cmake: Collect headers for extracting translatable strings #33194
Conversation
|
The following sections might be updated with supplementary metadata relevant to reviewers and maintainers. Code Coverage & BenchmarksFor details see: https://corecheck.dev/bitcoin/bitcoin/pulls/33194. ReviewsSee the guideline for information on the review process. ConflictsReviewers, this pull request conflicts with the following ones:
If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first. |
|
There is nothing wrong with globbing per se. The problem with globbing in the build analysis stage is that it is slows down each individual build. The problem with globbing in the build system generator is that the result may be out of date when the build runs. This is the reason why globbing is discouraged in Since the result of the glob is only used inside a single custom target, the best approach would be to actually perform the globbing inside the script that is run as the custom target. That means: either do it in the python script or add a cmake script. |
|
Closing in favour of #33200. |
While updating the translation source file for Transifex after #31296, I noticed that headers are not processed by the
extract_strings_qt.pyscript.Adding headers to build targets’ sources doesn’t seem like a good solution, because it would be hard to maintain, as CMake does not require the presence of headers in a target’s sources.
Notably, the
translatebuild target differs from others in the following ways:Considering the above, we can safely collect input files for processing by
extract_strings_qt.pyusing thefile(GLOB_RECURSE ...)command.Required for #33193.