gh-92455: Respect case-sensitive mimetype suffixes#148782
Conversation
…sensitive-add-type
…sensitive-add-type
…sensitive-add-type
ifrh
left a comment
ifrh
left a comment
There was a problem hiding this comment.
I cannot approve the changes formaly, but they look good for me.
| eq(self.db.guess_type("scheme:foobar.tar.z"), (None, None)) | ||
|
|
||
| def test_added_types_case_sensitive_preferred(self): | ||
| self.db.add_type("text/x-r-script", ".R") |
There was a problem hiding this comment.
"text/x-r-script" could be added to the database in future. It is better to use something less likely to be added.
Please add also tests for strict=False.
There was a problem hiding this comment.
Thanks for the review. I changed the test type and added tests for strict=False.
|
#30229 changed also the code related to while True:
if ext in self.suffix_map:
base, ext = posixpath.splitext(base + self.suffix_map[ext])
continue
ext_lower = ext.lower():
if ext_lower in self.suffix_map:
base, ext = posixpath.splitext(base + self.suffix_map[ext_lower])
continue
break@kumaraditya303, what are your thoughts? |
…sensitive-add-type # Conflicts: # Lib/test/test_mimetypes.py
I think we should do the same for suffix_map i.e. first try case sensitively then case-insensitively. |
|
Would you prefer handling |
|
It is better to handle all in one PR. Also please update the documentation. |
|
Added a failing test, then updated |
Documentation build overview
193 files changed ·
|
…sensitive-add-type
4 participants
Fixes gh-92455.
As described in the issue, the current implementation cannot find uppercase
extensions added via
mimetypes.add_type(). This appears to be a regression from gh-30229.This restores the lookup order documented for
mimetypes.guess_type():type suffixes are tried case-sensitively first, then case-insensitively.