]> git.ipfire.org Git - thirdparty/Python/cpython.git/commit
gh-118221: Always use the default row factory in sqlite3.iterdump() (#118223)
authorErlend E. Aasland <erlend@python.org>
Thu, 25 Apr 2024 08:11:45 +0000 (10:11 +0200)
committerGitHub <noreply@github.com>
Thu, 25 Apr 2024 08:11:45 +0000 (10:11 +0200)
commite38b43c213a8ab2ad9748bac2732af9b58c816ae
treef3244a81e7f38983d6f09dea67e9ec418deff1ef
parent796b3fb28057948ea5b98f7eb0c0f3af6a1e276e
gh-118221: Always use the default row factory in sqlite3.iterdump() (#118223)

sqlite3.iterdump() depends on the row factory returning resulting rows
as tuples; it will fail with custom row factories like for example a
dict factory.

With this commit, we explicitly reset the row factory of the cursor used
by iterdump(), so we always get predictable results. This does not
affect the row factory of the parent connection.

Co-authored-by: Mariusz Felisiak <felisiak.mariusz@gmail.com>
Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>
Lib/sqlite3/dump.py
Lib/test/test_sqlite3/test_dump.py
Misc/NEWS.d/next/Library/2024-04-24-12-29-33.gh-issue-118221.2k_bac.rst [new file with mode: 0644]