generalize utf8 decoding error handling to any field specified as bytess in converters
Closes T435.
Details
Details
- Reviewers
olasd - Group Reviewers
Reviewers - Maniphest Tasks
- T435: Fail gracefully if the revision decoding process fails
- Commits
- R65:110ba85e2ae6: templates: show if decoding errors occurred
R65:05d37892bc8a: converters: add an extra step to byte decoding in case they are non-utf8 bytes
R65:65aaf922a2f7: converters: byte decoding now tracks which fields generated an error
rDWAPPS110ba85e2ae6: templates: show if decoding errors occurred
rDWAPPS65aaf922a2f7: converters: byte decoding now tracks which fields generated an error
rDWAPPS05d37892bc8a: converters: add an extra step to byte decoding in case they are non-utf8 bytes
Diff Detail
Diff Detail
- Repository
- rDWAPPS Web applications
- Branch
- T435
- Lint
No Linters Available - Unit
No Unit Test Coverage - Build Status
Buildable 175 Build 265: Software Heritage Python tests Build 264: arc lint + arc unit
Event Timeline
Comment Actions
Hey,
You should somehow mark what fields failed to decode and got escaped; in the current implementation there is no way to distinguish between
some_input = {
'a': 'something',
'b': 'someone',
'c': b'a name \xff',
'd': b'an email \xff',
}and
some_input = {
'a': 'something',
'b': 'someone',
'c': b'a name \\xff',
'd': b'an email \\xff',
}