sqlite-utils json dogs.db "select * from dogs"

This commit is contained in:
Simon Willison 2019-01-25 18:06:29 -08:00
commit 5466c9745d
4 changed files with 158 additions and 6 deletions

View file

@ -11,11 +11,59 @@ Running queries and returning CSV
You can execute a SQL query against a database and get the results back as CSV like this::
$ sqlite-utils csv docs.db "select id, title, author from documents"
$ sqlite-utils csv dogs.db "select * from dogs"
id,age,name
1,4,Cleo
2,2,Pancakes
This will default to including the column names as a header row. To exclude the headers, use ``--no-headers``::
$ sqlite-utils csv docs.db "select id, title, author from documents" --no-headers
$ sqlite-utils csv dogs.db "select * from dogs" --no-headers
1,4,Cleo
2,2,Pancakes
Running queries and returning JSON
==================================
You can execute a SQL query against a database and get the results back as JSON like this::
$ sqlite-utils json dogs.db "select * from dogs"
[{"id": 1, "age": 4, "name": "Cleo"},
{"id": 2, "age": 2, "name": "Pancakes"}]
Use ``--nl`` to get back newline-delimited JSON objects::
$ sqlite-utils json --nl dogs.db "select * from dogs"
{"id": 1, "age": 4, "name": "Cleo"}
{"id": 2, "age": 2, "name": "Pancakes"}
You can use ``--arrays`` to request ararys instead of objects::
$ sqlite-utils json --arrays dogs.db "select * from dogs"
[[1, 4, "Cleo"],
[2, 2, "Pancakes"]]
You can also combine ``--arrays`` and ``--nl``::
$ sqlite-utils json --arrays --nl dogs.db "select * from dogs"
[1, 4, "Cleo"]
[2, 2, "Pancakes"]
If you want to pretty-print the output further, you can pipe it through ``python -mjson.tool``::
$ sqlite-utils json dogs.db "select * from dogs" | python -mjson.tool
[
{
"id": 1,
"age": 4,
"name": "Cleo"
},
{
"id": 2,
"age": 2,
"name": "Pancakes"
}
]
Listing tables
==============
@ -71,7 +119,6 @@ You can import all three records into an automatically created ``dogs`` table an
$ sqlite-utils insert dogs.db dogs dogs.json --pk=id
Upserting data
==============

View file

@ -1,8 +1,10 @@
import click
import sqlite_utils
import json
from sqlite_utils.utils import iter_pairs
import json as json_std
import sys
import csv as csv_std
import sqlite3
@click.group()
@ -72,7 +74,7 @@ def optimize(path, no_vacuum):
def insert(path, table, json_file, pk):
"Insert records from JSON file into the table, create table if it is missing"
db = sqlite_utils.Database(path)
docs = json.load(json_file)
docs = json_std.load(json_file)
if isinstance(docs, dict):
docs = [docs]
db[table].insert_all(docs, pk=pk)
@ -90,7 +92,7 @@ def insert(path, table, json_file, pk):
def upsert(path, table, json_file, pk):
"Upsert records based on their primary key"
db = sqlite_utils.Database(path)
docs = json.load(json_file)
docs = json_std.load(json_file)
if isinstance(docs, dict):
docs = [docs]
db[table].upsert_all(docs, pk=pk)
@ -115,3 +117,42 @@ def csv(path, sql, no_headers):
writer.writerow([c[0] for c in cursor.description])
for row in cursor:
writer.writerow(row)
@cli.command()
@click.argument(
"path",
type=click.Path(file_okay=True, dir_okay=False, allow_dash=False),
required=True,
)
@click.argument("sql")
@click.option("--nl", help="Output newline-delimited JSON", is_flag=True, default=False)
@click.option(
"--arrays",
help="Output rows as arrays instead of objects",
is_flag=True,
default=False,
)
def json(path, sql, nl, arrays):
"Execute SQL query and return the results as JSON"
db = sqlite_utils.Database(path)
cursor = iter(db.conn.execute(sql))
# We have to iterate two-at-a-time so we can know if we
# should output a trailing comma or if we have reached
# the last row.
row = None
first = True
headers = [c[0] for c in cursor.description]
for row, next_row, is_last in iter_pairs(cursor):
# We now reliably have row and next_row
data = row
if not arrays:
data = dict(zip(headers, row))
line = "{firstchar}{serialized}{maybecomma}{lastchar}".format(
firstchar=("[" if first else " ") if not nl else "",
serialized=json_std.dumps(data),
maybecomma="," if (not nl and not is_last) else "",
lastchar="]" if (is_last and not nl) else "",
)
click.echo(line)
first = False

28
sqlite_utils/utils.py Normal file
View file

@ -0,0 +1,28 @@
def iter_pairs(iterator):
# Yields (item, next_item, is_last) pairs
# Last row is (item, None, True)
first = True
next_item = None
while True:
next_done = False
if first:
item, done = next_with_done(iterator)
if done:
break
next_item, next_done = next_with_done(iterator)
first = False
else:
item = next_item
next_item, next_done = next_with_done(iterator)
yield item, next_item, next_done
if next_done:
break
def next_with_done(iterator):
try:
return next(iterator), False
except StopIteration:
return None, True

View file

@ -151,3 +151,39 @@ def test_csv(db_path):
cli.cli, ["csv", db_path, "select id, name, age from dogs", "--no-headers"]
)
assert "1,Cleo,4\n2,Pancakes,2\n" == result.output
@pytest.mark.parametrize(
"args,expected",
[
(
[],
'[{"id": 1, "name": "Cleo", "age": 4},\n {"id": 2, "name": "Pancakes", "age": 2}]',
),
(
["--nl"],
'{"id": 1, "name": "Cleo", "age": 4}\n{"id": 2, "name": "Pancakes", "age": 2}',
),
(["--arrays"], '[[1, "Cleo", 4],\n [2, "Pancakes", 2]]'),
(["--arrays", "--nl"], '[1, "Cleo", 4]\n[2, "Pancakes", 2]'),
],
)
def test_json(db_path, args, expected):
db = Database(db_path)
with db.conn:
db["dogs"].insert_all(
[
{"id": 1, "age": 4, "name": "Cleo"},
{"id": 2, "age": 2, "name": "Pancakes"},
]
)
result = CliRunner().invoke(
cli.cli, ["csv", db_path, "select id, name, age from dogs"]
)
assert 0 == result.exit_code
assert "id,name,age\n1,Cleo,4\n2,Pancakes,2\n" == result.output
# Test the no-headers option:
result = CliRunner().invoke(
cli.cli, ["json", db_path, "select id, name, age from dogs"] + args
)
assert expected == result.output.strip()