[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#960193: marked as done (transition: icu)

Your message dated Sat, 11 Jul 2020 12:13:19 +0200
with message-id <20200711101319.GA1954528@ramacher.at>
and subject line Re: Bug#960193: transition: icu
has caused the Debian Bug report #960193,
regarding transition: icu
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
960193: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=960193
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: release.debian.org
Severity: normal
User: release.debian.org@packages.debian.org
Usertags: transition

Hi RMs,

Quite a big transition of ICU I would like to start.
The build test is done on amd64 only and has the following FTBFS reasons.
Level 1:
casparcg-server: linking problem with Boost, already reported and it's
sid only already due to this,
dino-im tied to a specific version of libsignal-protocol-c for an
unknown reason:
CMake Error at /usr/share/cmake-3.16/Modules/FindPackageHandleStandardArgs.cmake:146
(message):
  Could NOT find SignalProtocol: Found unsuitable version "2.3.3", but
  required is exact version "2.3.2" (found
  /usr/lib/x86_64-linux-gnu/libsignal-protocol-c.so)
But libsignal-protocol-c 2.3.3-1 was uploaded recently as the CMake
error shows that.

freshplayerplugin: missing build dependency of libdrm which is known,
filed and due to this sid only already.
ledger: fails with the current ICU release in Sid as well, hence
unrelated to the new version:
        226 - RegressTest_1057 (Failed)

libvmime: this is ICU specific and upstream has a patch[1].
simpleitk: self-test failures, known, reported and due to these
already sid only.

Level 1.5 is needed for Boost packages to be binNMUed:
Both boost1.67 and boost1.71 build correctly.

Level 2:
clickhouse: has a missing build dependcy on tzdata with that it
builds, otherwise:
Poco::Exception. Code: 1000, e.code() = 0, e.displayText() =
Exception: Could not determine time zone from TZ variable value:
`Europe/Moscow': boost::filesystem::canonical: No such file or
directory: "/usr/share/zoneinfo/", e.what() = Exception

mongo-c-driver: seems to be Sphinx related: didn't check without the old ICU:
[Sphinx]
Warning, treated as error:
the taglist extension is not safe for parallel reading
Not creating man page for index

mozjs68: fails with the current ICU release in Sid as well, seems lz4 broke it:
/usr/bin/ld: ../Unified_cpp_mfbt0.o: in function `mozPoisonValueInit':
./debian/build/mfbt/./mfbt/Poison.cpp:120: undefined reference to `sysconf'
/usr/bin/ld: ../Unified_cpp_mfbt0.o: in function `mozilla::RandomUint64()':
./debian/build/mfbt/./mfbt/RandomNum.cpp:118: undefined reference to `syscall'
/usr/bin/ld: ./debian/build/mfbt/./mfbt/RandomNum.cpp:136: undefined
reference to `close'
/usr/bin/ld: ../lz4.o: in function `LZ4_createStream':
./debian/build/mfbt/./mfbt/lz4.c:1288: undefined reference to `malloc'
/usr/bin/ld: ../lz4.o: in function `LZ4_freeStream':
./debian/build/mfbt/./mfbt/lz4.c:1336: undefined reference to `free'
/usr/bin/ld: ../lz4.o: in function `LZ4_createStreamDecode':
./debian/build/mfbt/./mfbt/lz4.c:2075: undefined reference to `calloc'
/usr/bin/ld: ../lz4.o: in function `LZ4_freeStreamDecode':
./debian/build/mfbt/./mfbt/lz4.c:2083: undefined reference to `free'
/usr/bin/ld: TestAlgorithm: hidden symbol `syscall' isn't defined
/usr/bin/ld: final link failed: bad value

zimlib: due to the new ICU package release (updating the symbols file
make it builds):
dpkg-gensymbols: error: some symbols or patterns disappeared in the
symbols file: see diff output below
[ICU related symbols]

Level 2.5:
haskell-text-icu needs to be binNMUed first for level 3 Haskell packages.

Level 3:
chromium: embeds the V8 JavaSript engine which FTBFS:
[37842/38456] CXX obj/v8/v8_base_without_compiler/js-number-format.o
../../v8/src/objects/js-number-format.cc:1223:17: error: no member
named 'getAllFieldPositions' in 'icu_67::number::FormattedNumber'
      formatted.getAllFieldPositions(*fp_iter, status);
      ~~~~~~~~~ ^
../../v8/src/objects/js-number-format.cc:1238:17: error: no member
named 'getAllFieldPositions' in 'icu_67::number::FormattedNumber'
      formatted.getAllFieldPositions(*fp_iter, status);
      ~~~~~~~~~ ^
2 errors generated.

These seems to be fixed in its upstream[2] tree with other patches.
The problem is that I couldn't backport those into the current
Chromium package in Sid. It's two security releases behind of
upstream. Seems there's no enough manpower to maintain that package.
Problably that's the reason it was removed from Ubuntu, I'm not sure.
Tried to package the new upstream release but failed for the first try
and didn't tried further.

Level 4 to level 7 are all OK.

Thanks for considering,
Laszlo/GCS
[1] https://github.com/kisli/vmime/commit/e96aeeb14dc51deeea70e6fdffa95f80af78fdfc
[2] https://github.com/v8/v8/commit/3f8dc4b2e5baf77b463334c769af85b79d8c1463

--- End Message ---
--- Begin Message --- 
On 2020-05-10 15:40:43 +0200, László Böszörményi wrote:
> Package: release.debian.org
> Severity: normal
> User: release.debian.org@packages.debian.org
> Usertags: transition
> 
> Hi RMs,
> 
> Quite a big transition of ICU I would like to start.
> The build test is done on amd64 only and has the following FTBFS reasons.
> Level 1:
> casparcg-server: linking problem with Boost, already reported and it's
> sid only already due to this,
> dino-im tied to a specific version of libsignal-protocol-c for an
> unknown reason:
> CMake Error at /usr/share/cmake-3.16/Modules/FindPackageHandleStandardArgs.cmake:146
> (message):
>   Could NOT find SignalProtocol: Found unsuitable version "2.3.3", but
>   required is exact version "2.3.2" (found
>   /usr/lib/x86_64-linux-gnu/libsignal-protocol-c.so)
> But libsignal-protocol-c 2.3.3-1 was uploaded recently as the CMake
> error shows that.
> 
> freshplayerplugin: missing build dependency of libdrm which is known,
> filed and due to this sid only already.
> ledger: fails with the current ICU release in Sid as well, hence
> unrelated to the new version:
>         226 - RegressTest_1057 (Failed)
> 
> libvmime: this is ICU specific and upstream has a patch[1].
> simpleitk: self-test failures, known, reported and due to these
> already sid only.
> 
> Level 1.5 is needed for Boost packages to be binNMUed:
> Both boost1.67 and boost1.71 build correctly.
> 
> Level 2:
> clickhouse: has a missing build dependcy on tzdata with that it
> builds, otherwise:
> Poco::Exception. Code: 1000, e.code() = 0, e.displayText() =
> Exception: Could not determine time zone from TZ variable value:
> `Europe/Moscow': boost::filesystem::canonical: No such file or
> directory: "/usr/share/zoneinfo/", e.what() = Exception
> 
> mongo-c-driver: seems to be Sphinx related: didn't check without the old ICU:
> [Sphinx]
> Warning, treated as error:
> the taglist extension is not safe for parallel reading
> Not creating man page for index
> 
> mozjs68: fails with the current ICU release in Sid as well, seems lz4 broke it:
> /usr/bin/ld: ../Unified_cpp_mfbt0.o: in function `mozPoisonValueInit':
> ./debian/build/mfbt/./mfbt/Poison.cpp:120: undefined reference to `sysconf'
> /usr/bin/ld: ../Unified_cpp_mfbt0.o: in function `mozilla::RandomUint64()':
> ./debian/build/mfbt/./mfbt/RandomNum.cpp:118: undefined reference to `syscall'
> /usr/bin/ld: ./debian/build/mfbt/./mfbt/RandomNum.cpp:136: undefined
> reference to `close'
> /usr/bin/ld: ../lz4.o: in function `LZ4_createStream':
> ./debian/build/mfbt/./mfbt/lz4.c:1288: undefined reference to `malloc'
> /usr/bin/ld: ../lz4.o: in function `LZ4_freeStream':
> ./debian/build/mfbt/./mfbt/lz4.c:1336: undefined reference to `free'
> /usr/bin/ld: ../lz4.o: in function `LZ4_createStreamDecode':
> ./debian/build/mfbt/./mfbt/lz4.c:2075: undefined reference to `calloc'
> /usr/bin/ld: ../lz4.o: in function `LZ4_freeStreamDecode':
> ./debian/build/mfbt/./mfbt/lz4.c:2083: undefined reference to `free'
> /usr/bin/ld: TestAlgorithm: hidden symbol `syscall' isn't defined
> /usr/bin/ld: final link failed: bad value
> 
> zimlib: due to the new ICU package release (updating the symbols file
> make it builds):
> dpkg-gensymbols: error: some symbols or patterns disappeared in the
> symbols file: see diff output below
> [ICU related symbols]
> 
> Level 2.5:
> haskell-text-icu needs to be binNMUed first for level 3 Haskell packages.
> 
> Level 3:
> chromium: embeds the V8 JavaSript engine which FTBFS:
> [37842/38456] CXX obj/v8/v8_base_without_compiler/js-number-format.o
> ../../v8/src/objects/js-number-format.cc:1223:17: error: no member
> named 'getAllFieldPositions' in 'icu_67::number::FormattedNumber'
>       formatted.getAllFieldPositions(*fp_iter, status);
>       ~~~~~~~~~ ^
> ../../v8/src/objects/js-number-format.cc:1238:17: error: no member
> named 'getAllFieldPositions' in 'icu_67::number::FormattedNumber'
>       formatted.getAllFieldPositions(*fp_iter, status);
>       ~~~~~~~~~ ^
> 2 errors generated.
> 
> These seems to be fixed in its upstream[2] tree with other patches.
> The problem is that I couldn't backport those into the current
> Chromium package in Sid. It's two security releases behind of
> upstream. Seems there's no enough manpower to maintain that package.
> Problably that's the reason it was removed from Ubuntu, I'm not sure.
> Tried to package the new upstream release but failed for the first try
> and didn't tried further.
> 
> Level 4 to level 7 are all OK.

Since there was no progress on grcompiler, I binNMUed it in tpu.
libicu63 got finally removed from testing.

Cheers
-- 
Sebastian Ramacher

Attachment: signature.asc
Description: PGP signature


--- End Message ---
Reply to: