mirror of
https://github.com/simonw/datasette.git
synced 2025-12-10 16:51:24 +01:00
Switch to dash encoding for table/database/row-pk in paths
* Dash encoding functions, tests and docs, refs #1439 * dash encoding is now like percent encoding but with dashes * Use dash-encoding for row PKs and ?_next=, refs #1439 * Use dash encoding for table names, refs #1439 * Use dash encoding for database names, too, refs #1439 See also https://simonwillison.net/2022/Mar/5/dash-encoding/
This commit is contained in:
Simon Willison
GitHub
parent
de810f49cc
commit
1baa030eca
13 changed files with 173 additions and 53 deletions
|
|
@ -1,4 +1,4 @@
|
|||
from .utils import path_with_format, HASH_LENGTH, PrefixedUrlString
|
||||
from .utils import dash_encode, path_with_format, HASH_LENGTH, PrefixedUrlString
|
||||
import urllib
|
||||
|
||||
|
||||
|
|
@ -31,20 +31,20 @@ class Urls:
|
|||
db = self.ds.databases[database]
|
||||
if self.ds.setting("hash_urls") and db.hash:
|
||||
path = self.path(
|
||||
f"{urllib.parse.quote(database)}-{db.hash[:HASH_LENGTH]}", format=format
|
||||
f"{dash_encode(database)}-{db.hash[:HASH_LENGTH]}", format=format
|
||||
)
|
||||
else:
|
||||
path = self.path(urllib.parse.quote(database), format=format)
|
||||
path = self.path(dash_encode(database), format=format)
|
||||
return path
|
||||
|
||||
def table(self, database, table, format=None):
|
||||
path = f"{self.database(database)}/{urllib.parse.quote_plus(table)}"
|
||||
path = f"{self.database(database)}/{dash_encode(table)}"
|
||||
if format is not None:
|
||||
path = path_with_format(path=path, format=format)
|
||||
return PrefixedUrlString(path)
|
||||
|
||||
def query(self, database, query, format=None):
|
||||
path = f"{self.database(database)}/{urllib.parse.quote_plus(query)}"
|
||||
path = f"{self.database(database)}/{dash_encode(query)}"
|
||||
if format is not None:
|
||||
path = path_with_format(path=path, format=format)
|
||||
return PrefixedUrlString(path)
|
||||
|
|
|
|||
|
|
@ -112,12 +112,12 @@ async def await_me_maybe(value: typing.Any) -> typing.Any:
|
|||
|
||||
|
||||
def urlsafe_components(token):
|
||||
"""Splits token on commas and URL decodes each component"""
|
||||
return [urllib.parse.unquote_plus(b) for b in token.split(",")]
|
||||
"""Splits token on commas and dash-decodes each component"""
|
||||
return [dash_decode(b) for b in token.split(",")]
|
||||
|
||||
|
||||
def path_from_row_pks(row, pks, use_rowid, quote=True):
|
||||
"""Generate an optionally URL-quoted unique identifier
|
||||
"""Generate an optionally dash-quoted unique identifier
|
||||
for a row from its primary keys."""
|
||||
if use_rowid:
|
||||
bits = [row["rowid"]]
|
||||
|
|
@ -126,7 +126,7 @@ def path_from_row_pks(row, pks, use_rowid, quote=True):
|
|||
row[pk]["value"] if isinstance(row[pk], dict) else row[pk] for pk in pks
|
||||
]
|
||||
if quote:
|
||||
bits = [urllib.parse.quote_plus(str(bit)) for bit in bits]
|
||||
bits = [dash_encode(str(bit)) for bit in bits]
|
||||
else:
|
||||
bits = [str(bit) for bit in bits]
|
||||
|
||||
|
|
@ -1140,3 +1140,36 @@ def add_cors_headers(headers):
|
|||
headers["Access-Control-Allow-Origin"] = "*"
|
||||
headers["Access-Control-Allow-Headers"] = "Authorization"
|
||||
headers["Access-Control-Expose-Headers"] = "Link"
|
||||
|
||||
|
||||
_DASH_ENCODING_SAFE = frozenset(
|
||||
b"ABCDEFGHIJKLMNOPQRSTUVWXYZ"
|
||||
b"abcdefghijklmnopqrstuvwxyz"
|
||||
b"0123456789_"
|
||||
# This is the same as Python percent-encoding but I removed
|
||||
# '.' and '-' and '~'
|
||||
)
|
||||
|
||||
|
||||
class DashEncoder(dict):
|
||||
# Keeps a cache internally, via __missing__
|
||||
def __missing__(self, b):
|
||||
# Handle a cache miss, store encoded string in cache and return.
|
||||
res = chr(b) if b in _DASH_ENCODING_SAFE else "-{:02X}".format(b)
|
||||
self[b] = res
|
||||
return res
|
||||
|
||||
|
||||
_dash_encoder = DashEncoder().__getitem__
|
||||
|
||||
|
||||
@documented
|
||||
def dash_encode(s: str) -> str:
|
||||
"Returns dash-encoded string - for example ``/foo/bar`` -> ``-2Ffoo-2Fbar``"
|
||||
return "".join(_dash_encoder(char) for char in s.encode("utf-8"))
|
||||
|
||||
|
||||
@documented
|
||||
def dash_decode(s: str) -> str:
|
||||
"Decodes a dash-encoded string, so ``-2Ffoo-2Fbar`` -> ``/foo/bar``"
|
||||
return urllib.parse.unquote(s.replace("-", "%"))
|
||||
|
|
|
|||
|
|
@ -17,6 +17,8 @@ from datasette.utils import (
|
|||
InvalidSql,
|
||||
LimitedWriter,
|
||||
call_with_supported_arguments,
|
||||
dash_decode,
|
||||
dash_encode,
|
||||
path_from_row_pks,
|
||||
path_with_added_args,
|
||||
path_with_removed_args,
|
||||
|
|
@ -203,17 +205,17 @@ class DataView(BaseView):
|
|||
async def resolve_db_name(self, request, db_name, **kwargs):
|
||||
hash = None
|
||||
name = None
|
||||
db_name = urllib.parse.unquote_plus(db_name)
|
||||
if db_name not in self.ds.databases and "-" in db_name:
|
||||
decoded_name = dash_decode(db_name)
|
||||
if decoded_name not in self.ds.databases and "-" in db_name:
|
||||
# No matching DB found, maybe it's a name-hash?
|
||||
name_bit, hash_bit = db_name.rsplit("-", 1)
|
||||
if name_bit not in self.ds.databases:
|
||||
if dash_decode(name_bit) not in self.ds.databases:
|
||||
raise NotFound(f"Database not found: {name}")
|
||||
else:
|
||||
name = name_bit
|
||||
name = dash_decode(name_bit)
|
||||
hash = hash_bit
|
||||
else:
|
||||
name = db_name
|
||||
name = decoded_name
|
||||
|
||||
try:
|
||||
db = self.ds.databases[name]
|
||||
|
|
@ -233,9 +235,7 @@ class DataView(BaseView):
|
|||
return await db.table_exists(t)
|
||||
|
||||
table, _format = await resolve_table_and_format(
|
||||
table_and_format=urllib.parse.unquote_plus(
|
||||
kwargs["table_and_format"]
|
||||
),
|
||||
table_and_format=dash_decode(kwargs["table_and_format"]),
|
||||
table_exists=async_table_exists,
|
||||
allowed_formats=self.ds.renderers.keys(),
|
||||
)
|
||||
|
|
@ -243,11 +243,11 @@ class DataView(BaseView):
|
|||
if _format:
|
||||
kwargs["as_format"] = f".{_format}"
|
||||
elif kwargs.get("table"):
|
||||
kwargs["table"] = urllib.parse.unquote_plus(kwargs["table"])
|
||||
kwargs["table"] = dash_decode(kwargs["table"])
|
||||
|
||||
should_redirect = self.ds.urls.path(f"{name}-{expected}")
|
||||
if kwargs.get("table"):
|
||||
should_redirect += "/" + urllib.parse.quote_plus(kwargs["table"])
|
||||
should_redirect += "/" + dash_encode(kwargs["table"])
|
||||
if kwargs.get("pk_path"):
|
||||
should_redirect += "/" + kwargs["pk_path"]
|
||||
if kwargs.get("as_format"):
|
||||
|
|
@ -467,7 +467,7 @@ class DataView(BaseView):
|
|||
return await db.table_exists(t)
|
||||
|
||||
table, _ext_format = await resolve_table_and_format(
|
||||
table_and_format=urllib.parse.unquote_plus(args["table_and_format"]),
|
||||
table_and_format=dash_decode(args["table_and_format"]),
|
||||
table_exists=async_table_exists,
|
||||
allowed_formats=self.ds.renderers.keys(),
|
||||
)
|
||||
|
|
@ -475,7 +475,7 @@ class DataView(BaseView):
|
|||
args["table"] = table
|
||||
del args["table_and_format"]
|
||||
elif "table" in args:
|
||||
args["table"] = urllib.parse.unquote_plus(args["table"])
|
||||
args["table"] = dash_decode(args["table"])
|
||||
return _format, args
|
||||
|
||||
async def view_get(self, request, database, hash, correct_hash_provided, **kwargs):
|
||||
|
|
|
|||
|
|
@ -12,6 +12,7 @@ from datasette.utils import (
|
|||
MultiParams,
|
||||
append_querystring,
|
||||
compound_keys_after_sql,
|
||||
dash_encode,
|
||||
escape_sqlite,
|
||||
filters_should_redirect,
|
||||
is_url,
|
||||
|
|
@ -142,7 +143,7 @@ class RowTableShared(DataView):
|
|||
'<a href="{base_url}{database}/{table}/{flat_pks_quoted}">{flat_pks}</a>'.format(
|
||||
base_url=base_url,
|
||||
database=database,
|
||||
table=urllib.parse.quote_plus(table),
|
||||
table=dash_encode(table),
|
||||
flat_pks=str(markupsafe.escape(pk_path)),
|
||||
flat_pks_quoted=path_from_row_pks(row, pks, not pks),
|
||||
)
|
||||
|
|
@ -199,8 +200,8 @@ class RowTableShared(DataView):
|
|||
link_template.format(
|
||||
database=database,
|
||||
base_url=base_url,
|
||||
table=urllib.parse.quote_plus(other_table),
|
||||
link_id=urllib.parse.quote_plus(str(value)),
|
||||
table=dash_encode(other_table),
|
||||
link_id=dash_encode(str(value)),
|
||||
id=str(markupsafe.escape(value)),
|
||||
label=str(markupsafe.escape(label)) or "-",
|
||||
)
|
||||
|
|
@ -765,7 +766,7 @@ class TableView(RowTableShared):
|
|||
if prefix is None:
|
||||
prefix = "$null"
|
||||
else:
|
||||
prefix = urllib.parse.quote_plus(str(prefix))
|
||||
prefix = dash_encode(str(prefix))
|
||||
next_value = f"{prefix},{next_value}"
|
||||
added_args = {"_next": next_value}
|
||||
if sort:
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue