K4/zig:fix_doc into master
Resolves https://github.com/ziglang/zig/issues/21101
Resolves #35900
I also found that in src/Zir.zig sometimes the word Implement is used, but othertimes the word Implements is used. Is this for a reason?
Edit: I opened an issue with a more detailed write-up: #36064.
I marked this WIP because I want to backticks and single quotes in some missing places.
I'm pretty sure that I should add them to doc comments, but what about regular comments and stuff that gets printed to stdout or stderr?
There are only a few locations were they are added already (maybe I missed some):
https://codeberg.org/ziglang/zig/src/branch/master/test/standalone/build.zig#L83
https://codeberg.org/ziglang/zig/src/branch/master/test/behavior/typename.zig#L10
https://codeberg.org/ziglang/zig/src/branch/master/src/main.zig#L4980-L4981
https://codeberg.org/ziglang/zig/src/branch/master/test/tests.zig#L2534
https://codeberg.org/ziglang/zig/src/branch/master/test/tests.zig#L2557
https://codeberg.org/ziglang/zig/src/branch/master/test/tests.zig#L2573
https://codeberg.org/ziglang/zig/src/branch/master/test/standalone/build.zig#L108
b46d45c12c
b9bbeaee3c
This is now ready for review after the force-push, but this question remains:
I also found that in
src/Zir.zigsometimes the wordImplementis used, but othertimes the wordImplementsis used. Is this for a reason?
@ -1,4 +1,5 @@
// zig run -O ReleaseFast --zig-lib-dir ../.. benchmark.zig
// cd ..
// zig run -OReleaseFast --zig-lib-dir .. Random/benchmark.zig
--zig-lib-dir is removed so this is still wrong
No, --zig-lib-dir was removed for the build subcommand. Iniatialy, I thought that I need to set ZIG_LIB_DIR, but I found it isn't necessary.
Closing because it has problems and a third run would be a waste of CI resources. Thanks for the effort but please re-read #31297.
@alichraghi wrote in #35985 (comment):
Thanks for the effort but please re-read #31297.
This PR is about semantic errors and redundancies, not spelling errors.
Edit: !35985 (commit dba8487721) could be considered to be about spelling errors, but for example adding @ to memset builtins in not trivial since it is used in other ways (lib/c/string.zig#L12).
I don't want to run CI on this PR because it doesn't change code, but @andrewrk wrote in #30875 (comment):
I actually prefer for nobody to use this
[skip ci]feature, because the PR queue is supposed to be for stuff that is ready to be merged. If your push is not expected to pass the CI, then simply don't push it! If your PR is not ready to be merged, then please work on it locally or in a fork, and then open a PR when it is ready. I don't think there is ever a reason to use[skip ci]. I would disable that feature on forgejo if I could.
I personally think this feature is useful for PRs like this one.
No due date set.
No dependencies set.
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?