docs: compiler.find_library: discourage lib prefix #10838
Conversation
|
CI failure is very relevant. |
The docs generator was previously restricted to ASCII, preventing the use of nicer typography characters. Since ASCII is a subset of UTF-8, expanding should be safe.
|
I do not want you to use unicode in the docs, NACK to the first commit. :) |
|
unicode is already used in the docs, for instance in the reference tables. No idea why we aren't just using |
|
Idk about that, but I do know that frivolously using So anything that catches them and makes them error out is good in my book. Now if only we could remove them from the Unicode standard... I think in general if your documentation relies on Unicode then you need to do some deep introspection on why you think you need it and whether that's in fact true. It probably isn't. |
|
|
|
I also hate unicode quotes, but I don't think automatic detection is a reason to forbid all of unicode. That is just a reason to force use of regular quotes |
…ut it
We need to support cases where the library might be called "foo.so" and
therefore we check for exact matches too. But this also allows
`cc.find_library('libfoo')` to find libfoo.so, which is strange and
won't work in many cases. Emit a warning when this happens.
Fixes mesonbuild#10838
|
Anyway this approach is totally wrong, I submitted a PR to do this correctly, by warning about it instead of burying it in docs where no one will see it. |
…ut it
We need to support cases where the library might be called "foo.so" and
therefore we check for exact matches too. But this also allows
`cc.find_library('libfoo')` to find libfoo.so, which is strange and
won't work in many cases. Emit a warning when this happens.
Fixes #10838
3 participants
I continue seeing it in the wild.