[MacPorts] #68919: brotli @1.1.0_0+universal: Failed to destroot

MacPorts noreply at macports.org
Sat Dec 16 11:07:28 UTC 2023


#68919: brotli @1.1.0_0+universal: Failed to destroot
-----------------------+-------------------------------
  Reporter:  thetrial  |      Owner:  raimue
      Type:  defect    |     Status:  assigned
  Priority:  Normal    |  Milestone:
 Component:  ports     |    Version:  2.8.1
Resolution:            |   Keywords:  legacy-os, sierra
      Port:  brotli    |
-----------------------+-------------------------------
Changes (by ryandesign):

 * cc: herbygillot (added)


Comment:

 Ok, the log shows merging this file the intended way (using `lipo`) failed
 because the two files `have the same architectures (i386) and can't be in
 the same fat output file`. That shouldn't be, of course. This port
 includes the muniversal portgroup which means it intends to build multiple
 times, once for each architecture. What has actually happened, though, is
 that each of the two builds have built for both architectures already:

 {{{
 % lipo -info brotli/work/destroot-*86*/opt/local/lib/lib*1.1.0.dylib
 Architectures in the fat file:
 brotli/work/destroot-i386/opt/local/lib/libbrotlicommon.1.1.0.dylib are:
 i386 x86_64
 Architectures in the fat file:
 brotli/work/destroot-i386/opt/local/lib/libbrotlidec.1.1.0.dylib are: i386
 x86_64
 Architectures in the fat file:
 brotli/work/destroot-i386/opt/local/lib/libbrotlienc.1.1.0.dylib are: i386
 x86_64
 Architectures in the fat file: brotli/work/destroot-
 x86_64/opt/local/lib/libbrotlicommon.1.1.0.dylib are: x86_64 i386
 Architectures in the fat file: brotli/work/destroot-
 x86_64/opt/local/lib/libbrotlidec.1.1.0.dylib are: x86_64 i386
 Architectures in the fat file: brotli/work/destroot-
 x86_64/opt/local/lib/libbrotlienc.1.1.0.dylib are: x86_64 i386
 }}}

 A probable cause for this is where the log shows the use of
 `-DCMAKE_OSX_ARCHITECTURES="x86_64;i386"`. While we could investigate how
 to remove that, the fact that the build succeeded for both architectures
 simultaneously raises the question about why the muniversal portgroup is
 being used at all. It was just
 [changeset:ceb560896d734cf29b3a7659b3315de76b8a144a/macports-ports added
 to the port yesterday] and apparently it was intended to resolve #64784,
 but obviously doesn't do so correctly.

-- 
Ticket URL: <https://trac.macports.org/ticket/68919#comment:3>
MacPorts <https://www.macports.org/>
Ports system for macOS


More information about the macports-tickets mailing list