var row = ev.currentTarget.closest(filterRowSelector(manager));
if (row.querySelector("select").name === "_filter_column") {
setFilterRowNumber(row, nextFilterRowNumber(manager));
}
var clone = cloneFilterRow(row);
addFilterRowButtons(clone, manager);
row.parentNode.insertBefore(clone, row.nextSibling);
updateFilterRowButtons(manager);
clone.querySelector("select").focus();
});
row.appendChild(addButton);
row.querySelector("select").addEventListener("change", () => {
updateFilterRowButtons(manager);
});
}
/* Add buttons to the filter rows */
function addButtonsToFilterRows(manager) {
var rows = filterRowsWithControls(manager);
rows.forEach((row) => {
addFilterRowButtons(row, manager);
});
updateFilterRowButtons(manager);
} }
/* Set up datalist autocomplete for filter values */ /* Set up datalist autocomplete for filter values */
@ -806,11 +687,11 @@ function initAutocompleteForFilterValues(manager) {
}); });
} }
createDataLists(); createDataLists();
// When any filter column select changes, update the datalist // When any select with name=_filter_column changes, update the datalist
document.body.addEventListener("change", function (event) { document.body.addEventListener("change", function (event) {
if (event.target.name && event.target.name.startsWith("_filter_column")) { if (event.target.name === "_filter_column") {
event.target event.target
.closest(filterRowSelector(manager)) .closest(manager.selectors.filterRow)
.querySelector(".filter-value") .querySelector(".filter-value")
.setAttribute("list", "datalist-" + event.target.value); .setAttribute("list", "datalist-" + event.target.value);
} }
@ -860,45 +741,10 @@ function openColumnChooser() {
}); });
} }
function initCountAll() {
var button = document.querySelector(".count-all");
if (!button) {
return;
}
button.addEventListener("click", async function () {
var count = document.querySelector(".table-count");
var error = document.querySelector(".count-error");
button.disabled = true;
button.textContent = "Counting…";
error.textContent = "";
try {
var response = await fetch(button.dataset.countUrl + location.search, {
method: "POST",
headers: {
Accept: "application/json",
},
});
var data = await response.json();
if (!response.ok || !data.ok) {
throw new Error((data.errors || ["Count failed"]).join(" "));
}
count.textContent =
data.count.toLocaleString("en-US") +
(data.count === 1 ? " row" : " rows");
button.remove();
} catch (ex) {
error.textContent = ex.message || "Count failed";
button.disabled = false;
button.textContent = "count all";
}
});
}
// Ensures Table UI is initialized only after the Manager is ready. // Ensures Table UI is initialized only after the Manager is ready.
document.addEventListener("datasette_init", function (evt) { document.addEventListener("datasette_init", function (evt) {
const { detail: manager } = evt; const { detail: manager } = evt;
initCountAll();
initializeColumnActions(manager); initializeColumnActions(manager);
// Main table // Main table

View file

@ -1,9 +1,8 @@
from __future__ import annotations from __future__ import annotations
import json
from collections.abc import Iterable
from dataclasses import dataclass from dataclasses import dataclass
from typing import Any import json
from typing import Any, Iterable
from .utils import tilde_encode, urlsafe_components from .utils import tilde_encode, urlsafe_components
@ -63,6 +62,7 @@ def stored_query_to_dict(query: StoredQuery) -> dict[str, Any]:
"description_html": query.description_html, "description_html": query.description_html,
"hide_sql": query.hide_sql, "hide_sql": query.hide_sql,
"fragment": query.fragment, "fragment": query.fragment,
"params": list(query.parameters),
"parameters": list(query.parameters), "parameters": list(query.parameters),
"is_write": query.is_write, "is_write": query.is_write,
"is_private": query.is_private, "is_private": query.is_private,
@ -84,6 +84,7 @@ def stored_query_page_to_dict(page: StoredQueryPage) -> dict[str, Any]:
return { return {
"queries": [stored_query_to_dict(query) for query in page.queries], "queries": [stored_query_to_dict(query) for query in page.queries],
"next": page.next, "next": page.next,
"has_more": page.has_more,
"limit": page.limit, "limit": page.limit,
} }
@ -387,7 +388,7 @@ async def count_queries(
OR q.sql LIKE :query_search OR q.sql LIKE :query_search
) )
""") """)
params["query_search"] = f"%{q}%" params["query_search"] = "%{}%".format(q)
if is_write is not None: if is_write is not None:
where_clauses.append("q.is_write = :query_is_write") where_clauses.append("q.is_write = :query_is_write")
params["query_is_write"] = int(bool(is_write)) params["query_is_write"] = int(bool(is_write))
@ -463,7 +464,7 @@ async def list_queries(
except ValueError: except ValueError:
components = [] components = []
if database is None and len(components) == 3: if database is None and len(components) == 3:
where_clauses.append(f""" where_clauses.append("""
( (
q.database_name > :cursor_database q.database_name > :cursor_database
OR ( OR (
@ -477,12 +478,12 @@ async def list_queries(
) )
) )
) )
""") """.format(sort_key_sql=sort_key_sql))
params["cursor_database"] = components[0] params["cursor_database"] = components[0]
params["cursor_sort_key"] = components[1] params["cursor_sort_key"] = components[1]
params["cursor_name"] = components[2] params["cursor_name"] = components[2]
elif database is not None and len(components) == 2: elif database is not None and len(components) == 2:
where_clauses.append(f""" where_clauses.append("""
( (
{sort_key_sql} > :cursor_sort_key {sort_key_sql} > :cursor_sort_key
OR ( OR (
@ -490,7 +491,7 @@ async def list_queries(
AND q.name > :cursor_name AND q.name > :cursor_name
) )
) )
""") """.format(sort_key_sql=sort_key_sql))
params["cursor_sort_key"] = components[0] params["cursor_sort_key"] = components[0]
params["cursor_name"] = components[1] params["cursor_name"] = components[1]
@ -503,7 +504,7 @@ async def list_queries(
OR q.sql LIKE :query_search OR q.sql LIKE :query_search
) )
""") """)
params["query_search"] = f"%{q}%" params["query_search"] = "%{}%".format(q)
if is_write is not None: if is_write is not None:
where_clauses.append("q.is_write = :query_is_write") where_clauses.append("q.is_write = :query_is_write")
params["query_is_write"] = int(bool(is_write)) params["query_is_write"] = int(bool(is_write))

View file

@ -1,40 +0,0 @@
"""
Index of the documented template contexts for Datasette's core HTML pages.
This module deliberately contains no documentation strings of its own -
the documentation lives next to the code it describes:
- Every page renders a Context dataclass defined in its view module
(DatabaseContext, QueryContext in views/database.py, TableContext in
views/table.py, RowContext in views/row.py). Fields added by view code
carry ``help`` metadata; fields declared with from_extra() take their
documentation from the description on the matching Extra class in
views/table_extras.py.
- The keys render_template() adds to every page are documented in
TEMPLATE_BASE_CONTEXT in datasette/app.py, next to the code that adds
them.
The contract tests in tests/test_template_context.py assert that the real
rendered context for each page exactly matches what is documented, and
docs/template_context_doc.py generates docs/template_context.rst from the
same classes.
"""
from datasette.app import TEMPLATE_BASE_CONTEXT
from datasette.views.database import DatabaseContext, QueryContext
from datasette.views.row import RowContext
from datasette.views.table import TableContext
PAGES = {
"database": DatabaseContext,
"query": QueryContext,
"table": TableContext,
"row": RowContext,
}
def documented_context_keys(page_name):
"Set of every documented key for the named page, including base context keys"
return set(TEMPLATE_BASE_CONTEXT) | {
f.name for f in PAGES[page_name].documented_fields()
}

View file

@ -1,5 +1,5 @@
<script src="{{ static('sql-formatter-2.3.3.min.js') }}" defer></script> <script src="{{ base_url }}-/static/sql-formatter-2.3.3.min.js" defer></script>
<script src="{{ static('cm-editor-6.0.1.bundle.js') }}"></script> <script src="{{ base_url }}-/static/cm-editor-6.0.1.bundle.js"></script>
<style> <style>
.cm-editor { .cm-editor {
resize: both; resize: both;

View file

@ -6,20 +6,8 @@
padding: 1.5em; padding: 1.5em;
margin-bottom: 2em; margin-bottom: 2em;
} }
.permission-form form {
max-width: 60rem;
}
.permission-form-grid {
display: grid;
gap: 1.5rem;
grid-template-columns: repeat(2, minmax(0, 1fr));
}
.permission-form-result {
margin-top: 1rem;
max-width: 60rem;
}
.form-section { .form-section {
margin-bottom: 1.25em; margin-bottom: 1em;
} }
.form-section label { .form-section label {
display: block; display: block;
@ -27,51 +15,22 @@
font-weight: bold; font-weight: bold;
} }
.form-section input[type="text"], .form-section input[type="text"],
.form-section input[type="number"],
.form-section select,
.permission-textarea {
background-color: #fff;
border: 1px solid #aaa;
border-radius: 4px;
box-sizing: border-box;
box-shadow: inset 0 1px 2px rgba(0, 0, 0, 0.08);
color: #222;
font-family: inherit;
font-size: 1rem;
line-height: 1.4;
max-width: none;
width: 100%;
}
.form-section input[type="text"] {
height: 3rem;
padding: 0.6rem 0.75rem;
}
.form-section input[type="number"] {
height: 3rem;
max-width: 7rem;
padding: 0.6rem 0.75rem;
}
.form-section select { .form-section select {
height: 3rem; width: 100%;
padding: 0.6rem 0.75rem; max-width: 500px;
} padding: 0.5em;
.permission-textarea { box-sizing: border-box;
font-family: monospace; border: 1px solid #ccc;
min-height: 12rem; border-radius: 3px;
padding: 0.75rem;
resize: vertical;
} }
.form-section input[type="text"]:focus, .form-section input[type="text"]:focus,
.form-section input[type="number"]:focus, .form-section select:focus {
.form-section select:focus, outline: 2px solid #0066cc;
.permission-textarea:focus {
border-color: #0066cc; border-color: #0066cc;
box-shadow: 0 0 0 3px rgba(0, 102, 204, 0.18);
outline: none;
} }
.form-section small { .form-section small {
display: block; display: block;
margin-top: 0.45em; margin-top: 0.3em;
color: #666; color: #666;
} }
.form-actions { .form-actions {
@ -183,9 +142,4 @@
text-align: center; text-align: center;
color: #666; color: #666;
} }
@media only screen and (max-width: 576px) {
.permission-form-grid {
grid-template-columns: minmax(0, 1fr);
}
}
</style> </style>

View file

@ -44,10 +44,10 @@
</style> </style>
<nav class="permissions-debug-tabs"> <nav class="permissions-debug-tabs">
<a href="{{ urls.path('-/check') }}{{ query_string }}" {% if current_tab == "check" %}class="active"{% endif %}>Explain</a> <a href="{{ urls.path('-/permissions') }}" {% if current_tab == "permissions" %}class="active"{% endif %}>Playground</a>
<a href="{{ urls.path('-/allowed') }}{{ query_string }}" {% if current_tab == "allowed" %}class="active"{% endif %}>Access map</a> <a href="{{ urls.path('-/check') }}{{ query_string }}" {% if current_tab == "check" %}class="active"{% endif %}>Check</a>
<a href="{{ urls.path('-/rules') }}{{ query_string }}" {% if current_tab == "rules" %}class="active"{% endif %}>Rule explorer</a> <a href="{{ urls.path('-/allowed') }}{{ query_string }}" {% if current_tab == "allowed" %}class="active"{% endif %}>Allowed</a>
<a href="{{ urls.path('-/permissions') }}" {% if current_tab == "permissions" %}class="active"{% endif %}>Activity</a> <a href="{{ urls.path('-/rules') }}{{ query_string }}" {% if current_tab == "rules" %}class="active"{% endif %}>Rules</a>
<a href="{{ urls.path('-/actions') }}" {% if current_tab == "actions" %}class="active"{% endif %}>Actions</a> <a href="{{ urls.path('-/actions') }}" {% if current_tab == "actions" %}class="active"{% endif %}>Actions</a>
<a href="{{ urls.path('-/allow-debug') }}" {% if current_tab == "allow_debug" %}class="active"{% endif %}>Allow debug</a> <a href="{{ urls.path('-/allow-debug') }}" {% if current_tab == "allow_debug" %}class="active"{% endif %}>Allow debug</a>
</nav> </nav>

View file

@ -3,11 +3,29 @@
{% block title %}Debug allow rules{% endblock %} {% block title %}Debug allow rules{% endblock %}
{% block extra_head %} {% block extra_head %}
{% include "_permission_ui_styles.html" %}
<style> <style>
textarea {
height: 10em;
width: 95%;
box-sizing: border-box;
padding: 0.5em;
border: 2px dotted black;
}
.two-col {
display: inline-block;
width: 48%;
}
.two-col label {
width: 48%;
}
p.message-warning { p.message-warning {
white-space: pre-wrap; white-space: pre-wrap;
} }
@media only screen and (max-width: 576px) {
.two-col {
width: 100%;
}
}
</style> </style>
{% endblock %} {% endblock %}
@ -20,28 +38,24 @@ p.message-warning {
<p>Use this tool to try out different actor and allow combinations. See <a href="https://docs.datasette.io/en/stable/authentication.html#defining-permissions-with-allow-blocks">Defining permissions with "allow" blocks</a> for documentation.</p> <p>Use this tool to try out different actor and allow combinations. See <a href="https://docs.datasette.io/en/stable/authentication.html#defining-permissions-with-allow-blocks">Defining permissions with "allow" blocks</a> for documentation.</p>
<div class="permission-form"> <form class="core" action="{{ urls.path('-/allow-debug') }}" method="get" style="margin-bottom: 1em">
<form class="core" action="{{ urls.path('-/allow-debug') }}" method="get"> <div class="two-col">
<div class="permission-form-grid"> <p><label>Allow block</label></p>
<div class="form-section"> <textarea name="allow">{{ allow_input }}</textarea>
<label for="allow-block">Allow block</label>
<textarea class="permission-textarea" id="allow-block" name="allow">{{ allow_input }}</textarea>
</div> </div>
<div class="form-section"> <div class="two-col">
<label for="allow-actor">Actor</label> <p><label>Actor</label></p>
<textarea class="permission-textarea" id="allow-actor" name="actor">{{ actor_input }}</textarea> <textarea name="actor">{{ actor_input }}</textarea>
</div> </div>
</div> <div style="margin-top: 1em;">
<div class="form-actions"> <input type="submit" value="Apply allow block to actor">
<button type="submit" class="submit-btn">Apply allow block to actor</button>
</div> </div>
</form> </form>
{% if error %}<p class="message-warning permission-form-result">{{ error }}</p>{% endif %} {% if error %}<p class="message-warning">{{ error }}</p>{% endif %}
{% if result == "True" %}<p class="message-info permission-form-result">Result: allow</p>{% endif %} {% if result == "True" %}<p class="message-info">Result: allow</p>{% endif %}
{% if result == "False" %}<p class="message-error permission-form-result">Result: deny</p>{% endif %} {% if result == "False" %}<p class="message-error">Result: deny</p>{% endif %}
</div>
{% endblock %} {% endblock %}

View file

@ -3,6 +3,7 @@
{% block title %}API Explorer{% endblock %} {% block title %}API Explorer{% endblock %}
{% block extra_head %} {% block extra_head %}
<script src="{{ base_url }}-/static/json-format-highlight-1.0.1.js"></script>
{% endblock %} {% endblock %}
{% block content %} {% block content %}
@ -125,7 +126,7 @@ getForm.addEventListener("submit", (ev) => {
document.getElementById('response-status').textContent = response.status; document.getElementById('response-status').textContent = response.status;
return response.json(); return response.json();
}).then((data) => { }).then((data) => {
output.querySelector('pre').textContent = JSON.stringify(data, null, 2); output.querySelector('pre').innerHTML = jsonFormatHighlight(data);
errorList.style.display = 'none'; errorList.style.display = 'none';
}).catch((error) => { }).catch((error) => {
alert(error); alert(error);
@ -173,7 +174,7 @@ postForm.addEventListener("submit", (ev) => {
} else { } else {
errorList.style.display = 'none'; errorList.style.display = 'none';
} }
output.querySelector('pre').textContent = JSON.stringify(data, null, 2); output.querySelector('pre').innerHTML = jsonFormatHighlight(data);
output.style.display = 'block'; output.style.display = 'block';
}).catch(err => { }).catch(err => {
alert("Error: " + err); alert("Error: " + err);

View file

@ -2,13 +2,18 @@
<html lang="en"> <html lang="en">
<head> <head>
<title>{% block title %}{% endblock %}</title> <title>{% block title %}{% endblock %}</title>
<link rel="stylesheet" href="{{ static('app.css') }}"> <link rel="stylesheet" href="{{ urls.static('app.css') }}?{{ app_css_hash }}">
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"> <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
{% for url in extra_css_urls %} {% for url in extra_css_urls %}
<link rel="stylesheet" href="{{ url.url }}"{% if url.get("sri") %} integrity="{{ url.sri }}" crossorigin="anonymous"{% endif %}> <link rel="stylesheet" href="{{ url.url }}"{% if url.get("sri") %} integrity="{{ url.sri }}" crossorigin="anonymous"{% endif %}>
{% endfor %} {% endfor %}
<script>window.datasetteVersion = '{{ datasette_version }}';</script> <script>
<script src="{{ static('datasette-manager.js') }}" defer></script> window.datasetteVersion = {{ datasette_version|tojson }};
window.datasetteBaseUrl = {{ base_url|tojson }};
window.datasetteEditToolsUrl = {{ (urls.static('edit-tools.js') ~ '?hash=' ~ edit_tools_js_hash)|tojson }};
window.datasetteAutocompleteUrl = {{ urls.static('autocomplete.js')|tojson }};
</script>
<script src="{{ urls.static('datasette-manager.js') }}" defer></script>
{% for url in extra_js_urls %} {% for url in extra_js_urls %}
<script {% if url.module %}type="module" {% endif %}src="{{ url.url }}"{% if url.get("sri") %} integrity="{{ url.sri }}" crossorigin="anonymous"{% endif %}></script> <script {% if url.module %}type="module" {% endif %}src="{{ url.url }}"{% if url.get("sri") %} integrity="{{ url.sri }}" crossorigin="anonymous"{% endif %}></script>
{% endfor %} {% endfor %}
@ -70,7 +75,7 @@
{% endfor %} {% endfor %}
{% if select_templates %}<!-- Templates considered: {{ select_templates|join(", ") }} -->{% endif %} {% if select_templates %}<!-- Templates considered: {{ select_templates|join(", ") }} -->{% endif %}
<script src="{{ static('navigation-search.js') }}" defer></script> <script src="{{ urls.static('navigation-search.js') }}" defer></script>
<navigation-search url="{{ urls.path("/-/jump") }}"></navigation-search> <navigation-search url="{{ urls.path("/-/jump") }}"></navigation-search>
</body> </body>
</html> </html>

View file

@ -6,10 +6,6 @@
{{- super() -}} {{- super() -}}
{% include "_codemirror.html" %} {% include "_codemirror.html" %}
{% include "_sql_parameter_styles.html" %} {% include "_sql_parameter_styles.html" %}
{% if database_page_data.createTable %}
<script>window._datasetteDatabaseData = {{ database_page_data|tojson }};</script>
<script src="{{ static('edit-tools.js') }}" defer></script>
{% endif %}
{% endblock %} {% endblock %}
{% block body_class %}db db-{{ database|to_css_class }}{% endblock %} {% block body_class %}db db-{{ database|to_css_class }}{% endblock %}
@ -76,7 +72,7 @@
<div class="db-table"> <div class="db-table">
<h3><a href="{{ urls.table(database, table.name) }}">{{ table.name }}</a>{% if table.private %} 🔒{% endif %}{% if table.hidden %}<em> (hidden)</em>{% endif %}</h3> <h3><a href="{{ urls.table(database, table.name) }}">{{ table.name }}</a>{% if table.private %} 🔒{% endif %}{% if table.hidden %}<em> (hidden)</em>{% endif %}</h3>
<p><em>{% for column in table.columns %}{{ column }}{% if not loop.last %}, {% endif %}{% endfor %}</em></p> <p><em>{% for column in table.columns %}{{ column }}{% if not loop.last %}, {% endif %}{% endfor %}</em></p>
<p>{% if table.count is none %}Many rows{% elif table.count_truncated %}&gt;{{ "{:,}".format(table.count - 1) }} rows{% else %}{{ "{:,}".format(table.count) }} row{% if table.count == 1 %}{% else %}s{% endif %}{% endif %}</p> <p>{% if table.count is none %}Many rows{% elif table.count == count_limit + 1 %}&gt;{{ "{:,}".format(count_limit) }} rows{% else %}{{ "{:,}".format(table.count) }} row{% if table.count == 1 %}{% else %}s{% endif %}{% endif %}</p>
</div> </div>
{% endif %} {% endif %}
{% endfor %} {% endfor %}

View file

@ -9,7 +9,7 @@
{% include "_permissions_debug_tabs.html" %} {% include "_permissions_debug_tabs.html" %}
<p style="margin-bottom: 2em;"> <p style="margin-bottom: 2em;">
This Datasette instance has registered {{ data.actions|length }} action{{ data.actions|length != 1 and "s" or "" }}. This Datasette instance has registered {{ data|length }} action{{ data|length != 1 and "s" or "" }}.
Actions are used by the permission system to control access to different features. Actions are used by the permission system to control access to different features.
</p> </p>
@ -26,7 +26,7 @@
</tr> </tr>
</thead> </thead>
<tbody> <tbody>
{% for action in data.actions %} {% for action in data %}
<tr> <tr>
<td><strong>{{ action.name }}</strong></td> <td><strong>{{ action.name }}</strong></td>
<td>{% if action.abbr %}<code>{{ action.abbr }}</code>{% endif %}</td> <td>{% if action.abbr %}<code>{{ action.abbr }}</code>{% endif %}</td>

View file

@ -3,6 +3,7 @@
{% block title %}Allowed Resources{% endblock %} {% block title %}Allowed Resources{% endblock %}
{% block extra_head %} {% block extra_head %}
<script src="{{ base_url }}-/static/json-format-highlight-1.0.1.js"></script>
{% include "_permission_ui_styles.html" %} {% include "_permission_ui_styles.html" %}
{% include "_debug_common_functions.html" %} {% include "_debug_common_functions.html" %}
{% endblock %} {% endblock %}
@ -48,7 +49,7 @@
<div class="form-section"> <div class="form-section">
<label for="page_size">Page size:</label> <label for="page_size">Page size:</label>
<input type="number" id="page_size" name="_size" value="50" min="1" max="200"> <input type="number" id="page_size" name="page_size" value="50" min="1" max="200" style="max-width: 100px;">
<small>Number of results per page (max 200)</small> <small>Number of results per page (max 200)</small>
</div> </div>
@ -87,7 +88,7 @@ const hasDebugPermission = {{ 'true' if has_debug_permission else 'false' }};
(function() { (function() {
const params = populateFormFromURL(); const params = populateFormFromURL();
const action = params.get('action'); const action = params.get('action');
const page = params.get('_page'); const page = params.get('page');
if (action) { if (action) {
fetchResults(page ? parseInt(page) : 1); fetchResults(page ? parseInt(page) : 1);
} }
@ -101,14 +102,14 @@ async function fetchResults(page = 1) {
const params = new URLSearchParams(); const params = new URLSearchParams();
for (const [key, value] of formData.entries()) { for (const [key, value] of formData.entries()) {
if (value && key !== '_size' && key !== '_page') { if (value && key !== 'page_size') {
params.append(key, value); params.append(key, value);
} }
} }
const pageSize = document.getElementById('page_size').value || '50'; const pageSize = document.getElementById('page_size').value || '50';
params.append('_page', page.toString()); params.append('page', page.toString());
params.append('_size', pageSize); params.append('page_size', pageSize);
try { try {
const response = await fetch('{{ urls.path("-/allowed.json") }}?' + params.toString(), { const response = await fetch('{{ urls.path("-/allowed.json") }}?' + params.toString(), {
@ -197,7 +198,7 @@ function displayResults(data) {
} }
// Update raw JSON // Update raw JSON
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2); document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
} }
function displayError(data) { function displayError(data) {
@ -207,7 +208,7 @@ function displayError(data) {
resultsContent.innerHTML = `<div class="error-message">Error: ${escapeHtml(data.error || 'Unknown error')}</div>`; resultsContent.innerHTML = `<div class="error-message">Error: ${escapeHtml(data.error || 'Unknown error')}</div>`;
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2); document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
} }
// Disable child input if parent is empty // Disable child input if parent is empty

View file

@ -4,7 +4,7 @@
{% block extra_head %} {% block extra_head %}
{{ super() }} {{ super() }}
<script src="{{ static('autocomplete.js') }}" defer></script> <script src="{{ urls.static('autocomplete.js') }}" defer></script>
{% endblock %} {% endblock %}
{% block content %} {% block content %}

View file

@ -1,8 +1,9 @@
{% extends "base.html" %} {% extends "base.html" %}
{% block title %}Explain a permission decision{% endblock %} {% block title %}Permission Check{% endblock %}
{% block extra_head %} {% block extra_head %}
<script src="{{ base_url }}-/static/json-format-highlight-1.0.1.js"></script>
{% include "_permission_ui_styles.html" %} {% include "_permission_ui_styles.html" %}
{% include "_debug_common_functions.html" %} {% include "_debug_common_functions.html" %}
<style> <style>
@ -12,35 +13,29 @@
border-radius: 5px; border-radius: 5px;
} }
#output.allowed { #output.allowed {
background-color: #f3fbf4; background-color: #e8f5e9;
border: 2px solid #4caf50; border: 2px solid #4caf50;
} }
#output.denied { #output.denied {
background-color: #fff7f7; background-color: #ffebee;
border: 2px solid #f44336; border: 2px solid #f44336;
} }
#output h2 { #output h2 {
margin-top: 0; margin-top: 0;
} }
#output h3 { #output .result-badge {
margin-bottom: 0.5em;
}
#output .result-badge,
.effect-badge,
.rule-status {
display: inline-block; display: inline-block;
padding: 0.2em 0.5em; padding: 0.3em 0.8em;
border-radius: 3px; border-radius: 3px;
font-weight: bold; font-weight: bold;
font-size: 1.1em;
} }
#output .allowed-badge, #output .allowed-badge {
.effect-allow { background-color: #4caf50;
background-color: #2e7d32;
color: white; color: white;
} }
#output .denied-badge, #output .denied-badge {
.effect-deny { background-color: #f44336;
background-color: #c62828;
color: white; color: white;
} }
.details-section { .details-section {
@ -53,130 +48,70 @@
.details-section dd { .details-section dd {
margin-left: 1em; margin-left: 1em;
} }
.explanation-section {
background: rgba(255, 255, 255, 0.75);
border: 1px solid #ddd;
border-radius: 4px;
margin-top: 1em;
padding: 0 1em 1em;
}
.rules-table {
border-collapse: collapse;
width: 100%;
}
.rules-table th,
.rules-table td {
border-bottom: 1px solid #ddd;
padding: 0.5em;
text-align: left;
vertical-align: top;
}
.rule-status {
background: #e8f5e9;
color: #1b5e20;
}
.rule-ignored {
background: #eee;
color: #555;
font-weight: normal;
}
.requirement-allowed {
color: #1b5e20;
}
.requirement-denied {
color: #b71c1c;
}
@media only screen and (max-width: 576px) {
.rules-table,
.rules-table tbody,
.rules-table tr,
.rules-table td {
display: block;
}
.rules-table thead {
display: none;
}
.rules-table td::before {
content: attr(data-label) ": ";
font-weight: bold;
}
}
</style> </style>
{% endblock %} {% endblock %}
{% block content %} {% block content %}
<h1>Explain a permission decision</h1> <h1>Permission check</h1>
{% set current_tab = "check" %} {% set current_tab = "check" %}
{% include "_permissions_debug_tabs.html" %} {% include "_permissions_debug_tabs.html" %}
<p>Test an actor, action and resource. The result explains which rules matched, which specificity level won, and whether actor restrictions or required actions changed the verdict.</p> <p>Use this tool to test permission checks for the current actor. It queries the <code>/-/check.json</code> API endpoint.</p>
{% if request.actor %}
<p>Current actor: <strong>{{ request.actor.get("id", "anonymous") }}</strong></p>
{% else %}
<p>Current actor: <strong>anonymous (not logged in)</strong></p>
{% endif %}
<div class="permission-form"> <div class="permission-form">
<form id="check-form" method="get" action="{{ urls.path('-/check') }}"> <form id="check-form" method="get" action="{{ urls.path("-/check") }}">
<div class="form-section"> <div class="form-section">
<label for="actor">Actor JSON:</label> <label for="action">Action (permission name):</label>
<textarea class="permission-textarea" id="actor" name="actor">{{ actor_json }}</textarea>
<small>Use <code>null</code> for an anonymous actor. This actor is simulated; it does not change who you are signed in as.</small>
</div>
<div class="form-section">
<label for="action">Action:</label>
<select id="action" name="action" required> <select id="action" name="action" required>
<option value="">Select an action...</option> <option value="">Select an action...</option>
{% for action in actions %} {% for action_name in sorted_actions %}
<option value="{{ action.name }}">{{ action.name }}{% if action.description %} — {{ action.description }}{% endif %}</option> <option value="{{ action_name }}">{{ action_name }}</option>
{% endfor %} {% endfor %}
</select> </select>
<small id="action-help">The operation to evaluate</small> <small>The permission action to check</small>
</div> </div>
<div class="form-section" id="parent-section"> <div class="form-section">
<label for="parent">Parent resource:</label> <label for="parent">Parent resource (optional):</label>
<input type="text" id="parent" name="parent" placeholder="e.g., database name"> <input type="text" id="parent" name="parent" placeholder="e.g., database name">
<small>The database or other parent resource</small> <small>For database-level permissions, specify the database name</small>
</div> </div>
<div class="form-section" id="child-section"> <div class="form-section">
<label for="child">Child resource:</label> <label for="child">Child resource (optional):</label>
<input type="text" id="child" name="child" placeholder="e.g., table or query name"> <input type="text" id="child" name="child" placeholder="e.g., table name">
<small>The table, query or other child resource</small> <small>For table-level permissions, specify the table name (requires parent)</small>
</div> </div>
<div class="form-actions"> <div class="form-actions">
<button type="submit" class="submit-btn" id="submit-btn">Explain decision</button> <button type="submit" class="submit-btn" id="submit-btn">Check Permission</button>
</div> </div>
</form> </form>
</div> </div>
<div id="output" style="display: none;"> <div id="output" style="display: none;">
<h2>Result: <span class="result-badge" id="result-badge"></span></h2> <h2>Result: <span class="result-badge" id="result-badge"></span></h2>
<p id="result-summary"></p>
<dl class="details-section"> <dl class="details-section">
<dt>Actor:</dt>
<dd><code id="result-actor"></code></dd>
<dt>Action:</dt> <dt>Action:</dt>
<dd><code id="result-action"></code></dd> <dd id="result-action"></dd>
<dt>Resource:</dt>
<dd><code id="result-resource"></code></dd> <dt>Resource Path:</dt>
<dd id="result-resource"></dd>
<dt>Actor ID:</dt>
<dd id="result-actor"></dd>
<div id="additional-details"></div>
</dl> </dl>
<section class="explanation-section">
<h3>Matching rules</h3>
<div id="matching-rules"></div>
</section>
<section class="explanation-section" id="restrictions-section">
<h3>Actor restrictions</h3>
<div id="restriction-results"></div>
</section>
<section class="explanation-section" id="requirements-section">
<h3>Required actions</h3>
<div id="requirement-results"></div>
</section>
<details style="margin-top: 1em;"> <details style="margin-top: 1em;">
<summary style="cursor: pointer; font-weight: bold;">Raw JSON response</summary> <summary style="cursor: pointer; font-weight: bold;">Raw JSON response</summary>
<pre id="raw-json" style="margin-top: 1em; padding: 1em; background-color: #f5f5f5; border: 1px solid #ddd; border-radius: 3px; overflow-x: auto;"></pre> <pre id="raw-json" style="margin-top: 1em; padding: 1em; background-color: #f5f5f5; border: 1px solid #ddd; border-radius: 3px; overflow-x: auto;"></pre>
@ -184,134 +119,152 @@
</div> </div>
<script> <script>
const actions = Object.fromEntries({{ actions|tojson }}.map(action => [action.name, action]));
const form = document.getElementById('check-form'); const form = document.getElementById('check-form');
const output = document.getElementById('output'); const output = document.getElementById('output');
const submitBtn = document.getElementById('submit-btn'); const submitBtn = document.getElementById('submit-btn');
const actionSelect = document.getElementById('action');
function updateResourceFields() {
const action = actions[actionSelect.value];
document.getElementById('parent-section').style.display = action && action.takes_parent ? 'block' : 'none';
document.getElementById('child-section').style.display = action && action.takes_child ? 'block' : 'none';
let help = action && action.description ? action.description : 'The operation to evaluate';
if (action && action.also_requires) {
help += `; also requires ${action.also_requires}`;
}
document.getElementById('action-help').textContent = help;
}
async function performCheck() { async function performCheck() {
submitBtn.disabled = true; submitBtn.disabled = true;
submitBtn.textContent = 'Explaining...'; submitBtn.textContent = 'Checking...';
const params = new URLSearchParams(new FormData(form));
const formData = new FormData(form);
const params = new URLSearchParams();
for (const [key, value] of formData.entries()) {
if (value) {
params.append(key, value);
}
}
try { try {
const response = await fetch('{{ urls.path("-/check.json") }}?' + params.toString(), { const response = await fetch('{{ urls.path("-/check.json") }}?' + params.toString(), {
headers: {'Accept': 'application/json'} method: 'GET',
headers: {
'Accept': 'application/json',
}
}); });
const data = await response.json(); const data = await response.json();
if (response.ok) { if (response.ok) {
displayResult(data); displayResult(data);
} else { } else {
displayError(data); displayError(data);
} }
} catch (error) { } catch (error) {
displayError({error: error.message}); alert('Error: ' + error.message);
} finally { } finally {
submitBtn.disabled = false; submitBtn.disabled = false;
submitBtn.textContent = 'Explain decision'; submitBtn.textContent = 'Check Permission';
} }
} }
// Populate form on initial load
(function() {
const params = populateFormFromURL();
const action = params.get('action');
if (action) {
performCheck();
}
})();
function displayResult(data) { function displayResult(data) {
output.style.display = 'block'; output.style.display = 'block';
// Set badge and styling
const resultBadge = document.getElementById('result-badge'); const resultBadge = document.getElementById('result-badge');
output.className = data.allowed ? 'allowed' : 'denied'; if (data.allowed) {
resultBadge.className = `result-badge ${data.allowed ? 'allowed-badge' : 'denied-badge'}`; output.className = 'allowed';
resultBadge.textContent = data.allowed ? 'ALLOWED ✓' : 'DENIED ✗'; resultBadge.className = 'result-badge allowed-badge';
document.getElementById('result-summary').textContent = data.explanation.summary; resultBadge.textContent = 'ALLOWED ✓';
document.getElementById('result-actor').textContent = data.actor === null ? 'anonymous' : JSON.stringify(data.actor); } else {
document.getElementById('result-action').textContent = data.action; output.className = 'denied';
document.getElementById('result-resource').textContent = data.resource.path; resultBadge.className = 'result-badge denied-badge';
displayRules(data.explanation); resultBadge.textContent = 'DENIED ✗';
displayRestrictions(data.explanation.restrictions);
displayRequirements(data.explanation.required_actions);
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2);
} }
function displayRules(explanation) { // Basic details
const container = document.getElementById('matching-rules'); document.getElementById('result-action').textContent = data.action || 'N/A';
if (!explanation.matched_rules.length) { document.getElementById('result-resource').textContent = data.resource?.path || '/';
container.innerHTML = '<p>No rules matched. Datasette denies access when there is no matching rule.</p>'; document.getElementById('result-actor').textContent = data.actor_id || 'anonymous';
return;
} // Additional details
let html = '<table class="rules-table"><thead><tr><th>Effect</th><th>Scope</th><th>Source</th><th>Reason</th><th>Role in decision</th></tr></thead><tbody>'; const additionalDetails = document.getElementById('additional-details');
for (const rule of explanation.matched_rules) { additionalDetails.innerHTML = '';
const status = rule.decisive
? '<span class="rule-status">Decisive</span>' if (data.reason !== undefined) {
: `<span class="rule-status rule-ignored">${escapeHtml(rule.ignored_because)}</span>`; const dt = document.createElement('dt');
html += '<tr>'; dt.textContent = 'Reason:';
html += `<td data-label="Effect"><span class="effect-badge effect-${rule.effect}">${rule.effect.toUpperCase()}</span></td>`; const dd = document.createElement('dd');
html += `<td data-label="Scope">${escapeHtml(rule.scope)}</td>`; dd.textContent = data.reason || 'N/A';
html += `<td data-label="Source"><code>${escapeHtml(rule.source || 'unknown')}</code></td>`; additionalDetails.appendChild(dt);
html += `<td data-label="Reason">${escapeHtml(rule.reason || 'No reason supplied')}</td>`; additionalDetails.appendChild(dd);
html += `<td data-label="Role in decision">${status}</td>`;
html += '</tr>';
}
container.innerHTML = html + '</tbody></table>';
} }
function displayRestrictions(restrictions) { if (data.source_plugin !== undefined) {
const section = document.getElementById('restrictions-section'); const dt = document.createElement('dt');
const container = document.getElementById('restriction-results'); dt.textContent = 'Source Plugin:';
section.style.display = restrictions.length ? 'block' : 'none'; const dd = document.createElement('dd');
container.innerHTML = restrictions.map(restriction => { dd.textContent = data.source_plugin || 'N/A';
const className = restriction.allowed ? 'requirement-allowed' : 'requirement-denied'; additionalDetails.appendChild(dt);
const verdict = restriction.allowed ? 'INCLUDED ✓' : 'EXCLUDED ✗'; additionalDetails.appendChild(dd);
return `<p class="${className}"><strong>${verdict}</strong> by <code>${escapeHtml(restriction.source || 'unknown')}</code>: ${escapeHtml(restriction.reason)}</p>`;
}).join('');
} }
function displayRequirements(requirements) { if (data.used_default !== undefined) {
const section = document.getElementById('requirements-section'); const dt = document.createElement('dt');
const container = document.getElementById('requirement-results'); dt.textContent = 'Used Default:';
section.style.display = requirements.length ? 'block' : 'none'; const dd = document.createElement('dd');
container.innerHTML = requirements.map(requirement => { dd.textContent = data.used_default ? 'Yes' : 'No';
const className = requirement.allowed ? 'requirement-allowed' : 'requirement-denied'; additionalDetails.appendChild(dt);
const verdict = requirement.allowed ? 'ALLOWED ✓' : 'DENIED ✗'; additionalDetails.appendChild(dd);
return `<p class="${className}"><strong>${escapeHtml(requirement.action)}: ${verdict}</strong> — ${escapeHtml(requirement.summary)}</p>`; }
}).join('');
if (data.depth !== undefined) {
const dt = document.createElement('dt');
dt.textContent = 'Depth:';
const dd = document.createElement('dd');
dd.textContent = data.depth;
additionalDetails.appendChild(dt);
additionalDetails.appendChild(dd);
}
// Raw JSON
document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
// Scroll to output
output.scrollIntoView({ behavior: 'smooth', block: 'nearest' });
} }
function displayError(data) { function displayError(data) {
output.style.display = 'block'; output.style.display = 'block';
output.className = 'denied'; output.className = 'denied';
const resultBadge = document.getElementById('result-badge'); const resultBadge = document.getElementById('result-badge');
resultBadge.className = 'result-badge denied-badge'; resultBadge.className = 'result-badge denied-badge';
resultBadge.textContent = 'ERROR'; resultBadge.textContent = 'ERROR';
document.getElementById('result-summary').textContent = data.error || 'Unknown error';
document.getElementById('result-actor').textContent = '—'; document.getElementById('result-action').textContent = 'N/A';
document.getElementById('result-action').textContent = '—'; document.getElementById('result-resource').textContent = 'N/A';
document.getElementById('result-resource').textContent = '—'; document.getElementById('result-actor').textContent = 'N/A';
document.getElementById('matching-rules').innerHTML = '';
document.getElementById('restrictions-section').style.display = 'none'; const additionalDetails = document.getElementById('additional-details');
document.getElementById('requirements-section').style.display = 'none'; additionalDetails.innerHTML = '<dt>Error:</dt><dd>' + (data.error || 'Unknown error') + '</dd>';
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2);
document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
output.scrollIntoView({ behavior: 'smooth', block: 'nearest' });
} }
form.addEventListener('submit', event => { // Disable child input if parent is empty
event.preventDefault(); const parentInput = document.getElementById('parent');
performCheck(); const childInput = document.getElementById('child');
childInput.addEventListener('focus', () => {
if (!parentInput.value) {
alert('Please specify a parent resource first before adding a child resource.');
parentInput.focus();
}
}); });
actionSelect.addEventListener('change', updateResourceFields);
(function initializeFromUrl() {
const params = populateFormFromURL();
updateResourceFields();
if (params.get('action')) {
performCheck();
}
})();
</script> </script>
{% endblock %} {% endblock %}

View file

@ -1,6 +1,6 @@
{% extends "base.html" %} {% extends "base.html" %}
{% block title %}Permission activity{% endblock %} {% block title %}Debug permissions{% endblock %}
{% block extra_head %} {% block extra_head %}
{% include "_permission_ui_styles.html" %} {% include "_permission_ui_styles.html" %}
@ -20,29 +20,45 @@
.check-action, .check-when, .check-result { .check-action, .check-when, .check-result {
font-size: 1.3em; font-size: 1.3em;
} }
textarea {
height: 10em;
width: 95%;
box-sizing: border-box;
padding: 0.5em;
border: 2px dotted black;
}
.two-col {
display: inline-block;
width: 48%;
}
.two-col label {
width: 48%;
}
@media only screen and (max-width: 576px) {
.two-col {
width: 100%;
}
}
</style> </style>
{% endblock %} {% endblock %}
{% block content %} {% block content %}
<h1>Permission activity</h1> <h1>Permission playground</h1>
{% set current_tab = "permissions" %} {% set current_tab = "permissions" %}
{% include "_permissions_debug_tabs.html" %} {% include "_permissions_debug_tabs.html" %}
<h2>Raw simulator</h2> <p>This tool lets you simulate an actor and a permission check for that actor.</p>
<p>This form runs a hypothetical permission check and returns its raw explanation JSON. Use the <a href="{{ urls.path('-/check') }}">Explain tool</a> for a visual explanation of the same decision.</p>
<div class="permission-form"> <div class="permission-form">
<form action="{{ urls.path('-/permissions') }}" id="debug-post" method="post"> <form action="{{ urls.path('-/permissions') }}" id="debug-post" method="post">
<div class="permission-form-grid"> <div class="two-col">
<div>
<div class="form-section"> <div class="form-section">
<label for="activity-actor">Actor</label> <label>Actor</label>
<textarea class="permission-textarea" id="activity-actor" name="actor">{% if actor_input %}{{ actor_input }}{% else %}{"id": "root"}{% endif %}</textarea> <textarea name="actor">{% if actor_input %}{{ actor_input }}{% else %}{"id": "root"}{% endif %}</textarea>
</div> </div>
</div> </div>
<div> <div class="two-col" style="vertical-align: top">
<div class="form-section"> <div class="form-section">
<label for="permission">Action</label> <label for="permission">Action</label>
<select name="permission" id="permission"> <select name="permission" id="permission">
@ -60,7 +76,6 @@
<input type="text" id="resource_2" name="resource_2" placeholder="e.g., table name"> <input type="text" id="resource_2" name="resource_2" placeholder="e.g., table name">
</div> </div>
</div> </div>
</div>
<div class="form-actions"> <div class="form-actions">
<button type="submit" class="submit-btn">Simulate permission check</button> <button type="submit" class="submit-btn">Simulate permission check</button>
</div> </div>
@ -110,7 +125,7 @@ debugPost.addEventListener('submit', function(ev) {
}); });
</script> </script>
<h2>Recent permission checks</h2> <h1>Recent permissions checks</h1>
<p> <p>
{% if filter != "all" %}<a href="?filter=all">All</a>{% else %}<strong>All</strong>{% endif %}, {% if filter != "all" %}<a href="?filter=all">All</a>{% else %}<strong>All</strong>{% endif %},

View file

@ -3,6 +3,7 @@
{% block title %}Permission Rules{% endblock %} {% block title %}Permission Rules{% endblock %}
{% block extra_head %} {% block extra_head %}
<script src="{{ base_url }}-/static/json-format-highlight-1.0.1.js"></script>
{% include "_permission_ui_styles.html" %} {% include "_permission_ui_styles.html" %}
{% include "_debug_common_functions.html" %} {% include "_debug_common_functions.html" %}
{% endblock %} {% endblock %}
@ -36,7 +37,7 @@
<div class="form-section"> <div class="form-section">
<label for="page_size">Page size:</label> <label for="page_size">Page size:</label>
<input type="number" id="page_size" name="_size" value="50" min="1" max="200"> <input type="number" id="page_size" name="page_size" value="50" min="1" max="200" style="max-width: 100px;">
<small>Number of results per page (max 200)</small> <small>Number of results per page (max 200)</small>
</div> </div>
@ -74,7 +75,7 @@ const submitBtn = document.getElementById('submit-btn');
(function() { (function() {
const params = populateFormFromURL(); const params = populateFormFromURL();
const action = params.get('action'); const action = params.get('action');
const page = params.get('_page'); const page = params.get('page');
if (action) { if (action) {
fetchResults(page ? parseInt(page) : 1); fetchResults(page ? parseInt(page) : 1);
} }
@ -88,14 +89,14 @@ async function fetchResults(page = 1) {
const params = new URLSearchParams(); const params = new URLSearchParams();
for (const [key, value] of formData.entries()) { for (const [key, value] of formData.entries()) {
if (value && key !== '_size' && key !== '_page') { if (value && key !== 'page_size') {
params.append(key, value); params.append(key, value);
} }
} }
const pageSize = document.getElementById('page_size').value || '50'; const pageSize = document.getElementById('page_size').value || '50';
params.append('_page', page.toString()); params.append('page', page.toString());
params.append('_size', pageSize); params.append('page_size', pageSize);
try { try {
const response = await fetch('{{ urls.path("-/rules.json") }}?' + params.toString(), { const response = await fetch('{{ urls.path("-/rules.json") }}?' + params.toString(), {
@ -184,7 +185,7 @@ function displayResults(data) {
} }
// Update raw JSON // Update raw JSON
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2); document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
} }
function displayError(data) { function displayError(data) {
@ -194,7 +195,7 @@ function displayError(data) {
resultsContent.innerHTML = `<div class="error-message">Error: ${escapeHtml(data.error || 'Unknown error')}</div>`; resultsContent.innerHTML = `<div class="error-message">Error: ${escapeHtml(data.error || 'Unknown error')}</div>`;
document.getElementById('raw-json').textContent = JSON.stringify(data, null, 2); document.getElementById('raw-json').innerHTML = jsonFormatHighlight(data);
} }
</script> </script>

View file

@ -2,7 +2,7 @@
<html lang="en"> <html lang="en">
<head> <head>
<title>Datasette: Pattern Portfolio</title> <title>Datasette: Pattern Portfolio</title>
<link rel="stylesheet" href="{{ static('app.css') }}"> <link rel="stylesheet" href="{{ base_url }}-/static/app.css?{{ app_css_hash }}">
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no"> <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
<meta name="robots" content="noindex"> <meta name="robots" content="noindex">
<style></style> <style></style>
@ -202,9 +202,9 @@
<h3>3 rows <h3>3 rows
where characteristic_id = 2 where characteristic_id = 2
</h3> </h3>
<form class="core filters" action="{{ base_url }}fixtures/roadside_attraction_characteristics" method="get"> <form class="filters" action="{{ base_url }}fixtures/roadside_attraction_characteristics" method="get">
<div class="search-row"><label for="_search">Search:</label><input id="_search" type="search" name="_search" value=""></div> <div class="search-row"><label for="_search">Search:</label><input id="_search" type="search" name="_search" value=""></div>
<div class="filter-row filter-controls-row"> <div class="filter-row">
<div class="select-wrapper"> <div class="select-wrapper">
<select name="_filter_column_1"> <select name="_filter_column_1">
<option value="">- remove filter -</option> <option value="">- remove filter -</option>
@ -238,7 +238,7 @@
</select> </select>
</div><input type="text" name="_filter_value_1" class="filter-value" value="2"> </div><input type="text" name="_filter_value_1" class="filter-value" value="2">
</div> </div>
<div class="filter-row filter-controls-row"> <div class="filter-row">
<div class="select-wrapper"> <div class="select-wrapper">
<select name="_filter_column"> <select name="_filter_column">
<option value="">- column -</option> <option value="">- column -</option>
@ -272,8 +272,8 @@
</select> </select>
</div><input type="text" name="_filter_value" class="filter-value"> </div><input type="text" name="_filter_value" class="filter-value">
</div> </div>
<div class="filter-row filter-actions-row"> <div class="filter-row">
<div class="select-wrapper"> <div class="select-wrapper small-screen-only">
<select name="_sort" id="sort_by"> <select name="_sort" id="sort_by">
<option value="">Sort...</option> <option value="">Sort...</option>
<option value="rowid" selected>Sort by rowid</option> <option value="rowid" selected>Sort by rowid</option>
@ -281,8 +281,8 @@
<option value="characteristic_id">Sort by characteristic_id</option> <option value="characteristic_id">Sort by characteristic_id</option>
</select> </select>
</div> </div>
<label class="sort_by_desc"><input type="checkbox" name="_sort_by_desc"> descending</label> <label class="sort_by_desc small-screen-only"><input type="checkbox" name="_sort_by_desc"> descending</label>
<input type="submit" value="Apply filters"> <input type="submit" value="Apply">
</div> </div>
</form> </form>

View file

@ -7,9 +7,9 @@
{% if row_mutation_ui %} {% if row_mutation_ui %}
<script>window._datasetteTableData = {{ table_page_data|tojson }};</script> <script>window._datasetteTableData = {{ table_page_data|tojson }};</script>
{% if table_page_data.foreignKeys %} {% if table_page_data.foreignKeys %}
<script src="{{ static('autocomplete.js') }}" defer></script> <script src="{{ urls.static('autocomplete.js') }}" defer></script>
{% endif %} {% endif %}
<script src="{{ static('edit-tools.js') }}" defer></script> <script src="{{ urls.static('edit-tools.js') }}?hash={{ edit_tools_js_hash }}" defer></script>
{% endif %} {% endif %}
<style> <style>
@media only screen and (max-width: 576px) { @media only screen and (max-width: 576px) {
@ -42,12 +42,24 @@
{% if foreign_key_tables %} {% if foreign_key_tables %}
<h2>Links from other tables</h2> <h2>Links from other tables</h2>
<ul> <ul class="row-foreign-key-tables">
{% for other in foreign_key_tables %} {% for other in foreign_key_tables %}
<li> <li>
<a href="{{ other.link }}"> <a href="{{ other.link }}">
{{ "{:,}".format(other.count) }} row{% if other.count == 1 %}{% else %}s{% endif %}</a> {{ "{:,}".format(other.count) }} row{% if other.count == 1 %}{% else %}s{% endif %}</a>
from {{ other.other_column }} in {{ other.other_table }} from {{ other.other_column }} in {{ other.other_table }}
{% if other.can_insert %}
<button
type="button"
class="core row-foreign-key-insert"
data-insert-dialog
data-insert-dialog-reload
data-database="{{ database }}"
data-table="{{ other.other_table }}"
data-row="{{ other.insert_row|tojson|forceescape }}"
data-message="{{ other.insert_message }}"
>Insert</button>
{% endif %}
</li> </li>
{% endfor %} {% endfor %}
</ul> </ul>

View file

@ -1,17 +1,17 @@
{% extends "base.html" %} {% extends "base.html" %}
{% block title %}{{ database }}: {{ table }}: {% if count_truncated %}{{ "{:,}".format(count - 1) }}+ rows{% elif count or count == 0 %}{{ "{:,}".format(count) }} row{% if count == 1 %}{% else %}s{% endif %}{% endif %}{% if human_description_en %} {{ human_description_en }}{% endif %}{% endblock %} {% block title %}{{ database }}: {{ table }}: {% if count or count == 0 %}{{ "{:,}".format(count) }} row{% if count == 1 %}{% else %}s{% endif %}{% endif %}{% if human_description_en %} {{ human_description_en }}{% endif %}{% endblock %}
{% block extra_head %} {% block extra_head %}
{{- super() -}} {{- super() -}}
<script>window._datasetteTableData = {{ table_page_data|tojson }};</script> <script>window._datasetteTableData = {{ table_page_data|tojson }};</script>
<script src="{{ static('column-chooser.js') }}" defer></script> <script src="{{ urls.static('column-chooser.js') }}" defer></script>
{% if table_page_data.foreignKeys %} {% if table_page_data.foreignKeys %}
<script src="{{ static('autocomplete.js') }}" defer></script> <script src="{{ urls.static('autocomplete.js') }}" defer></script>
{% endif %} {% endif %}
<script src="{{ static('edit-tools.js') }}" defer></script> <script src="{{ urls.static('edit-tools.js') }}?hash={{ edit_tools_js_hash }}" defer></script>
<script src="{{ static('table.js') }}" defer></script> <script src="{{ urls.static('table.js') }}?hash={{ table_js_hash }}" defer></script>
<script src="{{ static('mobile-column-actions.js') }}" defer></script> <script src="{{ urls.static('mobile-column-actions.js') }}" defer></script>
<script>DATASETTE_ALLOW_FACET = {{ datasette_allow_facet }};</script> <script>DATASETTE_ALLOW_FACET = {{ datasette_allow_facet }};</script>
<style> <style>
@media only screen and (max-width: 576px) { @media only screen and (max-width: 576px) {
@ -47,21 +47,20 @@
{% endif %} {% endif %}
{% if count or human_description_en %} {% if count or human_description_en %}
<h3 class="table-summary"> <h3>
{% if count_truncated %}<span class="table-count" aria-live="polite">{{ "{:,}".format(count - 1) }}+ rows</span> {% if count == count_limit + 1 %}&gt;{{ "{:,}".format(count_limit) }} rows
<button type="button" class="count-all" data-count-url="{{ urls.table(database, table) }}/-/count">count all</button> {% if allow_execute_sql and query.sql %} <a class="count-sql" style="font-size: 0.8em;" href="{{ urls.database_query(database, count_sql) }}">count all</a>{% endif %}
<span class="count-error" role="alert"></span>
{% elif count or count == 0 %}{{ "{:,}".format(count) }} row{% if count == 1 %}{% else %}s{% endif %}{% endif %} {% elif count or count == 0 %}{{ "{:,}".format(count) }} row{% if count == 1 %}{% else %}s{% endif %}{% endif %}
{% if human_description_en %}<span class="table-summary-description">{{ human_description_en }}</span>{% endif %} {% if human_description_en %}{{ human_description_en }}{% endif %}
</h3> </h3>
{% endif %} {% endif %}
<form class="core filters" action="{{ urls.table(database, table) }}" method="get"> <form class="core" class="filters" action="{{ urls.table(database, table) }}" method="get">
{% if supports_search %} {% if supports_search %}
<div class="search-row"><label for="_search">Search:</label><input id="_search" type="search" name="_search" value="{{ search }}"></div> <div class="search-row"><label for="_search">Search:</label><input id="_search" type="search" name="_search" value="{{ search }}"></div>
{% endif %} {% endif %}
{% for column, lookup, value in filters.selections() %} {% for column, lookup, value in filters.selections() %}
<div class="filter-row filter-controls-row"> <div class="filter-row">
<div class="select-wrapper"> <div class="select-wrapper">
<select name="_filter_column_{{ loop.index }}"> <select name="_filter_column_{{ loop.index }}">
<option value="">- remove filter -</option> <option value="">- remove filter -</option>
@ -78,7 +77,7 @@
</div><input type="text" name="_filter_value_{{ loop.index }}" class="filter-value" value="{{ value }}"> </div><input type="text" name="_filter_value_{{ loop.index }}" class="filter-value" value="{{ value }}">
</div> </div>
{% endfor %} {% endfor %}
<div class="filter-row filter-controls-row"> <div class="filter-row">
<div class="select-wrapper"> <div class="select-wrapper">
<select name="_filter_column"> <select name="_filter_column">
<option value="">- column -</option> <option value="">- column -</option>
@ -94,9 +93,9 @@
</select> </select>
</div><input type="text" name="_filter_value" class="filter-value"> </div><input type="text" name="_filter_value" class="filter-value">
</div> </div>
<div class="filter-row filter-actions-row"> <div class="filter-row">
{% if is_sortable %} {% if is_sortable %}
<div class="select-wrapper"> <div class="select-wrapper small-screen-only">
<select name="_sort" id="sort_by"> <select name="_sort" id="sort_by">
<option value="">Sort...</option> <option value="">Sort...</option>
{% for column in display_columns %} {% for column in display_columns %}
@ -106,12 +105,12 @@
{% endfor %} {% endfor %}
</select> </select>
</div> </div>
<label class="sort_by_desc"><input type="checkbox" name="_sort_by_desc" tabindex="0"{% if sort_desc %} checked{% endif %}> descending</label> <label class="sort_by_desc small-screen-only"><input type="checkbox" name="_sort_by_desc"{% if sort_desc %} checked{% endif %}> descending</label>
{% endif %} {% endif %}
{% for key, value in form_hidden_args %} {% for key, value in form_hidden_args %}
<input type="hidden" name="{{ key }}" value="{{ value }}"> <input type="hidden" name="{{ key }}" value="{{ value }}">
{% endfor %} {% endfor %}
<input type="submit" value="Apply filters" tabindex="0"> <input type="submit" value="Apply">
</div> </div>
</form> </form>

View file

@ -10,7 +10,7 @@ from __future__ import annotations
import dataclasses import dataclasses
import time import time
from typing import TYPE_CHECKING from typing import TYPE_CHECKING, Optional
import itsdangerous import itsdangerous
@ -18,21 +18,6 @@ if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
class TokenInvalid(Exception):
"""
Raised by a TokenHandler when a token it recognizes is invalid -
for example a bad signature, malformed payload or expired token.
Datasette responds to this with an HTTP 401 error. Handlers should
return None instead for tokens they do not recognize at all, so that
other registered handlers get a chance to verify them.
"""
def __init__(self, message="Invalid token"):
self.message = message
super().__init__(message)
@dataclasses.dataclass @dataclasses.dataclass
class TokenRestrictions: class TokenRestrictions:
""" """
@ -50,24 +35,24 @@ class TokenRestrictions:
database: dict[str, list[str]] = dataclasses.field(default_factory=dict) database: dict[str, list[str]] = dataclasses.field(default_factory=dict)
resource: dict[str, dict[str, list[str]]] = dataclasses.field(default_factory=dict) resource: dict[str, dict[str, list[str]]] = dataclasses.field(default_factory=dict)
def allow_all(self, action: str) -> TokenRestrictions: def allow_all(self, action: str) -> "TokenRestrictions":
"""Allow an action across all databases and resources.""" """Allow an action across all databases and resources."""
self.all.append(action) self.all.append(action)
return self return self
def allow_database(self, database: str, action: str) -> TokenRestrictions: def allow_database(self, database: str, action: str) -> "TokenRestrictions":
"""Allow an action on a specific database.""" """Allow an action on a specific database."""
self.database.setdefault(database, []).append(action) self.database.setdefault(database, []).append(action)
return self return self
def allow_resource( def allow_resource(
self, database: str, resource: str, action: str self, database: str, resource: str, action: str
) -> TokenRestrictions: ) -> "TokenRestrictions":
"""Allow an action on a specific resource within a database.""" """Allow an action on a specific resource within a database."""
self.resource.setdefault(database, {}).setdefault(resource, []).append(action) self.resource.setdefault(database, {}).setdefault(resource, []).append(action)
return self return self
def abbreviated(self, datasette: Datasette) -> dict | None: def abbreviated(self, datasette: "Datasette") -> Optional[dict]:
""" """
Return the abbreviated ``_r`` dictionary shape for this set of Return the abbreviated ``_r`` dictionary shape for this set of
restrictions, using action abbreviations registered with ``datasette``. restrictions, using action abbreviations registered with ``datasette``.
@ -112,23 +97,19 @@ class TokenHandler:
async def create_token( async def create_token(
self, self,
datasette: Datasette, datasette: "Datasette",
actor_id: str, actor_id: str,
*, *,
expires_after: int | None = None, expires_after: Optional[int] = None,
restrictions: TokenRestrictions | None = None, restrictions: Optional[TokenRestrictions] = None,
) -> str: ) -> str:
"""Create and return a token string for the given actor.""" """Create and return a token string for the given actor."""
raise NotImplementedError raise NotImplementedError
async def verify_token(self, datasette: Datasette, token: str) -> dict | None: async def verify_token(self, datasette: "Datasette", token: str) -> Optional[dict]:
""" """
Verify a token and return an actor dict. Verify a token and return an actor dict, or None if this handler
does not recognize the token.
Return None if this handler does not recognize the token at all,
so other handlers can try it. Raise TokenInvalid if the token is
recognized but invalid (bad signature, malformed, expired) - the
request will fail with a 401 error.
""" """
raise NotImplementedError raise NotImplementedError
@ -142,11 +123,11 @@ class SignedTokenHandler(TokenHandler):
async def create_token( async def create_token(
self, self,
datasette: Datasette, datasette: "Datasette",
actor_id: str, actor_id: str,
*, *,
expires_after: int | None = None, expires_after: Optional[int] = None,
restrictions: TokenRestrictions | None = None, restrictions: Optional[TokenRestrictions] = None,
) -> str: ) -> str:
if not datasette.setting("allow_signed_tokens"): if not datasette.setting("allow_signed_tokens"):
raise ValueError( raise ValueError(
@ -163,35 +144,32 @@ class SignedTokenHandler(TokenHandler):
token["_r"] = abbreviated token["_r"] = abbreviated
return "dstok_{}".format(datasette.sign(token, namespace="token")) return "dstok_{}".format(datasette.sign(token, namespace="token"))
async def verify_token(self, datasette: Datasette, token: str) -> dict | None: async def verify_token(self, datasette: "Datasette", token: str) -> Optional[dict]:
prefix = "dstok_" prefix = "dstok_"
if not token.startswith(prefix): if not datasette.setting("allow_signed_tokens"):
# Not one of our tokens - leave it for other handlers
return None return None
if not datasette.setting("allow_signed_tokens"):
raise TokenInvalid(
"Signed tokens are not enabled for this Datasette instance"
)
max_signed_tokens_ttl = datasette.setting("max_signed_tokens_ttl") max_signed_tokens_ttl = datasette.setting("max_signed_tokens_ttl")
if not token.startswith(prefix):
return None
raw = token[len(prefix) :] raw = token[len(prefix) :]
try: try:
decoded = datasette.unsign(raw, namespace="token") decoded = datasette.unsign(raw, namespace="token")
except itsdangerous.BadSignature: except itsdangerous.BadSignature:
raise TokenInvalid("Invalid token signature") return None
if "t" not in decoded: if "t" not in decoded:
raise TokenInvalid("Invalid token: no timestamp") return None
created = decoded["t"] created = decoded["t"]
if not isinstance(created, int): if not isinstance(created, int):
raise TokenInvalid("Invalid token: invalid timestamp") return None
duration = decoded.get("d") duration = decoded.get("d")
if duration is not None and not isinstance(duration, int): if duration is not None and not isinstance(duration, int):
raise TokenInvalid("Invalid token: invalid duration") return None
if (duration is None and max_signed_tokens_ttl) or ( if (duration is None and max_signed_tokens_ttl) or (
duration is not None duration is not None
@ -200,8 +178,9 @@ class SignedTokenHandler(TokenHandler):
): ):
duration = max_signed_tokens_ttl duration = max_signed_tokens_ttl
if duration and time.time() - created > duration: if duration:
raise TokenInvalid("Token has expired") if time.time() - created > duration:
return None
actor = {"id": decoded["a"], "token": "dstok"} actor = {"id": decoded["a"], "token": "dstok"}

View file

@ -1,11 +1,10 @@
import asyncio import asyncio
import json
import time
import traceback
from contextlib import contextmanager from contextlib import contextmanager
from contextvars import ContextVar from contextvars import ContextVar
from markupsafe import escape from markupsafe import escape
import time
import json
import traceback
tracers = {} tracers = {}
@ -133,17 +132,17 @@ class AsgiTracer:
"num_traces": len(traces), "num_traces": len(traces),
"traces": traces, "traces": traces,
} }
content_type = next( try:
( content_type = [
v.decode("utf8") v.decode("utf8")
for k, v in response_headers for k, v in response_headers
if k.lower() == b"content-type" if k.lower() == b"content-type"
), ][0]
"", except IndexError:
) content_type = ""
if "text/html" in content_type and b"</body>" in accumulated_body: if "text/html" in content_type and b"</body>" in accumulated_body:
extra = escape(json.dumps(trace_info, indent=2)) extra = escape(json.dumps(trace_info, indent=2))
extra_html = f"<pre>{extra}</pre></body>".encode() extra_html = f"<pre>{extra}</pre></body>".encode("utf8")
accumulated_body = accumulated_body.replace(b"</body>", extra_html) accumulated_body = accumulated_body.replace(b"</body>", extra_html)
elif "json" in content_type and accumulated_body.startswith(b"{"): elif "json" in content_type and accumulated_body.startswith(b"{"):
data = json.loads(accumulated_body.decode("utf8")) data = json.loads(accumulated_body.decode("utf8"))

View file

@ -1,7 +1,6 @@
from .utils import tilde_encode, path_with_format, PrefixedUrlString
import urllib import urllib
from .utils import PrefixedUrlString, path_with_format, tilde_encode
class Urls: class Urls:
def __init__(self, ds): def __init__(self, ds):
@ -9,7 +8,8 @@ class Urls:
def path(self, path, format=None): def path(self, path, format=None):
if not isinstance(path, PrefixedUrlString): if not isinstance(path, PrefixedUrlString):
path = path.removeprefix("/") if path.startswith("/"):
path = path[1:]
path = self.ds.setting("base_url") + path path = self.ds.setting("base_url") + path
if format is not None: if format is not None:
path = path_with_format(path=path, format=format) path = path_with_format(path=path, format=format)
@ -56,7 +56,6 @@ class Urls:
return PrefixedUrlString(path) return PrefixedUrlString(path)
def row_blob(self, database, table, row_path, column): def row_blob(self, database, table, row_path, column):
return ( return self.table(database, table) + "/{}.blob?_blob_column={}".format(
self.table(database, table) row_path, urllib.parse.quote_plus(column)
+ f"/{row_path}.blob?_blob_column={urllib.parse.quote_plus(column)}"
) )

View file

@ -1,31 +1,28 @@
import asyncio import asyncio
import base64 from contextlib import contextmanager
import binascii import aiofiles
import click
from collections import OrderedDict, namedtuple, Counter
import copy import copy
import dataclasses import dataclasses
import base64
import hashlib import hashlib
import inspect import inspect
import json import json
import os
import re
import secrets
import shlex
import shutil
import tempfile
import time
import types
import typing
import urllib
from collections import Counter, OrderedDict, namedtuple
from collections.abc import Iterable
from contextlib import contextmanager
import aiofiles
import click
import markupsafe import markupsafe
import mergedeep import mergedeep
import os
import re
import shlex
import tempfile
import typing
import time
import types
import secrets
import shutil
from typing import Iterable, List, Tuple
import urllib
import yaml import yaml
from .shutil_backport import copytree from .shutil_backport import copytree
from .sqlite import sqlite3, supports_table_xinfo from .sqlite import sqlite3, supports_table_xinfo
@ -38,7 +35,7 @@ if typing.TYPE_CHECKING:
class PaginatedResources: class PaginatedResources:
"""Paginated results from allowed_resources query.""" """Paginated results from allowed_resources query."""
resources: list["Resource"] resources: List["Resource"]
next: str | None # Keyset token for next page (None if no more results) next: str | None # Keyset token for next page (None if no more results)
_datasette: typing.Any = dataclasses.field(default=None, repr=False) _datasette: typing.Any = dataclasses.field(default=None, repr=False)
_action: str = dataclasses.field(default=None, repr=False) _action: str = dataclasses.field(default=None, repr=False)
@ -85,132 +82,22 @@ class PaginatedResources:
# From https://www.sqlite.org/lang_keywords.html # From https://www.sqlite.org/lang_keywords.html
reserved_words = { reserved_words = set(
"abort", (
"action", "abort action add after all alter analyze and as asc attach autoincrement "
"add", "before begin between by cascade case cast check collate column commit "
"after", "conflict constraint create cross current_date current_time "
"all", "current_timestamp database default deferrable deferred delete desc detach "
"alter", "distinct drop each else end escape except exclusive exists explain fail "
"analyze", "for foreign from full glob group having if ignore immediate in index "
"and", "indexed initially inner insert instead intersect into is isnull join key "
"as", "left like limit match natural no not notnull null of offset on or order "
"asc", "outer plan pragma primary query raise recursive references regexp reindex "
"attach", "release rename replace restrict right rollback row savepoint select set "
"autoincrement", "table temp temporary then to transaction trigger union unique update using "
"before", "vacuum values view virtual when where with without"
"begin", ).split()
"between", )
"by",
"cascade",
"case",
"cast",
"check",
"collate",
"column",
"commit",
"conflict",
"constraint",
"create",
"cross",
"current_date",
"current_time",
"current_timestamp",
"database",
"default",
"deferrable",
"deferred",
"delete",
"desc",
"detach",
"distinct",
"drop",
"each",
"else",
"end",
"escape",
"except",
"exclusive",
"exists",
"explain",
"fail",
"for",
"foreign",
"from",
"full",
"glob",
"group",
"having",
"if",
"ignore",
"immediate",
"in",
"index",
"indexed",
"initially",
"inner",
"insert",
"instead",
"intersect",
"into",
"is",
"isnull",
"join",
"key",
"left",
"like",
"limit",
"match",
"natural",
"no",
"not",
"notnull",
"null",
"of",
"offset",
"on",
"or",
"order",
"outer",
"plan",
"pragma",
"primary",
"query",
"raise",
"recursive",
"references",
"regexp",
"reindex",
"release",
"rename",
"replace",
"restrict",
"right",
"rollback",
"row",
"savepoint",
"select",
"set",
"table",
"temp",
"temporary",
"then",
"to",
"transaction",
"trigger",
"union",
"unique",
"update",
"using",
"vacuum",
"values",
"view",
"virtual",
"when",
"where",
"with",
"without",
}
APT_GET_DOCKERFILE_EXTRAS = r""" APT_GET_DOCKERFILE_EXTRAS = r"""
RUN apt-get update && \ RUN apt-get update && \
@ -270,7 +157,7 @@ functions_marked_as_documented = []
def documented(fn=None, *, label=None): def documented(fn=None, *, label=None):
def decorate(fn): def decorate(fn):
fn._datasette_docs_label = label or f"internals_utils_{fn.__name__}" fn._datasette_docs_label = label or "internals_utils_{}".format(fn.__name__)
functions_marked_as_documented.append(fn) functions_marked_as_documented.append(fn)
return fn return fn
@ -337,35 +224,17 @@ def compound_keys_after_sql(pks, start_index=0):
return "({})".format("\n or\n".join(or_clauses)) return "({})".format("\n or\n".join(or_clauses))
@documented
class CustomJSONEncoder(json.JSONEncoder): class CustomJSONEncoder(json.JSONEncoder):
"""
The CustomJSONEncoder class handles serialization for objects commonly used by Datasette,
including SQLite cursors and binary blobs. Datasette uses it internally to serve .json endpoints,
and plugins that return JSON can use it to match Datasette's own handling.
Built-in types (text, numbers, lists, etc) are encoded the same as Python's built-in ``json`` module.
- ``sqlite3.Row`` becomes a tuple
- ``sqlite3.Cursor`` becomes a list
Binary blobs are encoded as an object, with the actual data base64-encoded,
like so: ::
{
"$base64": True,
"encoded": ...,
}
Example: https://latest.datasette.io/fixtures/binary_data.json
"""
def default(self, obj): def default(self, obj):
if isinstance(obj, sqlite3.Row): if isinstance(obj, sqlite3.Row):
return tuple(obj) return tuple(obj)
if isinstance(obj, sqlite3.Cursor): if isinstance(obj, sqlite3.Cursor):
return list(obj) return list(obj)
if isinstance(obj, bytes): if isinstance(obj, bytes):
# Does it encode to utf8?
try:
return obj.decode("utf8")
except UnicodeDecodeError:
return { return {
"$base64": True, "$base64": True,
"encoded": base64.b64encode(obj).decode("latin1"), "encoded": base64.b64encode(obj).decode("latin1"),
@ -373,35 +242,6 @@ class CustomJSONEncoder(json.JSONEncoder):
return json.JSONEncoder.default(self, obj) return json.JSONEncoder.default(self, obj)
class WriteJsonValueError(ValueError):
pass
def decode_write_json_cell(value):
if not isinstance(value, dict):
return value
keys = set(value.keys())
if keys == {"$raw"}:
return value["$raw"]
if keys == {"$base64", "encoded"} and value.get("$base64") is True:
encoded = value["encoded"]
if not isinstance(encoded, str):
raise WriteJsonValueError("$base64 encoded value must be a string")
try:
return base64.b64decode(encoded, validate=True)
except binascii.Error as ex:
raise WriteJsonValueError("Invalid $base64 encoded value") from ex
return value
def decode_write_json_row(row):
return {key: decode_write_json_cell(value) for key, value in row.items()}
def decode_write_json_rows(rows):
return [decode_write_json_row(row) for row in rows]
@contextmanager @contextmanager
def sqlite_timelimit(conn, ms): def sqlite_timelimit(conn, ms):
deadline = time.perf_counter() + (ms / 1000) deadline = time.perf_counter() + (ms / 1000)
@ -472,7 +312,7 @@ disallawed_sql_res = [
( (
re.compile(f"pragma(?!_({'|'.join(allowed_pragmas)}))"), re.compile(f"pragma(?!_({'|'.join(allowed_pragmas)}))"),
"Statement contained a disallowed PRAGMA. Allowed pragma functions are {}".format( "Statement contained a disallowed PRAGMA. Allowed pragma functions are {}".format(
", ".join(f"pragma_{pragma}()" for pragma in allowed_pragmas) ", ".join("pragma_{}()".format(pragma) for pragma in allowed_pragmas)
), ),
) )
] ]
@ -568,9 +408,14 @@ def escape_css_string(s):
def escape_sqlite(s): def escape_sqlite(s):
if _boring_keyword_re.fullmatch(s) and (s.lower() not in reserved_words): if _boring_keyword_re.match(s) and (s.lower() not in reserved_words):
return s return s
elif "]" in s:
# SQLite does not support escaping ] inside [bracket] quoting, so fall
# back to double-quote quoting (doubling any embedded ") - #2677
return '"{}"'.format(s.replace('"', '""')) return '"{}"'.format(s.replace('"', '""'))
else:
return f"[{s}]"
def make_dockerfile( def make_dockerfile(
@ -646,7 +491,10 @@ CMD {cmd}""".format(
else "" else ""
), ),
environment_variables="\n".join( environment_variables="\n".join(
[f"ENV {key} '{value}'" for key, value in environment_variables.items()] [
"ENV {} '{}'".format(key, value)
for key, value in environment_variables.items()
]
), ),
install_from=" ".join(install), install_from=" ".join(install),
files=" ".join(files), files=" ".join(files),
@ -745,11 +593,11 @@ def detect_primary_keys(conn, table):
def get_outbound_foreign_keys(conn, table): def get_outbound_foreign_keys(conn, table):
infos = conn.execute(f"PRAGMA foreign_key_list({escape_sqlite(table)})").fetchall() infos = conn.execute(f"PRAGMA foreign_key_list([{table}])").fetchall()
fks = [] fks = []
for info in infos: for info in infos:
if info is not None: if info is not None:
id, seq, table_name, from_, to_, _on_update, _on_delete, _match = info id, seq, table_name, from_, to_, on_update, on_delete, match = info
fks.append( fks.append(
{ {
"column": from_, "column": from_,
@ -820,8 +668,7 @@ def detect_spatialite(conn):
def detect_fts(conn, table): def detect_fts(conn, table):
"""Detect if table has a corresponding FTS virtual table and return it""" """Detect if table has a corresponding FTS virtual table and return it"""
sql, params = detect_fts_sql(table) rows = conn.execute(detect_fts_sql(table)).fetchall()
rows = conn.execute(sql, params).fetchall()
if len(rows) == 0: if len(rows) == 0:
return None return None
else: else:
@ -829,26 +676,18 @@ def detect_fts(conn, table):
def detect_fts_sql(table): def detect_fts_sql(table):
escaped_table = table.replace("\\", "\\\\").replace("%", "\\%").replace("_", "\\_") return r"""
return (
r"""
select name from sqlite_master select name from sqlite_master
where rootpage = 0 where rootpage = 0
and ( and (
sql like :fts_double_quoted escape char(92) sql like '%VIRTUAL TABLE%USING FTS%content="{table}"%'
or sql like :fts_bracket_quoted escape char(92) or sql like '%VIRTUAL TABLE%USING FTS%content=[{table}]%'
or ( or (
tbl_name = :table tbl_name = "{table}"
and sql like '%VIRTUAL TABLE%USING FTS%' and sql like '%VIRTUAL TABLE%USING FTS%'
) )
) )
""", """.format(table=table.replace("'", "''"))
{
"fts_double_quoted": f'%VIRTUAL TABLE%USING FTS%content="{escaped_table}"%',
"fts_bracket_quoted": f"%VIRTUAL TABLE%USING FTS%content=[{escaped_table}]%",
"table": table,
},
)
def detect_json1(conn=None): def detect_json1(conn=None):
@ -859,7 +698,7 @@ def detect_json1(conn=None):
try: try:
conn.execute("SELECT json('{}')") conn.execute("SELECT json('{}')")
return True return True
except sqlite3.Error: except Exception:
return False return False
finally: finally:
if close_conn: if close_conn:
@ -939,7 +778,9 @@ def is_url(value):
if not value.startswith("http://") and not value.startswith("https://"): if not value.startswith("http://") and not value.startswith("https://"):
return False return False
# Any whitespace at all is invalid # Any whitespace at all is invalid
return not whitespace_re.search(value) if whitespace_re.search(value):
return False
return True
css_class_re = re.compile(r"^[a-zA-Z]+[_a-zA-Z0-9-]*$") css_class_re = re.compile(r"^[a-zA-Z]+[_a-zA-Z0-9-]*$")
@ -992,9 +833,7 @@ def module_from_path(path, name):
mod.__file__ = path mod.__file__ = path
with open(path, "r") as file: with open(path, "r") as file:
code = compile(file.read(), path, "exec", dont_inherit=True) code = compile(file.read(), path, "exec", dont_inherit=True)
# Executing the file is the whole point - this is how --plugins-dir loads exec(code, mod.__dict__)
# plugins and how metadata/config .py files are evaluated
exec(code, mod.__dict__) # noqa: S102
return mod return mod
@ -1151,7 +990,9 @@ def escape_fts(query):
query += '"' query += '"'
bits = _escape_fts_re.split(query) bits = _escape_fts_re.split(query)
bits = [b for b in bits if b and b != '""'] bits = [b for b in bits if b and b != '""']
return " ".join(f'"{bit}"' if not bit.startswith('"') else bit for bit in bits) return " ".join(
'"{}"'.format(bit) if not bit.startswith('"') else bit for bit in bits
)
class MultiParams: class MultiParams:
@ -1163,7 +1004,7 @@ class MultiParams:
data[key], (list, tuple) data[key], (list, tuple)
), "dictionary data should be a dictionary of key => [list]" ), "dictionary data should be a dictionary of key => [list]"
self._data = data self._data = data
elif isinstance(data, (list, tuple)): elif isinstance(data, list) or isinstance(data, tuple):
new_data = {} new_data = {}
for item in data: for item in data:
assert ( assert (
@ -1253,7 +1094,9 @@ def _gather_arguments(fn, kwargs):
for parameter in parameters: for parameter in parameters:
if parameter not in kwargs: if parameter not in kwargs:
raise TypeError( raise TypeError(
f"{fn} requires parameters {tuple(parameters)}, missing: {set(parameters) - set(kwargs.keys())}" "{} requires parameters {}, missing: {}".format(
fn, tuple(parameters), set(parameters) - set(kwargs.keys())
)
) )
call_with.append(kwargs[parameter]) call_with.append(kwargs[parameter])
return call_with return call_with
@ -1322,9 +1165,9 @@ def resolve_env_secrets(config, environ):
"""Create copy that recursively replaces {"$env": "NAME"} with values from environ""" """Create copy that recursively replaces {"$env": "NAME"} with values from environ"""
if isinstance(config, dict): if isinstance(config, dict):
if list(config.keys()) == ["$env"]: if list(config.keys()) == ["$env"]:
return environ.get(next(iter(config.values()))) return environ.get(list(config.values())[0])
elif list(config.keys()) == ["$file"]: elif list(config.keys()) == ["$file"]:
with open(next(iter(config.values()))) as fp: with open(list(config.values())[0]) as fp:
return fp.read() return fp.read()
else: else:
return { return {
@ -1402,38 +1245,29 @@ class StartupError(Exception):
pass pass
# Comments and string literals, matched in a single pass so that whichever _single_line_comment_re = re.compile(r"--.*")
# construct starts first "wins" - this ensures a comment marker inside a string _multi_line_comment_re = re.compile(r"/\*.*?\*/", re.DOTALL)
# literal (or a quote inside a comment) does not confuse the parameter scan. _single_quote_re = re.compile(r"'(?:''|[^'])*'")
_comments_and_strings_re = re.compile( _double_quote_re = re.compile(r'"(?:\"\"|[^"])*"')
r"""
--[^\n]* # single line comment
| /\*.*?(?:\*/|\Z) # multi line comment, possibly to end-of-input
| '(?:''|[^'])*' # single quoted string ('' escapes a quote)
| "(?:""|[^"])*" # double quoted identifier ("" escapes a quote)
| \[(?:[^\]])*\] # square-bracket quoted identifier
| `(?:``|[^`])*` # backtick quoted identifier
""",
re.DOTALL | re.VERBOSE,
)
_named_param_re = re.compile(r":(\w+)") _named_param_re = re.compile(r":(\w+)")
@documented @documented
def named_parameters(sql: str) -> list[str]: def named_parameters(sql: str) -> List[str]:
""" """
Given a SQL statement, return a list of named parameters that are used in the statement Given a SQL statement, return a list of named parameters that are used in the statement
e.g. for ``select * from foo where id=:id`` this would return ``["id"]`` e.g. for ``select * from foo where id=:id`` this would return ``["id"]``
""" """
# Strip comments and string literals first so that any ":name" sequences sql = _single_line_comment_re.sub("", sql)
# inside them are not mistaken for named parameters sql = _multi_line_comment_re.sub("", sql)
sql = _comments_and_strings_re.sub("", sql) sql = _single_quote_re.sub("", sql)
sql = _double_quote_re.sub("", sql)
# Extract parameters from what is left # Extract parameters from what is left
return _named_param_re.findall(sql) return _named_param_re.findall(sql)
async def derive_named_parameters(db: "Database", sql: str) -> list[str]: async def derive_named_parameters(db: "Database", sql: str) -> List[str]:
""" """
This undocumented but stable method exists for backwards compatibility This undocumented but stable method exists for backwards compatibility
with plugins that were using it before it switched to named_parameters() with plugins that were using it before it switched to named_parameters()
@ -1441,54 +1275,6 @@ async def derive_named_parameters(db: "Database", sql: str) -> list[str]:
return named_parameters(sql) return named_parameters(sql)
def parse_size_limit(value, default, maximum, name="_size"):
"""
Parse a page-size parameter using the same semantics as the table
view's ?_size=: blank means default, "max" means maximum, integers
must be 0 or greater and no larger than maximum. Raises ValueError
with a message suitable for a 400 response.
"""
if value in (None, ""):
return default
if value == "max":
return maximum
try:
size = int(value)
if size < 0:
raise ValueError
except ValueError:
raise ValueError(f"{name} must be a positive integer")
if size > maximum:
raise ValueError(f"{name} must be <= {maximum}")
return size
UNSTABLE_API_MESSAGE = (
"This API is not part of Datasette's stable interface and may change at any time"
)
def error_body(messages, status):
"""
The canonical JSON error body used by every Datasette JSON error response:
{"ok": False, "error": "...", "errors": ["...", ...], "status": 400}
"error" is all of the messages joined with "; ", "errors" is the full
list, "status" matches the HTTP status code. Callers may add extra
context keys to the returned dictionary but must not remove these four.
"""
if isinstance(messages, str):
messages = [messages]
messages = [str(message) for message in messages]
return {
"ok": False,
"error": "; ".join(messages),
"errors": messages,
"status": status,
}
def add_cors_headers(headers): def add_cors_headers(headers):
headers["Access-Control-Allow-Origin"] = "*" headers["Access-Control-Allow-Origin"] = "*"
headers["Access-Control-Allow-Headers"] = "Authorization, Content-Type" headers["Access-Control-Allow-Headers"] = "Authorization, Content-Type"
@ -1517,7 +1303,7 @@ class TildeEncoder(dict):
elif b == _space: elif b == _space:
res = "+" res = "+"
else: else:
res = f"~{b:02X}" res = "~{:02X}".format(b)
self[b] = res self[b] = res
return res return res
@ -1566,13 +1352,7 @@ async def row_sql_params_pks(db, table, pk_values):
if use_rowid: if use_rowid:
select = "rowid, *" select = "rowid, *"
pks = ["rowid"] pks = ["rowid"]
wheres = [] wheres = [f'"{pk}"=:p{i}' for i, pk in enumerate(pks)]
for i, pk in enumerate(pks):
escaped_pk = escape_sqlite(pk)
# Preserve the historic always-quoted SQL exposed by _extra=query
if escaped_pk == pk:
escaped_pk = f'"{pk}"'
wheres.append(f"{escaped_pk}=:p{i}")
sql = f"select {select} from {escape_sqlite(table)} where {' AND '.join(wheres)}" sql = f"select {select} from {escape_sqlite(table)} where {' AND '.join(wheres)}"
params = {} params = {}
for i, pk_value in enumerate(pk_values): for i, pk_value in enumerate(pk_values):
@ -1618,7 +1398,7 @@ def _combine(base: dict, update: dict) -> dict:
return base return base
def pairs_to_nested_config(pairs: list[tuple[str, typing.Any]]) -> dict: def pairs_to_nested_config(pairs: typing.List[typing.Tuple[str, typing.Any]]) -> dict:
""" """
Parse a list of key-value pairs into a nested dictionary. Parse a list of key-value pairs into a nested dictionary.
""" """
@ -1633,7 +1413,7 @@ def make_slot_function(name, datasette, request, **kwargs):
from datasette.plugins import pm from datasette.plugins import pm
method = getattr(pm.hook, name, None) method = getattr(pm.hook, name, None)
assert method is not None, f"No hook found for {name}" assert method is not None, "No hook found for {}".format(name)
async def inner(): async def inner():
html_bits = [] html_bits = []
@ -1657,7 +1437,7 @@ def prune_empty_dicts(d: dict):
d.pop(key, None) d.pop(key, None)
def move_plugins_and_allow(source: dict, destination: dict) -> tuple[dict, dict]: def move_plugins_and_allow(source: dict, destination: dict) -> Tuple[dict, dict]:
""" """
Move 'plugins' and 'allow' keys from source to destination dictionary. Creates Move 'plugins' and 'allow' keys from source to destination dictionary. Creates
hierarchy in destination if needed. After moving, recursively remove any keys hierarchy in destination if needed. After moving, recursively remove any keys
@ -1744,7 +1524,7 @@ def redact_keys(original: dict, key_patterns: Iterable) -> dict:
return { return {
k: ( k: (
redact(v) redact(v)
if not any(pattern in k.casefold() for pattern in key_patterns) if not any(pattern in k for pattern in key_patterns)
else "***" else "***"
) )
for k, v in data.items() for k, v in data.items()
@ -1768,17 +1548,6 @@ def md5_not_usedforsecurity(s):
_etag_cache = {} _etag_cache = {}
def sha256_file(filepath, chunk_size=4096):
hasher = hashlib.sha256()
with open(filepath, "rb") as fp:
while True:
chunk = fp.read(chunk_size)
if not chunk:
break
hasher.update(chunk)
return hasher.hexdigest()
async def calculate_etag(filepath, chunk_size=4096): async def calculate_etag(filepath, chunk_size=4096):
if filepath in _etag_cache: if filepath in _etag_cache:
return _etag_cache[filepath] return _etag_cache[filepath]

View file

@ -29,15 +29,6 @@ from datasette.utils.permissions import gather_permission_sql_from_hooks
if TYPE_CHECKING: if TYPE_CHECKING:
from datasette.app import Datasette from datasette.app import Datasette
from datasette.permissions import Action
def _child_collation(action: "Action") -> str:
"""Match resource identity without changing the spelling returned by SQL."""
resource_class = action.resource_class
if resource_class is not None and resource_class.case_insensitive_child:
return "NOCASE"
return "BINARY"
async def build_allowed_resources_sql( async def build_allowed_resources_sql(
@ -158,7 +149,6 @@ async def _build_single_action_sql(
raise ValueError(f"Unknown action: {action}") raise ValueError(f"Unknown action: {action}")
# Get base resources SQL from the resource class # Get base resources SQL from the resource class
child_collation = _child_collation(action_obj)
base_resources_sql = await action_obj.resource_class.resources_sql( base_resources_sql = await action_obj.resource_class.resources_sql(
datasette, actor=actor datasette, actor=actor
) )
@ -195,7 +185,7 @@ async def _build_single_action_sql(
if permission_sql.sql is None: if permission_sql.sql is None:
continue continue
rule_sqls.append(f""" rule_sqls.append(f"""
SELECT parent, child COLLATE {child_collation} AS child, allow, reason, '{permission_sql.source}' AS source_plugin FROM ( SELECT parent, child, allow, reason, '{permission_sql.source}' AS source_plugin FROM (
{permission_sql.sql} {permission_sql.sql}
) )
""".strip()) """.strip())
@ -262,62 +252,88 @@ async def _build_single_action_sql(
] ]
) )
# Continue with the cascading logic. # Continue with the cascading logic
# Aggregate the RULES by cascade level (small), rather than grouping
# base x rules (which scales with the number of resources).
def _agg(select_key, where, group_by):
parts = [
f" SELECT {select_key}",
" MAX(CASE WHEN allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN allow = 1 THEN 1 ELSE 0 END) AS any_allow,",
" json_group_array(CASE WHEN allow = 0 THEN source_plugin || ': ' || reason END) AS deny_reasons,",
" json_group_array(CASE WHEN allow = 1 THEN source_plugin || ': ' || reason END) AS allow_reasons",
f" FROM all_rules WHERE {where}",
]
if group_by:
parts.append(f" GROUP BY {group_by}")
return parts
query_parts.extend( query_parts.extend(
["child_agg AS ("] [
+ _agg( "child_lvl AS (",
"parent, child,", " SELECT b.parent, b.child,",
"parent IS NOT NULL AND child IS NOT NULL", " MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
"parent, child", " MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow,",
) " json_group_array(CASE WHEN ar.allow = 0 THEN ar.source_plugin || ': ' || ar.reason END) AS deny_reasons,",
+ ["),", "parent_agg AS ("] " json_group_array(CASE WHEN ar.allow = 1 THEN ar.source_plugin || ': ' || ar.reason END) AS allow_reasons",
+ _agg("parent,", "parent IS NOT NULL AND child IS NULL", "parent") " FROM base b",
+ ["),", "global_agg AS ("] " LEFT JOIN all_rules ar ON ar.parent = b.parent AND ar.child = b.child",
+ _agg("", "parent IS NULL AND child IS NULL", None) " GROUP BY b.parent, b.child",
+ ["),"] "),",
"parent_lvl AS (",
" SELECT b.parent, b.child,",
" MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow,",
" json_group_array(CASE WHEN ar.allow = 0 THEN ar.source_plugin || ': ' || ar.reason END) AS deny_reasons,",
" json_group_array(CASE WHEN ar.allow = 1 THEN ar.source_plugin || ': ' || ar.reason END) AS allow_reasons",
" FROM base b",
" LEFT JOIN all_rules ar ON ar.parent = b.parent AND ar.child IS NULL",
" GROUP BY b.parent, b.child",
"),",
"global_lvl AS (",
" SELECT b.parent, b.child,",
" MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow,",
" json_group_array(CASE WHEN ar.allow = 0 THEN ar.source_plugin || ': ' || ar.reason END) AS deny_reasons,",
" json_group_array(CASE WHEN ar.allow = 1 THEN ar.source_plugin || ': ' || ar.reason END) AS allow_reasons",
" FROM base b",
" LEFT JOIN all_rules ar ON ar.parent IS NULL AND ar.child IS NULL",
" GROUP BY b.parent, b.child",
"),",
]
) )
# Add anonymous decision logic if needed # Add anonymous decision logic if needed
if include_is_private: if include_is_private:
def _anon_agg(select_key, where, group_by):
parts = [
f" SELECT {select_key}",
" MAX(CASE WHEN allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN allow = 1 THEN 1 ELSE 0 END) AS any_allow",
f" FROM anon_rules WHERE {where}",
]
if group_by:
parts.append(f" GROUP BY {group_by}")
return parts
query_parts.extend( query_parts.extend(
["anon_child_agg AS ("] [
+ _anon_agg( "anon_child_lvl AS (",
f"parent, child COLLATE {child_collation} AS child,", " SELECT b.parent, b.child,",
"parent IS NOT NULL AND child IS NOT NULL", " MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
f"parent, child COLLATE {child_collation}", " MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow",
) " FROM base b",
+ ["),", "anon_parent_agg AS ("] " LEFT JOIN anon_rules ar ON ar.parent = b.parent AND ar.child = b.child",
+ _anon_agg("parent,", "parent IS NOT NULL AND child IS NULL", "parent") " GROUP BY b.parent, b.child",
+ ["),", "anon_global_agg AS ("] "),",
+ _anon_agg("", "parent IS NULL AND child IS NULL", None) "anon_parent_lvl AS (",
+ ["),"] " SELECT b.parent, b.child,",
" MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow",
" FROM base b",
" LEFT JOIN anon_rules ar ON ar.parent = b.parent AND ar.child IS NULL",
" GROUP BY b.parent, b.child",
"),",
"anon_global_lvl AS (",
" SELECT b.parent, b.child,",
" MAX(CASE WHEN ar.allow = 0 THEN 1 ELSE 0 END) AS any_deny,",
" MAX(CASE WHEN ar.allow = 1 THEN 1 ELSE 0 END) AS any_allow",
" FROM base b",
" LEFT JOIN anon_rules ar ON ar.parent IS NULL AND ar.child IS NULL",
" GROUP BY b.parent, b.child",
"),",
"anon_decisions AS (",
" SELECT",
" b.parent, b.child,",
" CASE",
" WHEN acl.any_deny = 1 THEN 0",
" WHEN acl.any_allow = 1 THEN 1",
" WHEN apl.any_deny = 1 THEN 0",
" WHEN apl.any_allow = 1 THEN 1",
" WHEN agl.any_deny = 1 THEN 0",
" WHEN agl.any_allow = 1 THEN 1",
" ELSE 0",
" END AS anon_is_allowed",
" FROM base b",
" JOIN anon_child_lvl acl ON b.parent = acl.parent AND (b.child = acl.child OR (b.child IS NULL AND acl.child IS NULL))",
" JOIN anon_parent_lvl apl ON b.parent = apl.parent AND (b.child = apl.child OR (b.child IS NULL AND apl.child IS NULL))",
" JOIN anon_global_lvl agl ON b.parent = agl.parent AND (b.child = agl.child OR (b.child IS NULL AND agl.child IS NULL))",
"),",
]
) )
# Final decisions # Final decisions
@ -326,28 +342,31 @@ async def _build_single_action_sql(
"decisions AS (", "decisions AS (",
" SELECT", " SELECT",
" b.parent, b.child,", " b.parent, b.child,",
" -- Cascading permission logic: child -> parent -> global, DENY beats ALLOW at each level", " -- Cascading permission logic: child → parent → global, DENY beats ALLOW at each level",
" -- Priority order:", " -- Priority order:",
" -- 1. Child-level deny 2. Child-level allow", " -- 1. Child-level deny (most specific, blocks access)",
" -- 3. Parent-level deny 4. Parent-level allow", " -- 2. Child-level allow (most specific, grants access)",
" -- 5. Global-level deny 6. Global-level allow", " -- 3. Parent-level deny (intermediate, blocks access)",
" -- 4. Parent-level allow (intermediate, grants access)",
" -- 5. Global-level deny (least specific, blocks access)",
" -- 6. Global-level allow (least specific, grants access)",
" -- 7. Default deny (no rules match)", " -- 7. Default deny (no rules match)",
" CASE", " CASE",
" WHEN ca.any_deny = 1 THEN 0", " WHEN cl.any_deny = 1 THEN 0",
" WHEN ca.any_allow = 1 THEN 1", " WHEN cl.any_allow = 1 THEN 1",
" WHEN pa.any_deny = 1 THEN 0", " WHEN pl.any_deny = 1 THEN 0",
" WHEN pa.any_allow = 1 THEN 1", " WHEN pl.any_allow = 1 THEN 1",
" WHEN ga.any_deny = 1 THEN 0", " WHEN gl.any_deny = 1 THEN 0",
" WHEN ga.any_allow = 1 THEN 1", " WHEN gl.any_allow = 1 THEN 1",
" ELSE 0", " ELSE 0",
" END AS is_allowed,", " END AS is_allowed,",
" CASE", " CASE",
" WHEN ca.any_deny = 1 THEN ca.deny_reasons", " WHEN cl.any_deny = 1 THEN cl.deny_reasons",
" WHEN ca.any_allow = 1 THEN ca.allow_reasons", " WHEN cl.any_allow = 1 THEN cl.allow_reasons",
" WHEN pa.any_deny = 1 THEN pa.deny_reasons", " WHEN pl.any_deny = 1 THEN pl.deny_reasons",
" WHEN pa.any_allow = 1 THEN pa.allow_reasons", " WHEN pl.any_allow = 1 THEN pl.allow_reasons",
" WHEN ga.any_deny = 1 THEN ga.deny_reasons", " WHEN gl.any_deny = 1 THEN gl.deny_reasons",
" WHEN ga.any_allow = 1 THEN ga.allow_reasons", " WHEN gl.any_allow = 1 THEN gl.allow_reasons",
" ELSE '[]'", " ELSE '[]'",
" END AS reason", " END AS reason",
] ]
@ -355,34 +374,21 @@ async def _build_single_action_sql(
if include_is_private: if include_is_private:
query_parts.append( query_parts.append(
" , CASE WHEN (" " , CASE WHEN ad.anon_is_allowed = 0 THEN 1 ELSE 0 END AS is_private"
"CASE"
" WHEN aca.any_deny = 1 THEN 0"
" WHEN aca.any_allow = 1 THEN 1"
" WHEN apa.any_deny = 1 THEN 0"
" WHEN apa.any_allow = 1 THEN 1"
" WHEN aga.any_deny = 1 THEN 0"
" WHEN aga.any_allow = 1 THEN 1"
" ELSE 0 END"
") = 0 THEN 1 ELSE 0 END AS is_private"
) )
query_parts.extend( query_parts.extend(
[ [
" FROM base b", " FROM base b",
" LEFT JOIN child_agg ca ON ca.parent = b.parent AND ca.child = b.child", " JOIN child_lvl cl ON b.parent = cl.parent AND (b.child = cl.child OR (b.child IS NULL AND cl.child IS NULL))",
" LEFT JOIN parent_agg pa ON pa.parent = b.parent", " JOIN parent_lvl pl ON b.parent = pl.parent AND (b.child = pl.child OR (b.child IS NULL AND pl.child IS NULL))",
" CROSS JOIN global_agg ga", " JOIN global_lvl gl ON b.parent = gl.parent AND (b.child = gl.child OR (b.child IS NULL AND gl.child IS NULL))",
] ]
) )
if include_is_private: if include_is_private:
query_parts.extend( query_parts.append(
[ " JOIN anon_decisions ad ON b.parent = ad.parent AND (b.child = ad.child OR (b.child IS NULL AND ad.child IS NULL))"
" LEFT JOIN anon_child_agg aca ON aca.parent = b.parent AND aca.child = b.child",
" LEFT JOIN anon_parent_agg apa ON apa.parent = b.parent",
" CROSS JOIN anon_global_agg aga",
]
) )
query_parts.append(")") query_parts.append(")")
@ -392,31 +398,10 @@ async def _build_single_action_sql(
# Wrap each restriction_sql in a subquery to avoid operator precedence issues # Wrap each restriction_sql in a subquery to avoid operator precedence issues
# with UNION ALL inside the restriction SQL statements # with UNION ALL inside the restriction SQL statements
restriction_intersect = "\nINTERSECT\n".join( restriction_intersect = "\nINTERSECT\n".join(
f"SELECT parent, child COLLATE {child_collation} AS child FROM ({sql})" f"SELECT * FROM ({sql})" for sql in restriction_sqls
for sql in restriction_sqls
) )
# Decompose by NULL-pattern so the final filter can use pure-equality
# EXISTS lookups (satisfiable via automatic indexes) instead of a
# correlated OR-scan over the whole list.
query_parts.extend( query_parts.extend(
[ [",", "restriction_list AS (", f" {restriction_intersect}", ")"]
",",
"restriction_list AS (",
f" {restriction_intersect}",
"),",
"restriction_exact AS (",
" SELECT parent, child FROM restriction_list WHERE parent IS NOT NULL AND child IS NOT NULL",
"),",
"restriction_parent_any AS (",
" SELECT DISTINCT parent FROM restriction_list WHERE parent IS NOT NULL AND child IS NULL",
"),",
"restriction_child_any AS (",
" SELECT DISTINCT child FROM restriction_list WHERE parent IS NULL AND child IS NOT NULL",
"),",
"restriction_all AS (",
" SELECT 1 AS matched FROM restriction_list WHERE parent IS NULL AND child IS NULL LIMIT 1",
")",
]
) )
# Final SELECT # Final SELECT
@ -431,11 +416,10 @@ async def _build_single_action_sql(
# Add restriction filter if there are restrictions # Add restriction filter if there are restrictions
if restriction_sqls: if restriction_sqls:
query_parts.append(""" query_parts.append("""
AND ( AND EXISTS (
EXISTS (SELECT 1 FROM restriction_all) SELECT 1 FROM restriction_list r
OR EXISTS (SELECT 1 FROM restriction_parent_any r WHERE r.parent = decisions.parent) WHERE (r.parent = decisions.parent OR r.parent IS NULL)
OR EXISTS (SELECT 1 FROM restriction_child_any r WHERE r.child = decisions.child) AND (r.child = decisions.child OR r.child IS NULL)
OR EXISTS (SELECT 1 FROM restriction_exact r WHERE r.parent = decisions.parent AND r.child = decisions.child)
)""") )""")
# Add parent filter if specified # Add parent filter if specified
@ -491,7 +475,6 @@ async def build_permission_rules_sql(
union_parts = [] union_parts = []
all_params = {} all_params = {}
restriction_sqls = [] restriction_sqls = []
child_collation = _child_collation(action_obj)
for permission_sql in permission_sqls: for permission_sql in permission_sqls:
all_params.update(permission_sql.params or {}) all_params.update(permission_sql.params or {})
@ -505,7 +488,7 @@ async def build_permission_rules_sql(
continue continue
union_parts.append(f""" union_parts.append(f"""
SELECT parent, child COLLATE {child_collation} AS child, allow, reason, '{permission_sql.source}' AS source_plugin FROM ( SELECT parent, child, allow, reason, '{permission_sql.source}' AS source_plugin FROM (
{permission_sql.sql} {permission_sql.sql}
) )
""".strip()) """.strip())
@ -576,7 +559,6 @@ async def check_permissions_for_actions(
verdicts = {} verdicts = {}
for i, (action, permission_sqls) in enumerate(zip(unique_actions, gathered)): for i, (action, permission_sqls) in enumerate(zip(unique_actions, gathered)):
child_collation = _child_collation(datasette.actions[action])
prefix = f"a{i}_" prefix = f"a{i}_"
rule_parts = [] rule_parts = []
restriction_parts = [] restriction_parts = []
@ -602,7 +584,7 @@ async def check_permissions_for_actions(
if sql is None: if sql is None:
continue continue
rule_parts.append( rule_parts.append(
f"SELECT parent, child COLLATE {child_collation} AS child, allow, reason, '{permission_sql.source}' AS source_plugin FROM (\n{sql}\n)" f"SELECT parent, child, allow, reason, '{permission_sql.source}' AS source_plugin FROM (\n{sql}\n)"
) )
if not rule_parts: if not rule_parts:
@ -636,8 +618,7 @@ async def check_permissions_for_actions(
if restriction_parts: if restriction_parts:
# Database-level restrictions (parent, NULL) match all children # Database-level restrictions (parent, NULL) match all children
restriction_intersect = "\nINTERSECT\n".join( restriction_intersect = "\nINTERSECT\n".join(
f"SELECT parent, child COLLATE {child_collation} AS child FROM ({sql})" f"SELECT * FROM ({sql})" for sql in restriction_parts
for sql in restriction_parts
) )
ctes.append(f"a{i}_restriction AS (\n{restriction_intersect}\n)") ctes.append(f"a{i}_restriction AS (\n{restriction_intersect}\n)")
verdict_sql = f"""({verdict_sql}) AND EXISTS ( verdict_sql = f"""({verdict_sql}) AND EXISTS (
@ -692,240 +673,3 @@ async def check_permission_for_resource(
child=child, child=child,
) )
return results[action] return results[action]
async def explain_permission_for_resource(
*,
datasette: "Datasette",
actor: dict | None,
action: str,
parent: str | None,
child: str | None,
) -> dict:
"""Explain a permission decision for one action and resource.
This is intended for Datasette's permission debugging tools. It uses the
same ``permission_resources_sql`` hook results and the same resolution
rules as :func:`check_permissions_for_actions`, but also returns the
matching rules, actor restriction results and ``also_requires`` chain.
The returned dictionary is part of Datasette's unstable debugging API.
"""
action_obj = datasette.actions.get(action)
if action_obj is None:
raise ValueError(f"Unknown action: {action}")
explanation = await _explain_single_action(
datasette=datasette,
actor=actor,
action=action,
parent=parent,
child=child,
)
required_actions = []
if action_obj.also_requires:
required = await explain_permission_for_resource(
datasette=datasette,
actor=actor,
action=action_obj.also_requires,
parent=parent,
child=child,
)
required_actions.append(required)
explanation["required_actions"] = required_actions
explanation["allowed"] = bool(
explanation["rule_allowed"]
and explanation["restriction_allowed"]
and all(required["allowed"] for required in required_actions)
)
explanation["summary"] = _permission_explanation_summary(explanation)
return explanation
async def _explain_single_action(
*,
datasette: "Datasette",
actor: dict | None,
action: str,
parent: str | None,
child: str | None,
) -> dict:
"""Return matching rules and restrictions for a single action."""
from datasette.utils.permissions import SKIP_PERMISSION_CHECKS
permission_sqls = await gather_permission_sql_from_hooks(
datasette=datasette,
actor=actor,
action=action,
)
if permission_sqls is SKIP_PERMISSION_CHECKS:
return {
"action": action,
"rule_allowed": True,
"restriction_allowed": True,
"winning_scope": "global",
"matched_rules": [
{
"scope": "global",
"effect": "allow",
"source": "skip_permission_checks",
"reason": "Permission checks were explicitly skipped",
"decisive": True,
"ignored_because": None,
}
],
"restrictions": [],
}
db = datasette.get_internal_database()
matched_rules = []
restrictions = []
child_collation = _child_collation(datasette.actions[action])
for permission_sql in permission_sqls:
params = dict(permission_sql.params or {})
parent_param = _unused_parameter_name(params, "_explain_parent")
params[parent_param] = parent
child_param = _unused_parameter_name(params, "_explain_child")
params[child_param] = child
if permission_sql.sql:
rows = await db.execute(
f"""
SELECT parent, child, allow, reason
FROM ({permission_sql.sql}) AS permission_rules
WHERE (parent IS NULL OR parent = :{parent_param})
AND (child IS NULL OR child COLLATE {child_collation} = :{child_param})
""",
params,
)
for row in rows:
specificity = (
2
if row["child"] is not None
else 1 if row["parent"] is not None else 0
)
matched_rules.append(
{
"scope": ("resource", "parent", "global")[2 - specificity],
"effect": "allow" if row["allow"] else "deny",
"source": permission_sql.source,
"reason": row["reason"],
"_specificity": specificity,
}
)
if permission_sql.restriction_sql:
restriction_row = (
await db.execute(
f"""
SELECT EXISTS(
SELECT 1 FROM ({permission_sql.restriction_sql}) AS restriction_rules
WHERE (parent IS NULL OR parent = :{parent_param})
AND (child IS NULL OR child COLLATE {child_collation} = :{child_param})
) AS resource_is_in_allowlist
""",
params,
)
).first()
restriction_allowed = bool(restriction_row[0])
restrictions.append(
{
"source": permission_sql.source,
"allowed": restriction_allowed,
"reason": params.get("deny")
or (
"Resource is included in this restriction allowlist"
if restriction_allowed
else "Resource is not included in this restriction allowlist"
),
}
)
matched_rules.sort(
key=lambda rule: (
-rule["_specificity"],
0 if rule["effect"] == "deny" else 1,
rule["source"] or "",
rule["reason"] or "",
)
)
if matched_rules:
winning_specificity = matched_rules[0]["_specificity"]
winning_rules = [
rule
for rule in matched_rules
if rule["_specificity"] == winning_specificity
]
rule_allowed = not any(rule["effect"] == "deny" for rule in winning_rules)
winning_scope = winning_rules[0]["scope"]
else:
winning_specificity = None
rule_allowed = False
winning_scope = None
for rule in matched_rules:
specificity = rule.pop("_specificity")
if specificity != winning_specificity:
rule["decisive"] = False
rule["ignored_because"] = "A more specific rule matched"
elif not rule_allowed and rule["effect"] == "allow":
rule["decisive"] = False
rule["ignored_because"] = "A deny rule matched at the same scope"
else:
rule["decisive"] = True
rule["ignored_because"] = None
return {
"action": action,
"rule_allowed": rule_allowed,
"restriction_allowed": all(
restriction["allowed"] for restriction in restrictions
),
"winning_scope": winning_scope,
"matched_rules": matched_rules,
"restrictions": restrictions,
}
def _unused_parameter_name(params: dict, preferred: str) -> str:
"""Return a SQL parameter name that is not already in ``params``."""
candidate = preferred
suffix = 2
while candidate in params:
candidate = f"{preferred}_{suffix}"
suffix += 1
return candidate
def _permission_explanation_summary(explanation: dict) -> str:
denied_requirement = next(
(
required
for required in explanation["required_actions"]
if not required["allowed"]
),
None,
)
if denied_requirement:
return (
f"Denied because {explanation['action']} also requires "
f"{denied_requirement['action']}, which was denied."
)
if not explanation["matched_rules"]:
return "Denied because no permission rule matched this actor and resource."
if not explanation["rule_allowed"]:
return (
f"Denied by a {explanation['winning_scope']}-level rule. "
"Deny rules take precedence over allow rules at the same scope."
)
if not explanation["restriction_allowed"]:
return (
"Denied because the resource is not included in the actor's restrictions."
)
return f"Allowed by the matching {explanation['winning_scope']}-level rule."

View file

@ -1,30 +1,28 @@
import asyncio
import json import json
import re from typing import Optional
from http.cookies import Morsel, SimpleCookie from datasette.utils import MultiParams, calculate_etag
from mimetypes import guess_type
from pathlib import Path
from urllib.parse import parse_qs, parse_qsl, urlunparse
import aiofiles
import aiofiles.os
from datasette.utils import MultiParams, calculate_etag, error_body, sha256_file
from datasette.utils.multipart import ( from datasette.utils.multipart import (
DEFAULT_MAX_FIELD_SIZE, parse_form_data,
DEFAULT_MAX_FIELDS, MultipartParseError,
FormData,
DEFAULT_MAX_FILE_SIZE, DEFAULT_MAX_FILE_SIZE,
DEFAULT_MAX_REQUEST_SIZE,
DEFAULT_MAX_FIELDS,
DEFAULT_MAX_FILES, DEFAULT_MAX_FILES,
DEFAULT_MAX_PARTS,
DEFAULT_MAX_FIELD_SIZE,
DEFAULT_MAX_MEMORY_FILE_SIZE, DEFAULT_MAX_MEMORY_FILE_SIZE,
DEFAULT_MAX_PART_HEADER_BYTES, DEFAULT_MAX_PART_HEADER_BYTES,
DEFAULT_MAX_PART_HEADER_LINES, DEFAULT_MAX_PART_HEADER_LINES,
DEFAULT_MAX_PARTS,
DEFAULT_MAX_REQUEST_SIZE,
DEFAULT_MIN_FREE_DISK_BYTES, DEFAULT_MIN_FREE_DISK_BYTES,
FormData,
MultipartParseError,
parse_form_data,
) )
from mimetypes import guess_type
from urllib.parse import parse_qs, urlunparse, parse_qsl
from pathlib import Path
from http.cookies import SimpleCookie, Morsel
import aiofiles
import aiofiles.os
import re
# Workaround for adding samesite support to pre 3.8 python # Workaround for adding samesite support to pre 3.8 python
Morsel._reserved["samesite"] = "SameSite" Morsel._reserved["samesite"] = "SameSite"
@ -69,41 +67,16 @@ class BadRequest(Base400):
status = 400 status = 400
class PayloadTooLarge(Base400):
status = 413
SAMESITE_VALUES = ("strict", "lax", "none") SAMESITE_VALUES = ("strict", "lax", "none")
# Bodies read fully into memory (post_body/post_vars/json) are capped at this
# size unless the max_post_body_bytes setting says otherwise. Kept deliberately
# far below multipart's DEFAULT_MAX_REQUEST_SIZE: that parser streams to disk,
# while these bodies are held in RAM and json.loads() can multiply their
# footprint several times over.
DEFAULT_MAX_POST_BODY_BYTES = 2 * 1024 * 1024 # 2MB
class _RequestHeaders(dict):
"""Incoming headers with lowercase keys and case-insensitive lookups."""
def __getitem__(self, key):
return super().__getitem__(key.lower())
def get(self, key, default=None):
return super().get(key.lower(), default)
def __contains__(self, key):
return super().__contains__(key.lower())
class Request: class Request:
def __init__(self, scope, receive, max_post_body_bytes=DEFAULT_MAX_POST_BODY_BYTES): def __init__(self, scope, receive):
self.scope = scope self.scope = scope
self.receive = receive self.receive = receive
self.max_post_body_bytes = max_post_body_bytes
def __repr__(self): def __repr__(self):
return f'<asgi.Request method="{self.method}" url="{self.url}">' return '<asgi.Request method="{}" url="{}">'.format(self.method, self.url)
@property @property
def method(self): def method(self):
@ -125,10 +98,10 @@ class Request:
@property @property
def headers(self): def headers(self):
return _RequestHeaders( return {
(k.decode("latin-1").lower(), v.decode("latin-1")) k.decode("latin-1").lower(): v.decode("latin-1")
for k, v in self.scope.get("headers") or [] for k, v in self.scope.get("headers") or []
) }
@property @property
def host(self): def host(self):
@ -168,43 +141,15 @@ class Request:
def actor(self): def actor(self):
return self.scope.get("actor", None) return self.scope.get("actor", None)
async def post_body(self, max_bytes=None): async def post_body(self):
""" body = b""
Read the request body fully into memory.
The body is capped at max_bytes - or self.max_post_body_bytes
(default 2MB, set from the max_post_body_bytes setting for requests
created by Datasette) if max_bytes is not provided. Pass max_bytes=0
to disable the limit. Raises PayloadTooLarge (HTTP 413) if exceeded -
oversized bodies are rejected as soon as the limit is passed, without
buffering the rest.
"""
if max_bytes is None:
max_bytes = self.max_post_body_bytes
too_large = PayloadTooLarge(
f"Request body exceeded maximum size of {max_bytes} bytes"
)
if max_bytes:
# Reject early if the client declares an oversized body
try:
if int(self.headers.get("content-length", "")) > max_bytes:
raise too_large
except ValueError:
# Missing or malformed - the streaming check below still applies
pass
chunks = []
received = 0
more_body = True more_body = True
while more_body: while more_body:
message = await self.receive() message = await self.receive()
assert message["type"] == "http.request", message assert message["type"] == "http.request", message
chunk = message.get("body", b"") body += message.get("body", b"")
received += len(chunk)
if max_bytes and received > max_bytes:
raise too_large
chunks.append(chunk)
more_body = message.get("more_body", False) more_body = message.get("more_body", False)
return b"".join(chunks) return body
async def post_vars(self): async def post_vars(self):
body = await self.post_body() body = await self.post_body()
@ -221,7 +166,7 @@ class Request:
max_request_size: int = DEFAULT_MAX_REQUEST_SIZE, max_request_size: int = DEFAULT_MAX_REQUEST_SIZE,
max_fields: int = DEFAULT_MAX_FIELDS, max_fields: int = DEFAULT_MAX_FIELDS,
max_files: int = DEFAULT_MAX_FILES, max_files: int = DEFAULT_MAX_FILES,
max_parts: int | None = DEFAULT_MAX_PARTS, max_parts: Optional[int] = DEFAULT_MAX_PARTS,
max_field_size: int = DEFAULT_MAX_FIELD_SIZE, max_field_size: int = DEFAULT_MAX_FIELD_SIZE,
max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE, max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE,
max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES, max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES,
@ -314,24 +259,12 @@ class AsgiLifespan:
while True: while True:
message = await receive() message = await receive()
if message["type"] == "lifespan.startup": if message["type"] == "lifespan.startup":
try:
for fn in self.on_startup: for fn in self.on_startup:
await fn() await fn()
except Exception as e: # noqa: BLE001
await send(
{"type": "lifespan.startup.failed", "message": str(e)}
)
return
await send({"type": "lifespan.startup.complete"}) await send({"type": "lifespan.startup.complete"})
elif message["type"] == "lifespan.shutdown": elif message["type"] == "lifespan.shutdown":
try:
for fn in self.on_shutdown: for fn in self.on_shutdown:
await fn() await fn()
except Exception as e: # noqa: BLE001
await send(
{"type": "lifespan.shutdown.failed", "message": str(e)}
)
return
await send({"type": "lifespan.shutdown.complete"}) await send({"type": "lifespan.shutdown.complete"})
return return
else: else:
@ -464,9 +397,6 @@ async def asgi_send_file(
) )
HASHED_STATIC_CACHE_CONTROL = "max-age=31536000, immutable, public"
def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None): def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
root_path = Path(root_path) root_path = Path(root_path)
static_headers = {} static_headers = {}
@ -493,17 +423,11 @@ def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
return return
try: try:
# Calculate ETag for filepath # Calculate ETag for filepath
hash_value = request.args.get("_hash")
if (
hash_value
and hash_value == sha256_file(full_path, chunk_size=chunk_size)[:12]
):
headers["Cache-Control"] = HASHED_STATIC_CACHE_CONTROL
etag = await calculate_etag(full_path, chunk_size=chunk_size) etag = await calculate_etag(full_path, chunk_size=chunk_size)
headers["ETag"] = etag headers["ETag"] = etag
if_none_match = request.headers.get("if-none-match") if_none_match = request.headers.get("if-none-match")
if if_none_match and if_none_match == etag: if if_none_match and if_none_match == etag:
return await asgi_send(send, "", 304, headers=headers) return await asgi_send(send, "", 304)
await asgi_send_file( await asgi_send_file(
send, full_path, chunk_size=chunk_size, headers=headers send, full_path, chunk_size=chunk_size, headers=headers
) )
@ -511,8 +435,6 @@ def asgi_static(root_path, chunk_size=4096, headers=None, content_type=None):
await asgi_send_html(send, "404: File not found", 404) await asgi_send_html(send, "404: File not found", 404)
return return
# Only the actual static-file handler can bypass dynamic response privacy.
inner_static._datasette_static = True
return inner_static return inner_static
@ -558,9 +480,9 @@ class Response:
httponly=False, httponly=False,
samesite="lax", samesite="lax",
): ):
assert ( assert samesite in SAMESITE_VALUES, "samesite should be one of {}".format(
samesite in SAMESITE_VALUES SAMESITE_VALUES
), f"samesite should be one of {SAMESITE_VALUES}" )
cookie = SimpleCookie() cookie = SimpleCookie()
cookie[key] = value cookie[key] = value
for prop_name, prop_value in ( for prop_name, prop_value in (
@ -604,18 +526,6 @@ class Response:
content_type="application/json; charset=utf-8", content_type="application/json; charset=utf-8",
) )
@classmethod
def error(cls, messages, status=400, headers=None):
"""
A JSON error response using Datasette's standard error format.
messages can be a single string or a list of strings. For errors
that should content-negotiate between JSON and HTML, raise
Forbidden, NotFound, BadRequest or DatasetteError instead and let
Datasette's error handling hooks build the response.
"""
return cls.json(error_body(messages, status), status=status, headers=headers)
@classmethod @classmethod
def redirect(cls, path, status=302, headers=None): def redirect(cls, path, status=302, headers=None):
headers = headers or {} headers = headers or {}
@ -652,23 +562,10 @@ class AsgiRunOnFirstRequest:
self.asgi = asgi self.asgi = asgi
self.on_startup = on_startup self.on_startup = on_startup
self._started = False self._started = False
# Guards against concurrent early requests interleaving with startup:
# without this, several requests could all observe `_started is
# False` and proceed before any of them finish running the hooks.
self._lock = asyncio.Lock()
async def __call__(self, scope, receive, send): async def __call__(self, scope, receive, send):
# Leave "lifespan" scope events alone - this shim only exists as a
# fallback for hosts that never send them. It wraps AsgiLifespan, so
# if it ran on_startup here too, a startup exception would escape
# before AsgiLifespan's own try/except got a chance to turn it into
# a lifespan.startup.failed message.
if scope["type"] != "lifespan" and not self._started:
async with self._lock:
# Re-check: another request may have finished startup while
# we were waiting for the lock.
if not self._started: if not self._started:
self._started = True
for hook in self.on_startup: for hook in self.on_startup:
await hook() await hook()
self._started = True
return await self.asgi(scope, receive, send) return await self.asgi(scope, receive, send)

View file

@ -13,7 +13,7 @@ Originally shared here: https://www.djangosnippets.org/snippets/1431/
""" """
class BaseConverter: class BaseConverter(object):
decimal_digits = "0123456789" decimal_digits = "0123456789"
def __init__(self, digits): def __init__(self, digits):

View file

@ -1,6 +1,6 @@
import inspect import inspect
import types import types
from typing import Any, NamedTuple from typing import NamedTuple, Any
class CallableStatus(NamedTuple): class CallableStatus(NamedTuple):
@ -19,7 +19,7 @@ def check_callable(obj: Any) -> CallableStatus:
if isinstance(obj, types.FunctionType): if isinstance(obj, types.FunctionType):
return CallableStatus(True, inspect.iscoroutinefunction(obj)) return CallableStatus(True, inspect.iscoroutinefunction(obj))
if callable(obj): if hasattr(obj, "__call__"):
return CallableStatus(True, inspect.iscoroutinefunction(obj.__call__)) return CallableStatus(True, inspect.iscoroutinefunction(obj.__call__))
assert False, f"obj {obj!r} is somehow callable with no __call__ method" assert False, "obj {} is somehow callable with no __call__ method".format(repr(obj))

View file

@ -1,30 +1,9 @@
import textwrap import textwrap
from datasette.utils import table_column_details
from sqlite_utils import Database as SQLiteUtilsDatabase
from sqlite_utils import Migrations
from datasette.utils import escape_sqlite, table_column_details async def init_internal_db(db):
create_tables_sql = textwrap.dedent("""
INTERNAL_DB_SCHEMA_TABLES = {
"catalog_databases",
"catalog_tables",
"catalog_views",
"catalog_columns",
"catalog_indexes",
"catalog_foreign_keys",
"metadata_instance",
"metadata_databases",
"metadata_resources",
"metadata_columns",
"column_types",
"queries",
}
INTERNAL_DB_SCHEMA_INDEXES = {
"queries_owner_idx",
}
INTERNAL_DB_SCHEMA_SQL = textwrap.dedent("""
CREATE TABLE IF NOT EXISTS catalog_databases ( CREATE TABLE IF NOT EXISTS catalog_databases (
database_name TEXT PRIMARY KEY, database_name TEXT PRIMARY KEY,
path TEXT, path TEXT,
@ -88,7 +67,13 @@ INTERNAL_DB_SCHEMA_SQL = textwrap.dedent("""
FOREIGN KEY (database_name) REFERENCES catalog_databases(database_name), FOREIGN KEY (database_name) REFERENCES catalog_databases(database_name),
FOREIGN KEY (database_name, table_name) REFERENCES catalog_tables(database_name, table_name) FOREIGN KEY (database_name, table_name) REFERENCES catalog_tables(database_name, table_name)
); );
""").strip()
await db.execute_write_script(create_tables_sql)
await initialize_metadata_tables(db)
async def initialize_metadata_tables(db):
await db.execute_write_script(textwrap.dedent("""
CREATE TABLE IF NOT EXISTS metadata_instance ( CREATE TABLE IF NOT EXISTS metadata_instance (
key text, key text,
value text, value text,
@ -149,40 +134,32 @@ INTERNAL_DB_SCHEMA_SQL = textwrap.dedent("""
CREATE INDEX IF NOT EXISTS queries_owner_idx CREATE INDEX IF NOT EXISTS queries_owner_idx
ON queries(owner_id); ON queries(owner_id);
""").strip() """))
internal_migrations = Migrations("datasette_internal") async def populate_schema_tables(internal_db, db):
def _internal_schema_exists(db):
table_names = set(db.table_names())
if not INTERNAL_DB_SCHEMA_TABLES.issubset(table_names):
return False
index_names = {
row[0]
for row in db.execute("select name from sqlite_master where type = 'index'")
}
return INTERNAL_DB_SCHEMA_INDEXES.issubset(index_names)
@internal_migrations(name="0001_initial")
def initial_internal_schema(db):
if _internal_schema_exists(db):
return
db.executescript(INTERNAL_DB_SCHEMA_SQL)
async def init_internal_db(db):
def apply_migrations(conn):
internal_migrations.apply(SQLiteUtilsDatabase(conn, execute_plugins=False))
await db.execute_write_fn(apply_migrations, transaction=False)
async def populate_schema_tables(internal_db, db, schema_version):
database_name = db.name database_name = db.name
def delete_everything(conn):
conn.execute(
"DELETE FROM catalog_tables WHERE database_name = ?", [database_name]
)
conn.execute(
"DELETE FROM catalog_views WHERE database_name = ?", [database_name]
)
conn.execute(
"DELETE FROM catalog_columns WHERE database_name = ?", [database_name]
)
conn.execute(
"DELETE FROM catalog_foreign_keys WHERE database_name = ?",
[database_name],
)
conn.execute(
"DELETE FROM catalog_indexes WHERE database_name = ?", [database_name]
)
await internal_db.execute_write_fn(delete_everything)
tables = (await db.execute("select * from sqlite_master WHERE type = 'table'")).rows tables = (await db.execute("select * from sqlite_master WHERE type = 'table'")).rows
views = (await db.execute("select * from sqlite_master WHERE type = 'view'")).rows views = (await db.execute("select * from sqlite_master WHERE type = 'view'")).rows
@ -207,30 +184,25 @@ async def populate_schema_tables(internal_db, db, schema_version):
columns = table_column_details(conn, table_name) columns = table_column_details(conn, table_name)
columns_to_insert.extend( columns_to_insert.extend(
{ {
"database_name": database_name, **{"database_name": database_name, "table_name": table_name},
"table_name": table_name,
**column._asdict(), **column._asdict(),
} }
for column in columns for column in columns
) )
foreign_keys = conn.execute( foreign_keys = conn.execute(
f"PRAGMA foreign_key_list({escape_sqlite(table_name)})" f"PRAGMA foreign_key_list([{table_name}])"
).fetchall() ).fetchall()
foreign_keys_to_insert.extend( foreign_keys_to_insert.extend(
{ {
"database_name": database_name, **{"database_name": database_name, "table_name": table_name},
"table_name": table_name,
**dict(foreign_key), **dict(foreign_key),
} }
for foreign_key in foreign_keys for foreign_key in foreign_keys
) )
indexes = conn.execute( indexes = conn.execute(f"PRAGMA index_list([{table_name}])").fetchall()
f"PRAGMA index_list({escape_sqlite(table_name)})"
).fetchall()
indexes_to_insert.extend( indexes_to_insert.extend(
{ {
"database_name": database_name, **{"database_name": database_name, "table_name": table_name},
"table_name": table_name,
**dict(index), **dict(index),
} }
for index in indexes for index in indexes
@ -251,48 +223,21 @@ async def populate_schema_tables(internal_db, db, schema_version):
indexes_to_insert, indexes_to_insert,
) = await db.execute_fn(collect_info) ) = await db.execute_fn(collect_info)
def replace_catalog(conn): await internal_db.execute_write_many(
# Delete child rows before their catalog_tables parents so this also
# works if a prepare_connection plugin enables foreign key enforcement.
for table in (
"catalog_columns",
"catalog_foreign_keys",
"catalog_indexes",
"catalog_views",
"catalog_tables",
):
conn.execute(
f"DELETE FROM {table} WHERE database_name = ?",
[database_name],
)
conn.execute(
"""
INSERT OR REPLACE INTO catalog_databases (
database_name, path, is_memory, schema_version
) VALUES (?, ?, ?, ?)
""",
[
database_name,
str(db.path) if db.path is not None else None,
db.is_memory,
schema_version,
],
)
conn.executemany(
""" """
INSERT INTO catalog_tables (database_name, table_name, rootpage, sql) INSERT INTO catalog_tables (database_name, table_name, rootpage, sql)
values (?, ?, ?, ?) values (?, ?, ?, ?)
""", """,
tables_to_insert, tables_to_insert,
) )
conn.executemany( await internal_db.execute_write_many(
""" """
INSERT INTO catalog_views (database_name, view_name, rootpage, sql) INSERT INTO catalog_views (database_name, view_name, rootpage, sql)
values (?, ?, ?, ?) values (?, ?, ?, ?)
""", """,
views_to_insert, views_to_insert,
) )
conn.executemany( await internal_db.execute_write_many(
""" """
INSERT INTO catalog_columns ( INSERT INTO catalog_columns (
database_name, table_name, cid, name, type, "notnull", default_value, is_pk, hidden database_name, table_name, cid, name, type, "notnull", default_value, is_pk, hidden
@ -302,7 +247,7 @@ async def populate_schema_tables(internal_db, db, schema_version):
""", """,
columns_to_insert, columns_to_insert,
) )
conn.executemany( await internal_db.execute_write_many(
""" """
INSERT INTO catalog_foreign_keys ( INSERT INTO catalog_foreign_keys (
database_name, table_name, "id", seq, "table", "from", "to", on_update, on_delete, match database_name, table_name, "id", seq, "table", "from", "to", on_update, on_delete, match
@ -312,7 +257,7 @@ async def populate_schema_tables(internal_db, db, schema_version):
""", """,
foreign_keys_to_insert, foreign_keys_to_insert,
) )
conn.executemany( await internal_db.execute_write_many(
""" """
INSERT INTO catalog_indexes ( INSERT INTO catalog_indexes (
database_name, table_name, seq, name, "unique", origin, partial database_name, table_name, seq, name, "unique", origin, partial
@ -322,5 +267,3 @@ async def populate_schema_tables(internal_db, db, schema_version):
""", """,
indexes_to_insert, indexes_to_insert,
) )
await internal_db.execute_write_fn(replace_catalog)

View file

@ -11,10 +11,15 @@ Supports:
import asyncio import asyncio
import shutil import shutil
import tempfile import tempfile
from collections.abc import Callable
from dataclasses import dataclass, field from dataclasses import dataclass, field
from typing import ( from typing import (
Any, Any,
Callable,
Dict,
List,
Optional,
Tuple,
Union,
) )
from urllib.parse import parse_qsl from urllib.parse import parse_qsl
@ -24,7 +29,7 @@ DEFAULT_MAX_REQUEST_SIZE = 100 * 1024 * 1024 # 100MB
DEFAULT_MAX_FIELDS = 1000 DEFAULT_MAX_FIELDS = 1000
DEFAULT_MAX_FILES = 100 DEFAULT_MAX_FILES = 100
# If max_parts is not specified, it defaults to max_fields + max_files # If max_parts is not specified, it defaults to max_fields + max_files
DEFAULT_MAX_PARTS: int | None = None DEFAULT_MAX_PARTS: Optional[int] = None
DEFAULT_MAX_FIELD_SIZE = 100 * 1024 # 100KB DEFAULT_MAX_FIELD_SIZE = 100 * 1024 # 100KB
DEFAULT_MAX_MEMORY_FILE_SIZE = 1024 * 1024 # 1MB DEFAULT_MAX_MEMORY_FILE_SIZE = 1024 * 1024 # 1MB
DEFAULT_MAX_PART_HEADER_BYTES = 16 * 1024 # 16KB DEFAULT_MAX_PART_HEADER_BYTES = 16 * 1024 # 16KB
@ -35,6 +40,8 @@ DEFAULT_MIN_FREE_DISK_BYTES = 50 * 1024 * 1024 # 50MB
class MultipartParseError(Exception): class MultipartParseError(Exception):
"""Raised when multipart parsing fails.""" """Raised when multipart parsing fails."""
pass
@dataclass @dataclass
class UploadedFile: class UploadedFile:
@ -50,7 +57,7 @@ class UploadedFile:
name: str name: str
filename: str filename: str
content_type: str | None content_type: Optional[str]
size: int size: int
_file: tempfile.SpooledTemporaryFile = field(repr=False) _file: tempfile.SpooledTemporaryFile = field(repr=False)
@ -79,8 +86,7 @@ class UploadedFile:
def __del__(self): def __del__(self):
try: try:
self._file.close() self._file.close()
except Exception: # noqa: BLE001, S110 except Exception:
# __del__ must never raise
pass pass
@ -92,27 +98,27 @@ class FormData:
""" """
def __init__(self): def __init__(self):
self._data: list[tuple[str, str | UploadedFile]] = [] self._data: List[Tuple[str, Union[str, UploadedFile]]] = []
def append(self, key: str, value: str | UploadedFile) -> None: def append(self, key: str, value: Union[str, UploadedFile]) -> None:
"""Add a key-value pair.""" """Add a key-value pair."""
self._data.append((key, value)) self._data.append((key, value))
def __getitem__(self, key: str) -> str | UploadedFile: def __getitem__(self, key: str) -> Union[str, UploadedFile]:
"""Get the first value for a key.""" """Get the first value for a key."""
for k, v in self._data: for k, v in self._data:
if k == key: if k == key:
return v return v
raise KeyError(key) raise KeyError(key)
def get(self, key: str, default: Any = None) -> str | UploadedFile | None: def get(self, key: str, default: Any = None) -> Optional[Union[str, UploadedFile]]:
"""Get the first value for a key, or default if not found.""" """Get the first value for a key, or default if not found."""
try: try:
return self[key] return self[key]
except KeyError: except KeyError:
return default return default
def getlist(self, key: str) -> list[str | UploadedFile]: def getlist(self, key: str) -> List[Union[str, UploadedFile]]:
"""Get all values for a key.""" """Get all values for a key."""
return [v for k, v in self._data if k == key] return [v for k, v in self._data if k == key]
@ -136,15 +142,15 @@ class FormData:
"""Return unique keys.""" """Return unique keys."""
return list(self) return list(self)
def items(self) -> list[tuple[str, str | UploadedFile]]: def items(self) -> List[Tuple[str, Union[str, UploadedFile]]]:
"""Return all key-value pairs.""" """Return all key-value pairs."""
return list(self._data) return list(self._data)
def values(self) -> list[str | UploadedFile]: def values(self) -> List[Union[str, UploadedFile]]:
"""Return all values.""" """Return all values."""
return [v for _, v in self._data] return [v for _, v in self._data]
def _uploaded_files(self) -> list[UploadedFile]: def _uploaded_files(self) -> List[UploadedFile]:
"""Return UploadedFile instances contained in this form.""" """Return UploadedFile instances contained in this form."""
return [v for _, v in self._data if isinstance(v, UploadedFile)] return [v for _, v in self._data if isinstance(v, UploadedFile)]
@ -157,7 +163,7 @@ class FormData:
for uploaded in self._uploaded_files(): for uploaded in self._uploaded_files():
try: try:
uploaded.close_sync() uploaded.close_sync()
except Exception: # noqa: BLE001, S110 except Exception:
# Best-effort cleanup; ignore close errors # Best-effort cleanup; ignore close errors
pass pass
@ -166,7 +172,7 @@ class FormData:
for uploaded in self._uploaded_files(): for uploaded in self._uploaded_files():
try: try:
await uploaded.close() await uploaded.close()
except Exception: # noqa: BLE001, S110 except Exception:
# Best-effort cleanup; ignore close errors # Best-effort cleanup; ignore close errors
pass pass
@ -183,13 +189,13 @@ class FormData:
await self.aclose() await self.aclose()
def parse_content_disposition(header: str) -> dict[str, str | None]: def parse_content_disposition(header: str) -> Dict[str, Optional[str]]:
""" """
Parse Content-Disposition header value. Parse Content-Disposition header value.
Returns dict with 'name', 'filename' keys (filename may be None). Returns dict with 'name', 'filename' keys (filename may be None).
""" """
result: dict[str, str | None] = {"name": None, "filename": None} result: Dict[str, Optional[str]] = {"name": None, "filename": None}
# Split on semicolons, handling quoted strings # Split on semicolons, handling quoted strings
parts = [] parts = []
@ -232,8 +238,7 @@ def parse_content_disposition(header: str) -> dict[str, str | None]:
from urllib.parse import unquote from urllib.parse import unquote
result["filename"] = unquote(encoded, encoding="utf-8") result["filename"] = unquote(encoded, encoding="utf-8")
except Exception: # noqa: BLE001, S110 except Exception:
# Malformed RFC 5987 filename* - fall back to the plain filename
pass pass
continue continue
@ -245,8 +250,9 @@ def parse_content_disposition(header: str) -> dict[str, str | None]:
if key == "name": if key == "name":
result["name"] = value result["name"] = value
# Only set filename if filename* hasn't already set it elif key == "filename":
elif key == "filename" and result["filename"] is None: # Only set if filename* hasn't already set it
if result["filename"] is None:
# Strip path components (security) # Strip path components (security)
# Handle both Unix and Windows paths # Handle both Unix and Windows paths
value = value.replace("\\", "/") value = value.replace("\\", "/")
@ -257,7 +263,7 @@ def parse_content_disposition(header: str) -> dict[str, str | None]:
return result return result
def parse_content_type(header: str) -> tuple[str, dict[str, str]]: def parse_content_type(header: str) -> Tuple[str, Dict[str, str]]:
""" """
Parse Content-Type header value. Parse Content-Type header value.
@ -301,7 +307,7 @@ class MultipartParser:
max_request_size: int = DEFAULT_MAX_REQUEST_SIZE, max_request_size: int = DEFAULT_MAX_REQUEST_SIZE,
max_fields: int = DEFAULT_MAX_FIELDS, max_fields: int = DEFAULT_MAX_FIELDS,
max_files: int = DEFAULT_MAX_FILES, max_files: int = DEFAULT_MAX_FILES,
max_parts: int | None = DEFAULT_MAX_PARTS, max_parts: Optional[int] = DEFAULT_MAX_PARTS,
max_field_size: int = DEFAULT_MAX_FIELD_SIZE, max_field_size: int = DEFAULT_MAX_FIELD_SIZE,
max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE, max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE,
max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES, max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES,
@ -342,12 +348,12 @@ class MultipartParser:
self._tempdir = tempfile.gettempdir() self._tempdir = tempfile.gettempdir()
# Current part state # Current part state
self.current_headers: dict[str, str] = {} self.current_headers: Dict[str, str] = {}
self.current_file: tempfile.SpooledTemporaryFile | None = None self.current_file: Optional[tempfile.SpooledTemporaryFile] = None
self.current_body = bytearray() self.current_body = bytearray()
self.current_name: str | None = None self.current_name: Optional[str] = None
self.current_filename: str | None = None self.current_filename: Optional[str] = None
self.current_content_type: str | None = None self.current_content_type: Optional[str] = None
def feed(self, chunk: bytes) -> None: def feed(self, chunk: bytes) -> None:
"""Feed a chunk of data to the parser.""" """Feed a chunk of data to the parser."""
@ -358,13 +364,6 @@ class MultipartParser:
self.buffer.extend(chunk) self.buffer.extend(chunk)
self._process() self._process()
def close(self) -> None:
"""Discard completed uploads and any file still being received."""
if self.current_file is not None:
self.current_file.close()
self.current_file = None
self.form_data.close()
def _process(self) -> None: def _process(self) -> None:
"""Process buffered data.""" """Process buffered data."""
while True: while True:
@ -455,7 +454,7 @@ class MultipartParser:
# Parse header # Parse header
try: try:
line_str = line.decode("utf-8", errors="replace") line_str = line.decode("utf-8", errors="replace")
except UnicodeDecodeError: except Exception:
line_str = line.decode("latin-1") line_str = line.decode("latin-1")
if ":" in line_str: if ":" in line_str:
@ -482,9 +481,7 @@ class MultipartParser:
if self.file_count > self.max_files: if self.file_count > self.max_files:
raise MultipartParseError("Too many files") raise MultipartParseError("Too many files")
if self.handle_files: if self.handle_files:
# Outlives this method - it is filled in across parser callbacks self.current_file = tempfile.SpooledTemporaryFile(
# and then handed to the UploadedFile the caller consumes
self.current_file = tempfile.SpooledTemporaryFile( # noqa: SIM115
max_size=self.max_memory_file_size max_size=self.max_memory_file_size
) )
else: else:
@ -584,9 +581,6 @@ class MultipartParser:
def _finish_part(self) -> None: def _finish_part(self) -> None:
"""Finalize current part and add to form data.""" """Finalize current part and add to form data."""
if self.current_name is None: if self.current_name is None:
if self.current_file is not None:
self.current_file.close()
self.current_file = None
return return
if self.current_filename is not None: if self.current_filename is not None:
@ -650,7 +644,7 @@ async def parse_form_data(
max_request_size: int = DEFAULT_MAX_REQUEST_SIZE, max_request_size: int = DEFAULT_MAX_REQUEST_SIZE,
max_fields: int = DEFAULT_MAX_FIELDS, max_fields: int = DEFAULT_MAX_FIELDS,
max_files: int = DEFAULT_MAX_FILES, max_files: int = DEFAULT_MAX_FILES,
max_parts: int | None = DEFAULT_MAX_PARTS, max_parts: Optional[int] = DEFAULT_MAX_PARTS,
max_field_size: int = DEFAULT_MAX_FIELD_SIZE, max_field_size: int = DEFAULT_MAX_FIELD_SIZE,
max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE, max_memory_file_size: int = DEFAULT_MAX_MEMORY_FILE_SIZE,
max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES, max_part_header_bytes: int = DEFAULT_MAX_PART_HEADER_BYTES,
@ -732,29 +726,12 @@ async def parse_form_data(
batch_target = 64 * 1024 batch_target = 64 * 1024
batch = bytearray() batch = bytearray()
async def run_parser(fn, *args):
# Cancellation must not close files while a worker is using them.
task = asyncio.create_task(asyncio.to_thread(fn, *args))
try:
return await asyncio.shield(task)
except asyncio.CancelledError as cancelled:
try:
while not task.done():
try:
await asyncio.shield(task)
except asyncio.CancelledError:
continue
task.result()
finally:
raise cancelled
async def flush_batch() -> None: async def flush_batch() -> None:
if batch: if batch:
data = bytes(batch) data = bytes(batch)
batch.clear() batch.clear()
await run_parser(parser.feed, data) await asyncio.to_thread(parser.feed, data)
try:
while True: while True:
message = await receive() message = await receive()
message_type = message.get("type") message_type = message.get("type")
@ -771,11 +748,7 @@ async def parse_form_data(
break break
await flush_batch() await flush_batch()
return await run_parser(parser.finalize) return await asyncio.to_thread(parser.finalize)
except BaseException:
# No FormData is returned to the caller to take ownership on failure.
await asyncio.to_thread(parser.close)
raise
else: else:
raise MultipartParseError( raise MultipartParseError(

View file

@ -2,9 +2,8 @@
from __future__ import annotations from __future__ import annotations
import json import json
from typing import Any, Dict, Iterable, List, Sequence, Tuple
import sqlite3 import sqlite3
from collections.abc import Iterable, Sequence
from typing import Any
from datasette.permissions import PermissionSQL from datasette.permissions import PermissionSQL
from datasette.plugins import pm from datasette.plugins import pm
@ -16,7 +15,7 @@ SKIP_PERMISSION_CHECKS = object()
async def gather_permission_sql_from_hooks( async def gather_permission_sql_from_hooks(
*, datasette, actor: dict | None, action: str *, datasette, actor: dict | None, action: str
) -> list[PermissionSQL] | object: ) -> List[PermissionSQL] | object:
"""Collect PermissionSQL objects from the permission_resources_sql hook. """Collect PermissionSQL objects from the permission_resources_sql hook.
Ensures that each returned PermissionSQL has a populated ``source``. Ensures that each returned PermissionSQL has a populated ``source``.
@ -35,7 +34,7 @@ async def gather_permission_sql_from_hooks(
hookimpls = hook_caller.get_hookimpls() hookimpls = hook_caller.get_hookimpls()
hook_results = list(hook_caller(datasette=datasette, actor=actor, action=action)) hook_results = list(hook_caller(datasette=datasette, actor=actor, action=action))
collected: list[PermissionSQL] = [] collected: List[PermissionSQL] = []
actor_json = json.dumps(actor) if actor is not None else None actor_json = json.dumps(actor) if actor is not None else None
actor_id = actor.get("id") if isinstance(actor, dict) else None actor_id = actor.get("id") if isinstance(actor, dict) else None
@ -72,7 +71,7 @@ def _iter_permission_sql_from_result(
if isinstance(result, PermissionSQL): if isinstance(result, PermissionSQL):
return [result] return [result]
if isinstance(result, (list, tuple)): if isinstance(result, (list, tuple)):
collected: list[PermissionSQL] = [] collected: List[PermissionSQL] = []
for item in result: for item in result:
collected.extend(_iter_permission_sql_from_result(item, action=action)) collected.extend(_iter_permission_sql_from_result(item, action=action))
return collected return collected
@ -91,7 +90,7 @@ def _iter_permission_sql_from_result(
def build_rules_union( def build_rules_union(
actor: dict | None, plugins: Sequence[PermissionSQL] actor: dict | None, plugins: Sequence[PermissionSQL]
) -> tuple[str, dict[str, Any]]: ) -> Tuple[str, Dict[str, Any]]:
""" """
Compose plugin SQL into a UNION ALL. Compose plugin SQL into a UNION ALL.
@ -103,10 +102,10 @@ def build_rules_union(
The system reserves these parameter names: :actor, :actor_id, :action, :filter_parent The system reserves these parameter names: :actor, :actor_id, :action, :filter_parent
Plugin parameters should be prefixed with a unique identifier (e.g., source name). Plugin parameters should be prefixed with a unique identifier (e.g., source name).
""" """
parts: list[str] = [] parts: List[str] = []
actor_json = json.dumps(actor) if actor else None actor_json = json.dumps(actor) if actor else None
actor_id = actor.get("id") if actor else None actor_id = actor.get("id") if actor else None
params: dict[str, Any] = {"actor": actor_json, "actor_id": actor_id} params: Dict[str, Any] = {"actor": actor_json, "actor_id": actor_id}
for p in plugins: for p in plugins:
# No namespacing - just use plugin params as-is # No namespacing - just use plugin params as-is
@ -142,10 +141,10 @@ async def resolve_permissions_from_catalog(
plugins: Sequence[Any], plugins: Sequence[Any],
action: str, action: str,
candidate_sql: str, candidate_sql: str,
candidate_params: dict[str, Any] | None = None, candidate_params: Dict[str, Any] | None = None,
*, *,
implicit_deny: bool = True, implicit_deny: bool = True,
) -> list[dict[str, Any]]: ) -> List[Dict[str, Any]]:
""" """
Resolve permissions by embedding the provided *candidate_sql* in a CTE. Resolve permissions by embedding the provided *candidate_sql* in a CTE.
@ -169,8 +168,8 @@ async def resolve_permissions_from_catalog(
- parent, child, allow, reason, source_plugin, depth - parent, child, allow, reason, source_plugin, depth
- resource (rendered "/parent/child" or "/parent" or "/") - resource (rendered "/parent/child" or "/parent" or "/")
""" """
resolved_plugins: list[PermissionSQL] = [] resolved_plugins: List[PermissionSQL] = []
restriction_sqls: list[str] = [] restriction_sqls: List[str] = []
for plugin in plugins: for plugin in plugins:
if callable(plugin) and not isinstance(plugin, PermissionSQL): if callable(plugin) and not isinstance(plugin, PermissionSQL):
@ -399,11 +398,11 @@ async def resolve_permissions_with_candidates(
db, db,
actor: dict | None, actor: dict | None,
plugins: Sequence[Any], plugins: Sequence[Any],
candidates: list[tuple[str, str | None]], candidates: List[Tuple[str, str | None]],
action: str, action: str,
*, *,
implicit_deny: bool = True, implicit_deny: bool = True,
) -> list[dict[str, Any]]: ) -> List[Dict[str, Any]]:
""" """
Resolve permissions without any external candidate table by embedding Resolve permissions without any external candidate table by embedding
the candidates as a UNION of parameterized SELECTs in a CTE. the candidates as a UNION of parameterized SELECTs in a CTE.
@ -412,8 +411,8 @@ async def resolve_permissions_with_candidates(
actor: actor dict (or None), made available as :actor (JSON), :actor_id, and :action actor: actor dict (or None), made available as :actor (JSON), :actor_id, and :action
""" """
# Build a small CTE for candidates. # Build a small CTE for candidates.
cand_rows_sql: list[str] = [] cand_rows_sql: List[str] = []
cand_params: dict[str, Any] = {} cand_params: Dict[str, Any] = {}
for i, (parent, child) in enumerate(candidates): for i, (parent, child) in enumerate(candidates):
pkey = f"cand_p_{i}" pkey = f"cand_p_{i}"
ckey = f"cand_c_{i}" ckey = f"cand_c_{i}"

View file

@ -6,7 +6,7 @@ https://github.com/python/cpython/blob/v3.8.3/LICENSE
""" """
import os import os
from shutil import Error, copy, copy2, copystat from shutil import copy, copy2, copystat, Error
def _copytree( def _copytree(

View file

@ -1,8 +1,6 @@
import sys
from dataclasses import dataclass from dataclasses import dataclass
from typing import Literal from typing import Literal
from datasette.utils import escape_sqlite
from datasette.utils.sqlite import SQLiteTableType, sqlite3, sqlite_table_type from datasette.utils.sqlite import SQLiteTableType, sqlite3, sqlite_table_type
SQLOperation = Literal[ SQLOperation = Literal[
@ -152,6 +150,7 @@ _SQLITE_INTERNAL_SCHEMA_FUNCTIONS = {
"sqlite_rename_test", "sqlite_rename_test",
"substr", "substr",
} }
_AUTHORIZER_ACTION_NAMES = { _AUTHORIZER_ACTION_NAMES = {
getattr(sqlite3, name): name getattr(sqlite3, name): name
for name in ( for name in (
@ -197,16 +196,6 @@ def _allow_authorizer_action(*args):
return sqlite3.SQLITE_OK return sqlite3.SQLITE_OK
def _disable_authorizer(conn):
# Python 3.11 added support for unregistering an authorizer using None.
# On Python 3.10, None is installed as the callback instead, and the next
# statement fails with "not authorized" when sqlite3 tries to call it.
if sys.version_info >= (3, 11):
conn.set_authorizer(None)
else:
conn.set_authorizer(_allow_authorizer_action)
def analyze_sql_tables( def analyze_sql_tables(
conn, conn,
sql: str, sql: str,
@ -220,9 +209,7 @@ def analyze_sql_tables(
This function is synchronous and connection-based. It temporarily installs a This function is synchronous and connection-based. It temporarily installs a
SQLite authorizer, prepares ``EXPLAIN <sql>``, and returns the operation SQLite authorizer, prepares ``EXPLAIN <sql>``, and returns the operation
callbacks observed while SQLite compiles the statement. ``CREATE VIEW`` is callbacks observed while SQLite compiles the statement.
additionally executed inside a rolled-back savepoint so its source-table reads
can be discovered by analyzing a query against the temporary view.
""" """
operations: dict[OperationKey, set[str]] = {} operations: dict[OperationKey, set[str]] = {}
@ -404,10 +391,6 @@ def analyze_sql_tables(
) )
return sqlite3.SQLITE_OK return sqlite3.SQLITE_OK
if action == sqlite3.SQLITE_RECURSIVE:
# Recursive CTE bookkeeping; table reads are reported separately.
return sqlite3.SQLITE_OK
if action == sqlite3.SQLITE_FUNCTION and arg2 is not None: if action == sqlite3.SQLITE_FUNCTION and arg2 is not None:
record( record(
"function", "function",
@ -427,12 +410,12 @@ def analyze_sql_tables(
database=None, database=None,
table=None, table=None,
sqlite_schema=sqlite_schema, sqlite_schema=sqlite_schema,
target=f"{arg1} {arg2}" if arg2 is not None else arg1, target="{} {}".format(arg1, arg2) if arg2 is not None else arg1,
source=source, source=source,
) )
return sqlite3.SQLITE_OK return sqlite3.SQLITE_OK
action_name = _AUTHORIZER_ACTION_NAMES.get(action, f"SQLITE_{action}") action_name = _AUTHORIZER_ACTION_NAMES.get(action, "SQLITE_{}".format(action))
record( record(
"unknown", "unknown",
"unknown", "unknown",
@ -495,24 +478,24 @@ def analyze_sql_tables(
conn, key.table, schema=key.sqlite_schema conn, key.table, schema=key.sqlite_schema
) )
finally: finally:
_disable_authorizer(conn) conn.set_authorizer(None)
has_schema_operation = any( has_schema_operation = any(
key.target_type in {"table", "index", "view", "trigger", "virtual-table"} key.target_type in {"table", "index", "view", "trigger", "virtual-table"}
and key.operation in {"create", "alter", "drop"} and key.operation in {"create", "alter", "drop"}
for key in operations for key in operations
) )
dropped_tables_and_views = { dropped_tables = {
(key.database, key.table) (key.database, key.table)
for key in operations for key in operations
if key.operation == "drop" and key.target_type in {"table", "view"} if key.operation == "drop" and key.target_type == "table"
} }
def key_is_drop_table_delete(key: OperationKey) -> bool: def key_is_drop_table_delete(key: OperationKey) -> bool:
return ( return (
key.operation == "delete" key.operation == "delete"
and key.target_type == "table" and key.target_type == "table"
and (key.database, key.table) in dropped_tables_and_views and (key.database, key.table) in dropped_tables
) )
has_user_table_access_in_schema_operation = any( has_user_table_access_in_schema_operation = any(
@ -535,7 +518,9 @@ def analyze_sql_tables(
and key.target in _SQLITE_INTERNAL_SCHEMA_FUNCTIONS and key.target in _SQLITE_INTERNAL_SCHEMA_FUNCTIONS
): ):
return True return True
return bool(key_is_drop_table_delete(key)) if key_is_drop_table_delete(key):
return True
return False
def table_kind_for(key: OperationKey) -> SQLiteTableType | None: def table_kind_for(key: OperationKey) -> SQLiteTableType | None:
if ( if (
@ -546,7 +531,7 @@ def analyze_sql_tables(
return None return None
return table_kind_cache[(key.sqlite_schema, key.table)] return table_kind_cache[(key.sqlite_schema, key.table)]
analysis = SQLAnalysis( return SQLAnalysis(
operations=tuple( operations=tuple(
Operation( Operation(
operation=key.operation, operation=key.operation,
@ -563,58 +548,3 @@ def analyze_sql_tables(
for key, columns in operations.items() for key, columns in operations.items()
) )
) )
# SQLite does not resolve the SELECT body of a view when preparing CREATE
# VIEW, so its authorizer does not report reads from the view's source
# tables. Temporarily create the view, analyze a query against it (which
# does resolve the body), then roll the schema change back. Database-level
# callers use an isolated writable connection for this analysis.
create_view_operations = tuple(
operation
for operation in analysis.operations
if operation.operation == "create" and operation.target_type == "view"
)
if not create_view_operations:
return analysis
savepoint = "datasette_analyze_create_view"
conn.execute(f"SAVEPOINT {savepoint}")
try:
conn.execute(sql, params if params is not None else {})
dependency_reads = []
for view_operation in create_view_operations:
if view_operation.sqlite_schema is None or view_operation.table is None:
raise sqlite3.OperationalError(
"Could not determine the created view name"
)
quoted_schema = escape_sqlite(view_operation.sqlite_schema)
quoted_view = escape_sqlite(view_operation.table)
qualified_view = f"{quoted_schema}.{quoted_view}"
view_analysis = analyze_sql_tables(
conn,
f"SELECT * FROM {qualified_view}",
database_name=database_name,
schema_to_database=schema_to_database,
)
dependency_reads.extend(
operation
for operation in view_analysis.operations
if operation.operation == "read"
and not (
operation.sqlite_schema == view_operation.sqlite_schema
and operation.table == view_operation.table
)
)
finally:
conn.execute(f"ROLLBACK TO {savepoint}")
conn.execute(f"RELEASE {savepoint}")
existing_operations = set(analysis.operations)
return SQLAnalysis(
operations=analysis.operations
+ tuple(
operation
for operation in dependency_reads
if operation not in existing_operations
)
)

View file

@ -15,17 +15,8 @@ if hasattr(sqlite3, "enable_callback_tracebacks"):
_cached_sqlite_version = None _cached_sqlite_version = None
_cached_supports_returning = None _cached_supports_returning = None
SQLiteTableType = Literal["table", "view", "virtual", "shadow"] SQLiteTableType = Literal["table", "view", "virtual", "shadow"]
_SQLITE_IDENTIFIER_RE = (
r"""(?:"(?:[^"]|"")*"|'(?:[^']|'')*'|`(?:[^`]|``)*`|\[[^\]]*\]|[^\s.()'"`\[\]]+)"""
)
_VIRTUAL_TABLE_MODULE_RE = re.compile( _VIRTUAL_TABLE_MODULE_RE = re.compile(
r"^\s*CREATE\s+VIRTUAL\s+TABLE\b\s*(?:IF\s+NOT\s+EXISTS\s+)?" r"\bCREATE\s+VIRTUAL\s+TABLE\b.*?\bUSING\s+([^\s(]+)",
+ _SQLITE_IDENTIFIER_RE
+ r"(?:\s*\.\s*"
+ _SQLITE_IDENTIFIER_RE
+ r")?\s*\bUSING\b\s*("
+ _SQLITE_IDENTIFIER_RE
+ r")",
re.IGNORECASE | re.DOTALL, re.IGNORECASE | re.DOTALL,
) )
_VIRTUAL_TABLE_SHADOW_SUFFIXES = { _VIRTUAL_TABLE_SHADOW_SUFFIXES = {
@ -92,58 +83,24 @@ def sqlite_table_type(
) -> SQLiteTableType | None: ) -> SQLiteTableType | None:
if supports_table_list(): if supports_table_list():
try: try:
# Use the "PRAGMA table_list" statement form rather than the query = "select type from pragma_table_list where name = ?"
# pragma_table_list(...) table-valued function. The params: tuple[str, ...] = (table,)
# table-valued function is resolved like an ordinary relation
# name, so an attacker-created table or view literally named
# "pragma_table_list" can shadow it and spoof the reported
# type (e.g. claiming a virtual table is an ordinary table).
# The PRAGMA statement form is a distinct piece of SQL syntax
# that always invokes SQLite's built-in pragma, so it cannot
# be shadowed by a user-created relation.
if schema is not None: if schema is not None:
query = f"PRAGMA {_quote_identifier(schema)}.table_list" query += " and schema = ?"
else: params = (table, schema)
query = "PRAGMA table_list" row = conn.execute(query, params).fetchone()
cursor = conn.execute(query) if row is not None and row[0] in {"table", "view", "virtual", "shadow"}:
columns = [description[0] for description in cursor.description] return row[0]
for row in cursor.fetchall():
record = dict(zip(columns, row))
if record.get("name") != table:
continue
if schema is not None and record.get("schema") != schema:
continue
row_type = record.get("type")
if row_type in {"table", "view", "virtual", "shadow"}:
return row_type
except sqlite3.DatabaseError: except sqlite3.DatabaseError:
pass pass
return _sqlite_table_type_from_schema(conn, table, schema=schema) return _sqlite_table_type_from_schema(conn, table, schema=schema)
def check_structured_write_table(conn, table: str, *, allow_missing=False):
"""Validate a row-write target on the connection that will perform the write."""
# SQLite resolves identifiers case-insensitively. The create API must not
# treat a differently cased existing name as a missing table.
row = conn.execute(
"select name from main.sqlite_master where name = ? collate nocase "
"and type in ('table', 'view')",
(table,),
).fetchone()
if row is None and allow_missing:
return
if row is not None and sqlite_table_type(conn, row[0]) == "table":
return
# Virtual table modules can interpret row writes as administrative operations.
# Their shadow tables are internal storage, not independently writable data.
raise ValueError("Structured writes require an ordinary table")
def sqlite_hidden_table_names(conn, *, schema: str | None = "main") -> list[str]: def sqlite_hidden_table_names(conn, *, schema: str | None = "main") -> list[str]:
schema_table = _sqlite_schema_table(schema) schema_table = _sqlite_schema_table(schema)
try: try:
rows = conn.execute( rows = conn.execute(
f"select name, sql from {schema_table} where type = 'table'" "select name, sql from {} where type = 'table'".format(schema_table)
).fetchall() ).fetchall()
except sqlite3.DatabaseError: except sqlite3.DatabaseError:
return [] return []
@ -161,63 +118,6 @@ def sqlite_hidden_table_names(conn, *, schema: str | None = "main") -> list[str]
return sorted(hidden_tables) + content_fts_tables return sorted(hidden_tables) + content_fts_tables
def sqlite_derived_table_dependencies(
conn, *, schema: str | None = "main"
) -> dict[str, str]:
"""Return implementation table -> logical/content table dependencies.
``PRAGMA table_list`` safely identifies virtual and shadow tables, but
does not report which virtual table owns a shadow table or which table is
named by an FTS ``content=`` option. Derive those relationships from
``sqlite_master`` DDL and the documented shadow-table suffixes.
Database errors propagate: failed discovery must not be mistaken for an
empty dependency map and cached as permission to skip inheritance.
"""
schema_table = _sqlite_schema_table(schema)
rows = conn.execute(
f"select name, sql from {schema_table} where type = 'table'"
).fetchall()
table_names = {row[0] for row in rows}
# SQLite identifiers fold ASCII letters only.
identifier_case = str.maketrans(
"ABCDEFGHIJKLMNOPQRSTUVWXYZ", "abcdefghijklmnopqrstuvwxyz"
)
canonical_names = {name.translate(identifier_case): name for name in table_names}
dependencies = {}
for virtual_table, sql in rows:
module = _virtual_table_module(sql)
if module is None:
continue
# SQLite's documented shadow tables are implementation details of
# their logical virtual table.
for suffix in _VIRTUAL_TABLE_SHADOW_SUFFIXES.get(module, ()):
shadow_table = virtual_table + suffix
if shadow_table in table_names:
dependencies[shadow_table] = virtual_table
# An external-content FTS table can expose values fetched from its
# content table, so it must also depend on that table's permission.
if module in {"fts3", "fts4", "fts5"}:
content_table = _fts_external_content_table(sql)
if content_table:
dependencies[virtual_table] = content_table
if module in {"fts5vocab", "fts4aux"}:
source = _fts_vocabulary_source(sql, module, schema or "main")
source = (
canonical_names.get(source.translate(identifier_case))
if source
else None
)
# An unresolved source is itself derived, so the one-hop policy denies it.
dependencies[virtual_table] = source or virtual_table
return dependencies
def _sqlite_table_type_from_schema( def _sqlite_table_type_from_schema(
conn, conn,
table: str, table: str,
@ -227,7 +127,7 @@ def _sqlite_table_type_from_schema(
schema_table = _sqlite_schema_table(schema) schema_table = _sqlite_schema_table(schema)
try: try:
row = conn.execute( row = conn.execute(
f"select type, sql from {schema_table} where name = ?", "select type, sql from {} where name = ?".format(schema_table),
(table,), (table,),
).fetchone() ).fetchone()
except sqlite3.DatabaseError: except sqlite3.DatabaseError:
@ -255,7 +155,7 @@ def _is_known_shadow_table(
schema_table = _sqlite_schema_table(schema) schema_table = _sqlite_schema_table(schema)
try: try:
rows = conn.execute( rows = conn.execute(
f"select name, sql from {schema_table} where type = 'table'" "select name, sql from {} where type = 'table'".format(schema_table)
).fetchall() ).fetchall()
except sqlite3.DatabaseError: except sqlite3.DatabaseError:
return False return False
@ -274,7 +174,7 @@ def _sqlite_schema_table(schema: str | None) -> str:
return "sqlite_master" return "sqlite_master"
if schema == "temp": if schema == "temp":
return "sqlite_temp_master" return "sqlite_temp_master"
return f"{_quote_identifier(schema)}.sqlite_master" return "{}.sqlite_master".format(_quote_identifier(schema))
def _quote_identifier(value: str) -> str: def _quote_identifier(value: str) -> str:
@ -284,151 +184,10 @@ def _quote_identifier(value: str) -> str:
def _virtual_table_module(sql: str | None) -> str | None: def _virtual_table_module(sql: str | None) -> str | None:
if not sql: if not sql:
return None return None
match = _VIRTUAL_TABLE_MODULE_RE.search(_strip_sql_comments(sql))
if match is None:
return None
return _unquote_sql_value(match.group(1)).lower()
def _fts_external_content_table(sql: str | None) -> str | None:
"""Extract the external ``content=`` table from an FTS declaration."""
if not sql:
return None
sql = _strip_sql_comments(sql)
match = _VIRTUAL_TABLE_MODULE_RE.search(sql) match = _VIRTUAL_TABLE_MODULE_RE.search(sql)
if match is None: if match is None:
return None return None
open_paren = sql.find("(", match.end()) return match.group(1).strip("\"'[]`").lower()
if open_paren == -1:
return None
close_paren = sql.rfind(")")
if close_paren <= open_paren:
return None
for argument in _split_sql_arguments(sql[open_paren + 1 : close_paren]):
key, separator, value = argument.partition("=")
if not separator or key.strip().lower() != "content":
continue
return _unquote_sql_value(value.strip())
return None
def _fts_vocabulary_source(sql: str, module: str, schema: str) -> str | None:
"""Resolve a vocabulary source within the current SQLite schema.
Cross-schema sources cannot be represented by the dependency map and
are conservatively left unresolved.
"""
sql = _strip_sql_comments(sql)
match = _VIRTUAL_TABLE_MODULE_RE.search(sql)
if match is None:
return None
start = sql.find("(", match.end())
end = sql.rfind(")")
if start < 0 or end <= start:
return None
arguments = [
_unquote_sql_value(arg.strip())
for arg in _split_sql_arguments(sql[start + 1 : end])
]
expected = 2 if module == "fts5vocab" else 1
if len(arguments) == expected:
return arguments[0]
if len(arguments) == expected + 1 and arguments[0].lower() == schema.lower():
return arguments[1]
return None
def _split_sql_arguments(arguments: str) -> list[str]:
"""Split comma-separated SQLite arguments without splitting quoted text."""
parts = []
start = 0
quote = None
closing_quote = None
index = 0
while index < len(arguments):
char = arguments[index]
if quote is None:
if char in {"'", '"', "`", "["}:
quote = char
closing_quote = "]" if char == "[" else char
elif char == ",":
parts.append(arguments[start:index])
start = index + 1
elif char == closing_quote:
# Single/double/backtick quoting escapes the delimiter by
# doubling it. Square-bracket identifiers do not.
if (
quote != "["
and index + 1 < len(arguments)
and arguments[index + 1] == closing_quote
):
index += 1
else:
quote = None
closing_quote = None
index += 1
parts.append(arguments[start:])
return parts
def _strip_sql_comments(sql: str) -> str:
"""Remove SQLite comments while preserving quoted strings/identifiers."""
output = []
quote = None
closing_quote = None
index = 0
while index < len(sql):
char = sql[index]
next_char = sql[index + 1] if index + 1 < len(sql) else ""
if quote is None:
if char in {"'", '"', "`", "["}:
quote = char
closing_quote = "]" if char == "[" else char
output.append(char)
elif char == "-" and next_char == "-":
index += 2
while index < len(sql) and sql[index] not in "\r\n":
index += 1
output.append(" ")
continue
elif char == "/" and next_char == "*":
index += 2
while index + 1 < len(sql) and sql[index : index + 2] != "*/":
index += 1
index = min(index + 2, len(sql))
output.append(" ")
continue
else:
output.append(char)
else:
output.append(char)
if char == closing_quote:
if (
quote != "["
and index + 1 < len(sql)
and sql[index + 1] == closing_quote
):
output.append(sql[index + 1])
index += 1
else:
quote = None
closing_quote = None
index += 1
return "".join(output)
def _unquote_sql_value(value: str) -> str:
if len(value) < 2:
return value
pairs = {"'": "'", '"': '"', "`": "`", "[": "]"}
closing = pairs.get(value[0])
if closing is None or value[-1] != closing:
return value
unquoted = value[1:-1]
if value[0] != "[":
unquoted = unquoted.replace(closing * 2, closing)
return unquoted
def _is_fts_content_virtual_table(sql: str | None) -> bool: def _is_fts_content_virtual_table(sql: str | None) -> bool:

View file

@ -1,10 +1,9 @@
import json
from urllib.parse import urlencode
from asgiref.sync import async_to_sync from asgiref.sync import async_to_sync
from urllib.parse import urlencode
import json
# These wrapper classes pre-date the introduction of # These wrapper classes pre-date the introduction of
# datasette.client and httpx2 to Datasette. They could # datasette.client and httpx to Datasette. They could
# be removed if the Datasette tests are modified to # be removed if the Datasette tests are modified to
# call datasette.client directly. # call datasette.client directly.

View file

@ -1,2 +1,2 @@
__version__ = "1.0a40" __version__ = "1.0a34"
__version_info__ = tuple(__version__.split(".")) __version_info__ = tuple(__version__.split("."))

View file

@ -1,87 +1,2 @@
import dataclasses
import types
import typing
from dataclasses import dataclass
@dataclass(frozen=True)
class ContextField:
name: str
type_name: str
help: str
from_extra: bool = False
def _type_name(type_):
if type_ is type(None):
return "None"
origin = typing.get_origin(type_)
args = typing.get_args(type_)
if origin in (typing.Union, types.UnionType):
return " | ".join(_type_name(arg) for arg in args)
if origin is not None:
name = getattr(origin, "__name__", str(origin).removeprefix("typing."))
return "{}[{}]".format(name, ", ".join(_type_name(arg) for arg in args))
return getattr(type_, "__name__", str(type_).removeprefix("typing."))
def from_extra():
"""
Declare a Context dataclass field whose value comes from a registered
Extra of the same name - its documentation is the Extra description,
so the doc string lives next to the resolve() code rather than being
duplicated on the dataclass.
"""
return dataclasses.field(metadata={"from_extra": True})
class Context: class Context:
"Base class for all documented contexts" "Base class for all documented contexts"
# Set on subclasses whose from_extra() fields should be resolved
# against the extras registry for this scope
extras_scope = None
@classmethod
def documented_fields(cls):
"List of ContextField describing the documented fields of this context"
documented = []
for f in dataclasses.fields(cls):
if f.name.startswith("_"):
continue
is_from_extra = bool(f.metadata.get("from_extra"))
if is_from_extra:
help_text = cls._extra_description(f.name)
else:
help_text = f.metadata.get("help", "")
documented.append(
ContextField(
name=f.name,
type_name=_type_name(f.type),
help=help_text,
from_extra=is_from_extra,
)
)
return documented
@classmethod
def _extra_description(cls, name):
# Imported lazily - table_extras is not needed just to define
# Context subclasses
from datasette.views.table_extras import table_extra_registry
try:
extra_class = table_extra_registry.classes_by_name[name]
except KeyError:
raise KeyError(
f"{cls.__name__}.{name} is declared with from_extra() but there is no "
"registered extra of that name"
)
if cls.extras_scope is not None and not extra_class.available_for(
cls.extras_scope
):
raise ValueError(
f"{cls.__name__}.{name} is declared with from_extra() but the {name} extra is "
f"not available for scope {cls.extras_scope}"
)
return extra_class.description or ""

Some files were not shown because too many files have changed in this diff Show more