commits

Table actions

8 rows where author = 9599, "author_date" is on date 2017-11-17 and "committer_date" is on date 2017-11-17 sorted by author_date descending

✎ View and edit SQL

This data as json, CSV (advanced)

Suggested facets: raw_committer, committer, author_date (date), committer_date (date)

repo 1

author 1

  • simonw · 8 ✖
sha message author_date ▲ committer_date raw_author raw_committer repo author committer
0b702f3679a2ffd4e3efb5c34b9fe30221172ccb Fixed weird edge-case with foreign key detection It turns out it is possible for a SQLite table to define a foreign key relationship to a table that does not actually exist We should still be able to handle these databases. 2017-11-17T16:18:26Z 2017-11-17T16:18:26Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
45e502aace6cc1198cc5f9a04d61b4a1860a012b Added unit tests for inspect() foreign key detection Used them to fix a bug with it. Refs #85 2017-11-17T16:08:11Z 2017-11-17T16:08:11Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
e16ca1169cc49fbb5b0325564d248becad3a4b7f Refactored table column/row display logic Simplified the template and made the way for upcoming foreign key work. Refs #85 Also fixed   bug on database page - closes #113 2017-11-17T15:39:36Z 2017-11-17T15:39:36Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
084350b0f1343d988928cae63cfedbeb6205e25e Switched to gather_request=False for Sanic tests Gets rid of those ugly _, response = lines. 2017-11-17T14:53:37Z 2017-11-17T14:53:37Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
f3445e2d13e2c38bfbf75f35c2aa3e50df0040f1 Install python3-dev rothar than python-dev 2017-11-17T14:18:49Z 2017-11-17T14:18:49Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
1a6cfcc10f9abcc743b45437996c7264cc2a57b2 Added a .dockerignore Figured this would be useful while testing out 03572ae3557 2017-11-17T14:18:23Z 2017-11-17T14:18:23Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
03572ae35573c2ea802a540624ce116f540ba1ac Allow --load-extension to be set via environment variable I tesed this by first building and running a container using the new Dockerfile from #114: docker build . docker run -it -p 8001:8001 6c9ca7e29181 /bin/sh Then I ran this inside the container itself: apt update && apt-get install wget -y \ && wget http://www.gaia-gis.it/spatialite-2.3.1/test-2.3.sqlite.gz \ && gunzip test-2.3.sqlite.gz \ && mv test-2.3.sqlite test23.sqlite \ && datasette -h 0.0.0.0 test23.sqlite I visited this URL to confirm I got an error due to spatialite not being loaded: http://localhost:8001/test23-c88bc35?sql=select+ST_AsText%28Geometry%29+from+HighWays+limit+1 Then I checked that loading it with `--load-extension` worked correctly: datasette -h 0.0.0.0 test23.sqlite \ --load-extension=/usr/lib/x86_64-linux-gnu/mod_spatialite.so Then, finally, I tested it with the new environment variable option: SQLITE_EXTENSIONS=/usr/lib/x86_64-linux-gnu/mod_spatialite.so \ datasette -h 0.0.0.0 test23.sqlite Running it with an invalid environment variable option shows an error: $ SQLITE_EXTENSIONS=/usr/lib/x86_64-linux-gnu/blah.so datasette \ -h 0.0.0.0 test23.sqlite Usage: datasette -h [OPTIONS] [FILES]... Error: Invalid value for "--load-extension": Path "/usr/lib/x86_64-linux-gnu/blah.so" does not exist. Closes #112 2017-11-17T14:13:35Z 2017-11-17T14:13:35Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 datasette 107914493 simonw 9599 simonw 9599
b9af49be6c68d6e551c1ae23132711f22cb7c63c Fix case for Sanic dependency On PyPI it has a capital letter: https://pypi.python.org/pypi/Sanic I think this is why the list of "Requires Distributions" on https://pypi.python.org/pypi/datasette links to all of the other dependencies but does not link to Sanic. 2017-11-17T13:26:05Z 2017-11-17T13:26:05Z Simon Willison 2946d096d0cdefdc017559e6b57e87658736e843 GitHub cd792325681cbad9f663f2879d8b69f1edbb678f datasette 107914493 simonw 9599 web-flow 19864447

Advanced export

JSON shape: default, array, newline-delimited, object

CSV options: 

CREATE TABLE [commits] (
   [sha] TEXT PRIMARY KEY,
   [message] TEXT,
   [author_date] TEXT,
   [committer_date] TEXT,
   [raw_author] TEXT REFERENCES [raw_authors]([id]),
   [raw_committer] TEXT REFERENCES [raw_authors]([id]),
   [repo] INTEGER REFERENCES [repos]([id]),
   [author] INTEGER REFERENCES [users]([id]),
   [committer] INTEGER REFERENCES [users]([id])
);
CREATE INDEX [idx_commits_committer]
                ON [commits] ([committer]);
CREATE INDEX [idx_commits_author]
                ON [commits] ([author]);
CREATE INDEX [idx_commits_repo]
                ON [commits] ([repo]);
CREATE INDEX [idx_commits_raw_committer]
                ON [commits] ([raw_committer]);
CREATE INDEX [idx_commits_raw_author]
                ON [commits] ([raw_author]);