Previously, when running the Django test suite with warnings enabled,
the following was emitted:
/usr/lib64/python3.7/urllib/parse.py:915: BytesWarning: str() on a bytearray instance
v = quote_via(str(v), safe, encoding, errors)
This occurred due to the bytearray() being passed to
urllib.parse.urlencode() which eventually calls str() on it. The test
does not represent desired real world behavior. Rather than test for and
assert strange unspecified behavior that emits a warning, remove it.
This was also discussed in PR #11374.
email.headerregistry.parser.get_mailbox() returns a token with a `token_type` attribute.
If `token_type` is `’invalid-mailbox’` then RFC violations have been detected. Emails with only the local part, and no domain, are correctly parsed but are marked as `’invalid-mailbox’`.
As per #15042, local-only are supported, to enable sending to addresses on localhost.
sanitize_email() does not currently check `token_type`. This test is added to avoid a regression in case this is revisited in the future.
intword and filesizeformat passed floats to ngettext() which is
deprecated in Python 3.7. The rationale for this warning is documented
in BPO-28692: https://bugs.python.org/issue28692.
For filesizeformat, the filesize value is expected to be an int -- it
fills %d string formatting placeholders. It was likely coerced to a
float to ensure floating point division on Python 2. Python 3 always
does floating point division, so coerce to an int instead of a float to
fix the warning.
For intword, the number may contain a decimal component. In English, a
decimal component makes the noun plural. A helper function,
round_away_from_one(), was added to convert the float to an integer that
is appropriate for ngettext().
Filters in annotations crashed when used with numerical-type
aggregations (i.e. Avg, StdDev, and Variance). This was caused as the
source expressions no not necessarily have an output_field (such as the
filter field), which lead to an AttributeError: 'WhereNode' object has
no attribute output_field.
Thanks to Chuan-Zheng Lee for the report.
Regression in c690afb873 and two following
commits.
Previously we updated the file mtimes if the file has not been seen
before - i.e on the first iteration of the loop.
If the mtime has been changed we triggered the notify_file_changed()
method which in all cases except the translations will result in the
process being terminated. To be strictly correct we need to update the
mtime for either branch of the conditional.
Regression in 6754bffa2b.
Mariusz Felisiak
Andrew Godwin
Chris Jerdonek
Jon Dufresne
can
Min ho Kim
Sanyam Khurana
Carlton Gibson
Joachim Jablon
CruxBox
Aymeric Augustin
Hasan Ramezani
Alexandre Varas
Tobias Bengfort
Étienne Beaulé
Tim Graham
Jones Ambrosi
RobertAKARobin
Nathan Gaberel
Tom Forbes
Caio Ariede
Ivor Bosloper
Mattia Procopio
Johan Lübcke