Bug#879176: marked as done (transition: lz4)
Your message dated Thu, 12 Apr 2018 19:23:20 +0200
with message-id <dd691eb8-d636-65bb-7cdb-4085efe91dbf@debian.org>
and subject line Re: Bug#879176: transition: lz4
has caused the Debian Bug report #879176,
regarding transition: lz4
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.)
--
879176: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=879176
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
- To: Debian Bug Tracking System <submit@bugs.debian.org>
- Subject: transition: lz4
- From: Nobuhiro Iwamatsu <iwamatsu@debian.org>
- Date: Fri, 20 Oct 2017 12:32:06 +0900
- Message-id: <CABMQnVKDS68XdaAWSR-JRrV3dc8ucjTdPB7MbNLZahrq0Kvf0Q@mail.gmail.com>
Package: release.debian.org
Severity: normal
User: release.debian.org@packages.debian.org
Usertags: transition
Dear release team,
I'd like to push lz4 1.8.0 into unstable.
The library packge name has not changed since the previous version. However,
some API have been removed . Therefore, lz4 1.8.0 requires transition
processing,
I think.
-----
index f0ecbd4..ce4cf98 100644
--- a/debian/liblz4-1.symbols
+++ b/debian/liblz4-1.symbols
@@ -5,6 +5,7 @@ liblz4.so.1 liblz4-1 #MINVER#
LZ4F_compressFrame@Base 0.0~r127
LZ4F_compressFrameBound@Base 0.0~r127
LZ4F_compressUpdate@Base 0.0~r127
+ LZ4F_compressionLevel_max@Base 1.8.0
LZ4F_createCompressionContext@Base 0.0~r127
LZ4F_createDecompressionContext@Base 0.0~r127
LZ4F_decompress@Base 0.0~r127
@@ -13,7 +14,9 @@ liblz4.so.1 liblz4-1 #MINVER#
LZ4F_freeDecompressionContext@Base 0.0~r127
LZ4F_getErrorName@Base 0.0~r127
LZ4F_getFrameInfo@Base 0.0~r127
+ LZ4F_getVersion@Base 1.7.5
LZ4F_isError@Base 0.0~r127
+ LZ4F_resetDecompressionContext@Base 1.8.0
LZ4_compress@Base 0.0~r113
LZ4_compressBound@Base 0.0~r114
LZ4_compressHC2@Base 0.0~r113
@@ -37,8 +40,6 @@ liblz4.so.1 liblz4-1 #MINVER#
LZ4_compress_fast@Base 0.0~r130
LZ4_compress_fast_continue@Base 0.0~r130
LZ4_compress_fast_extState@Base 0.0~r130
- LZ4_compress_fast_force@Base 0.0~r130
- LZ4_compress_forceExtDict@Base 0.0~r119
LZ4_compress_limitedOutput@Base 0.0~r113
LZ4_compress_limitedOutput_continue@Base 0.0~r113
LZ4_compress_limitedOutput_withState@Base 0.0~r113
@@ -54,7 +55,6 @@ liblz4.so.1 liblz4-1 #MINVER#
LZ4_decompress_fast_withPrefix64k@Base 0.0~r113
LZ4_decompress_safe@Base 0.0~r113
LZ4_decompress_safe_continue@Base 0.0~r119
- LZ4_decompress_safe_forceExtDict@Base 0.0~r127
LZ4_decompress_safe_partial@Base 0.0~r113
LZ4_decompress_safe_usingDict@Base 0.0~r119
LZ4_decompress_safe_withPrefix64k@Base 0.0~r113
@@ -80,15 +80,4 @@ liblz4.so.1 liblz4-1 #MINVER#
LZ4_uncompress@Base 0.0~r114
LZ4_uncompress_unknownOutputSize@Base 0.0~r114
LZ4_versionNumber@Base 0.0~r122
- XXH32@Base 0.0~r127
- XXH32_createState@Base 0.0~r127
- XXH32_digest@Base 0.0~r127
- XXH32_freeState@Base 0.0~r127
- XXH32_reset@Base 0.0~r127
- XXH32_update@Base 0.0~r127
- XXH64@Base 0.0~r127
- XXH64_createState@Base 0.0~r127
- XXH64_digest@Base 0.0~r127
- XXH64_freeState@Base 0.0~r127
- XXH64_reset@Base 0.0~r127
- XXH64_update@Base 0.0~r127
+ LZ4_versionString@Base 1.8.0
-----
I tested building packages that depend on lz4 1.8.0.
apt
Build OK
borgbackup
Build OK
c-blosc
Build OK
dovecot
Build OK
groonga
Build OK
hhvm
Build OK
hilive
Build OK
imagevis3d
Build OK
libarchive
Build OK
librdkafka
Build OK
mtbl
FTBFS, Fixed in experimental.
Depend auto-mtbl transition.
mysql-5.7
FTBFS, not in testing. We can ignore this.
ocserv
Build OK
openconnect
Build OK
openvpn
Build OK
php-horde-lz4
Build OK
pytables
Build OK
python-lz4
Build OK
ros-ros-comm
Build OK
sambamba
FTBFS, ldc issue #878837. We can ignore this.
spice-gtk
Build OK
squashfs-tools
Build OK
systemd
Build OK
terminology
Build OK
spice
Build OK
wireshark
Build OK
Ben file:
title = "lz4";
is_affected = .depends ~ "/liblz4-1 \(>=/";
is_good = .depends ~ "/liblz4-1 \(>= 1.8.0\)/";
is_bad = .depends ~ "/liblz4-1 \(>= 0.0~r*\)/";
-- System Information:
Debian Release: buster/sid
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Kernel: Linux 4.11.0-2-amd64 (SMP w/8 CPU cores)
Locale: LANG=ja_JP.UTF-8, LC_CTYPE=ja_JP.UTF-8 (charmap=UTF-8),
LANGUAGE=ja_JP.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
--
Nobuhiro Iwamatsu
iwamatsu at {nigauri.org / debian.org}
GPG ID: 40AD1FA6
--- End Message ---
--- Begin Message ---
On 12/04/18 00:39, Nobuhiro Iwamatsu wrote:
> Hi, all.
>
> 2017-10-23 10:04 GMT+09:00 Nobuhiro Iwamatsu <iwamatsu@debian.org>:
>> Hi, all.
>>
>> Thanks for your comment.
>>
>> 2017-10-22 1:37 GMT+09:00 Emilio Pozuelo Monfort <pochu@debian.org>:
>>> Control: tags -1 moreinfo
>>>
>>> On 20/10/17 11:13, Mattia Rizzolo wrote:
>>>> On Fri, Oct 20, 2017 at 12:32:06PM +0900, Nobuhiro Iwamatsu wrote:
>>>>> I'd like to push lz4 1.8.0 into unstable.
>>>>> The library packge name has not changed since the previous version. However,
>>>>> some API have been removed . Therefore, lz4 1.8.0 requires transition
>>>>> processing,
>>>>
>>>> Since, as the symbols diff you attached, this version removes symbols,
>>>> therefore breaking the ABI, shouldn't the SONAME be changed as well,
>>>> together with the package name (i.e. liblz4-1 → liblz4-2).
>>>
>>> Yes, if symbols are removed, then the ABI is broken and the library needs to
>>> bump the SONAME. Worst case, if upstream doesn't want to bump it, you can rename
>>> the package to e.g. liblz4-1a, with Conflicts/Breaks/Replaces against liblz4-1.
>>> The SONAME bump is preferred though so that liblz4-1 and liblz4-2 are
>>> co-installable, which eases both the transition and upgrades.
>>
>> I see. I undetstood.
>>
>>>
>>> We could ignore the fact that symbols were removed IF those symbols weren't part
>>> of the public ABI (e.g. they weren't exported in the public headers). Which
>>> seems to be the case here from a quick grep, but I don't know this library well
>>> enough to affirm that...
>>
>> OK, I will check the dependencies of the library and the software that uses it.
>>
>
> I checked the binary and source code of packages that depend on lz4 and
> I confirmed that they do not affect the ABI changed this time.
> Therefore, we do not need to do transition.
> And Julian's comment, we can see that we do not need to do this.
> Thanks, Julian.
Alright, let's close this transition bug then.
Cheers,
Emilio
--- End Message ---
Reply to: