Skip to content

Fixes issue on MacOS arm chip with type detection - #128

Merged
MisterGC merged 5 commits into
ndsev:release/1.3from
Hsilgos:fix_any_macos_m1
Oct 17, 2026
Merged

MisterGC merged 5 commits into
ndsev:release/1.3from
Hsilgos:fix_any_macos_m1

Conversation

@Hsilgos

@Hsilgos Hsilgos commented Oct 6, 2026

Copy link
Copy Markdown
Contributor

Closes #127

@MisterGC
MisterGC changed the base branch from master to release/1.3 October 16, 2026 13:39
@MisterGC

MisterGC commented Oct 16, 2026 •

Copy link
Copy Markdown

Thanks @Hsilgos for the fix and the detailed write-up in #127! I could reproduce the problem on an arm64 Mac, and your change fixes it.

To get it into the 1.3 release, I've pushed two small commits to your branch and moved the PR to release/1.3:

  • The name comparison now only runs where it's needed: libc++ on arm64 Apple. Everywhere else, isType() uses the plain == as before. Without that limit, two different types with the same name (for example namespace { struct Foo {}; } in two source files) were treated as the same type, also on Linux.
  • A test for that case: AnyTest.sameNamedAnonymousTypes. It's skipped on arm64 Apple, where comparing names can't tell these types apart; the comment in Any.h mentions this.
  • Housekeeping: the missing #include <cstring> and the shifted clang-tidy suppression lines.

On macOS and iOS arm64 everything works exactly as with your original change. Thanks again!

@MisterGC MisterGC mentioned this pull request Oct 16, 2026
@MisterGC
MisterGC merged commit 943ac4f into ndsev:release/1.3 Oct 17, 2026
25 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Any type still doesn't work properly on arm64 Apple.

2 participants