------------------------------------------------------------------------------
To reply, visit https://hellosplat.com/s/beanbag/tickets/4988/
------------------------------------------------------------------------------
New update by benjackson
For Beanbag, Inc. > Review Board > Ticket #4988
Reply:
Found a regression with the fix, but we'll get reworked version in for
5.0.2 (which is going out tonight/tomorrow).
`force_text()`/`force_str()` will assume an encoding if the object is a
`bytes` or converts to a `bytes`. The problem with this is that we can't assume
at this particular stage. We instead need to use the configured encoding list.
This matters particularly for people working with some Chinese encodings.
`convert_to_unicode()` is responsible for trying the configured encodings, but
since the fix was passing in the forced-encode, it never got a chance.
That's not a problem for `Revision` instances, but is for raw revisions.
Probably not even a problem there (revisions are generally effectively ASCII in
most SCMs), but it's more of a problem when forcing the encoding for filenames
(which are completely at the whim of the developers committing code).
Making a change to specifically check for and cast the `Revision` before
storing it instead, which should be a bit safer.
--
You received this message because you are subscribed to the Google Groups
"reviewboard-issues" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/reviewboard-issues/20230207032257.30335.48780%40ip-10-1-54-209.ec2.internal.
