icu4c 74.2

[p-linnane]

• Have you followed the guidelines for contributing?
• Have you ensured that your commits follow the commit style guide?
• Have you checked that there aren't other open pull requests for the same formula update/change?
• Have you built your formula locally with HOMEBREW_NO_INSTALL_FROM_API=1 brew install --build-from-source <formula>, where <formula> is the name of the formula you're submitting?
• Is your test running fine brew test <formula>, where <formula> is the name of the formula you're submitting?
• Does your build pass brew audit --strict <formula> (after doing HOMEBREW_NO_INSTALL_FROM_API=1 brew install --build-from-source <formula>)? If this is a new formula, does it pass brew audit --new <formula>?

---

[Bo98]

Promising initial run.

Linux:

• gspell can't seem to find python3 (via glib I think).
• manticoresearch sha256 error.
• planck fails because it looks like webkitgtk needs a revision bump.
• qt (Chromium) is incompatible with ICU 74 due to the addition of five new line break types with Unicode 15.1. There'll probably be a patch soon, if not already. If it hasn't by the time we've fixed everything else here, we can patch ourselves e.g.:

#if U_ICU_VERSION_MAJOR_NUM >= 74
#define BA_LB_COUNT (U_LB_COUNT - 8)
...

• simdutf timed out for some reason - might need a lighter test or maybe it's signalling a genuine problem.
• tepl also can't find python3 (via glib too I think)
• vte3 also can't find python3 - I reckon there's something wrong with glib and we'll need to fix that before proceeding here.
• zobra maybe segfaulting

macOS:

• manticoresearch sha256 error.
• texlive - Perl mismatch somewhere: Parser.c: loadable library and perl binaries are mismatched (got first handshake key 0x10380080, needed 0x10200080)

Any non-disabled PHP versioned formula also needs revision bumping here: [email protected] and [email protected].

(Haven't checked dependent tests yet as they're still running)

[Porkepix]

Any non-disabled PHP versioned formula also needs revision bumping here: [email protected] and [email protected].

Note that active support for [email protected] is already gone, and security support will also be on 26th of November. Dunno if it calls for immediate deprecation/disabling, but it might save some time on it depending on when this PR is finalized.

[Bo98]

It'll probably just be a case of we'll do it if this PR ships before that date, and not if it doesn't. PHP's light enough to only take 10-20 minutes to build.

Focus first is fixing the other things though.

[iMichka]

Fix for gspell: #154128

[Bo98]

Looks like none of the PHP formulae are being revision bumped here. 8.0 isn't a concern anymore but anything newer, including the latest, is.

[iMichka]

Linux failures:

2023-11-26T22:49:55.1656031Z Warning: 2 failed steps ignored!
2023-11-26T22:49:55.1658020Z brew test --verbose simdutf
2023-11-26T22:49:55.1658927Z brew install --verbose --build-bottle zorba
2023-11-26T22:49:55.1666147Z Error: 3 failed steps!
2023-11-26T22:49:55.1666866Z brew test --verbose yaz
2023-11-26T22:49:55.1667427Z brew install --verbose --build-bottle planck
2023-11-26T22:49:55.1668176Z brew install --verbose --build-bottle qt

[cho-m]

Ignoring missing revision bumps and failures I already fixed in referenced PRs, the main failures are:

• qt - another libxml2 issue. Sadly 2.12 introduced a fair amount of breakage. Probably should try to fix outside this PR either in qt 6.6.1 #155921 or a separate PR. I think issue is fixed upstream of upstream (Chromium) but haven't checked what version newer qt is using.
• texlive - still the Perl mismatch issue when building HTML::TreeBuilder (with mismatch coming from previous resource HTML::Parser which is dependency?). I see we have a separate PR to work on this texlive: add Test::Fatal dependency and rebuild #155568.
  • EDIT: Also trying out texlive: fix build with Perl 5.38 #156019

---

Outside those, the rest look like flaky dependent tests or other unrelated issues:

• [email protected] (all macOS - skipped on Linux) - timing? Looks like we already added a timeout recently. Hard to tell exact issue as we run in background so logs get sent to separate file. On side note, it looks like retry test doesn't work as the previous test's process is still active.
  • Just increasing sleep some more in hadoop [email protected]: increase test sleep #156049
• hadoop (14-x86_64) - timing?
  • Just increasing sleep some more in hadoop [email protected]: increase test sleep #156049
• libosinfo (13-x86_64) - libosinfo: reduce test output to help avoid timeout #156053
• fastqc (14-arm64) - maybe bad JDK. Not sure on state of OpenJDK for Sonoma given other reported issues.
• stellar-core (13-x86_64)
• localtunnel (14-arm64, 13-arm64)

[cho-m]

Added some extra revision bumps.

Dealing with patches for qt in separate PR #156065.

Will need another rebase before running.

[cho-m]

Linux failed to build webkitgtk, planck (due to webkitgtk) , and swift:

• webkitgtk - another libxml2 issue - WebKit/WebKit@1bad176 - webkitgtk 2.42.3 #156786

  2023-12-03T00:43:41.6692807Z /tmp/webkitgtk-20231202-188180-corvlx/webkitgtk-2.40.5/Source/WebCore/xml/XSLStyleSheetLibxslt.cpp:143:106: error: invalid conversion from ‘void (*)(void*, xmlError*)’ {aka ‘void (*)(void*, _xmlError*)’} to ‘xmlStructuredErrorFunc’ {aka ‘void (*)(void*, const _xmlError*)’} [-fpermissive]
  2023-12-03T00:43:41.6695985Z   143 |     XMLDocumentParserScope scope(cachedResourceLoader(), XSLTProcessor::genericErrorFunc, XSLTProcessor::parseErrorFunc, console);
  2023-12-03T00:43:41.6697143Z       |                                                                                           ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~
  2023-12-03T00:43:41.6697780Z       |                                                                                                          |
  2023-12-03T00:43:41.6698650Z       |                                                                                                          void (*)(void*, xmlError*) {aka void (*)(void*, _xmlError*)}
  

• swift - it's accidentally picking up [email protected] from a dependency and failing without python-setuptools. We should use an argument to make sure the Python matches dependency we set:

  2023-12-03T17:53:20.9369231Z FAILED: lib/libc++.a 
  2023-12-03T17:53:20.9392590Z : && /home/linuxbrew/.linuxbrew/Cellar/cmake/3.27.9/bin/cmake -E rm -f lib/libc++.a && /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/bin/llvm-ar Dqc lib/libc++.a  libcxx/src/CMakeFiles/cxx_static.dir/algorithm.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/any.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/assert.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/atomic.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/barrier.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/bind.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/charconv.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/chrono.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/condition_variable.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/condition_variable_destructor.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/exception.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/functional.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/future.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/hash.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/legacy_pointer_safety.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/memory.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/mutex.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/mutex_destructor.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/new.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/optional.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/random_shuffle.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/ryu/d2fixed.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/ryu/d2s.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/ryu/f2s.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/shared_mutex.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/stdexcept.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/string.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/system_error.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/thread.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/typeinfo.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/utility.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/valarray.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/variant.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/vector.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/debug.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/legacy_debug_handler.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/random.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/ios.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/ios.instantiations.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/iostream.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/locale.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/regex.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/strstream.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/format.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/filesystem/operations.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/filesystem/directory_iterator.cpp.o libcxx/src/CMakeFiles/cxx_static.dir/filesystem/int128_builtins.cpp.o && /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/bin/llvm-ranlib -D lib/libc++.a && cd /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/tools/clang/runtime/compiler-rt-bins/lib/fuzzer/libcxx_fuzzer_x86_64/libcxx/src && /home/linuxbrew/.linuxbrew/bin/python3.12 /tmp/swift-20231203-1836061-etto1i/llvm-project/libcxx/utils/merge_archives.py -o /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/tools/clang/runtime/compiler-rt-bins/lib/fuzzer/libcxx_fuzzer_x86_64/lib/libc++.a --ar /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/bin/llvm-ar /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/tools/clang/runtime/compiler-rt-bins/lib/fuzzer/libcxx_fuzzer_x86_64/lib/libc++.a /tmp/swift-20231203-1836061-etto1i/build/llvm-linux-x86_64/tools/clang/runtime/compiler-rt-bins/lib/fuzzer/libcxx_fuzzer_x86_64/lib/libc++abi.a
  2023-12-03T17:53:20.9392949Z Traceback (most recent call last):
  2023-12-03T17:53:20.9393908Z   File "/tmp/swift-20231203-1836061-etto1i/llvm-project/libcxx/utils/merge_archives.py", line 11, in <module>
  2023-12-03T17:53:20.9394097Z     import distutils.spawn
  2023-12-03T17:53:20.9394458Z ModuleNotFoundError: No module named 'distutils'
  

[Porkepix]

74.2 is out.

[p-linnane]

Thanks to everyone who contributed to the effort to get this version bump merged!

[chenrui333]

Awesome work! 💯

[cho-m]

Follow ups:

• percona-xtrabackup - percona-xtrabackup: revision bump (icu4c 74.2) #164240
• libspelling - libspelling: revision bump (icu4c 74.2) #164241
• apngasm - apngasm: revision bump (icu4c 74.2) #164242
• freeling - freeling: revision bump (icu4c 74.2) #164243
• ncmpcpp - ncmpcpp: revision bump (icu4c 74.2) #164245
• mapnik - mapnik: revision bump (icu4c 74.2) #164246
• qt@5 - qt@5: revision bump (icu4c 74.2) #164247