[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Libcdio-devel] libcdio-0.91 - libiso9660 version
From: |
Chris Clayton |
Subject: |
[Libcdio-devel] libcdio-0.91 - libiso9660 version |
Date: |
Sun, 15 Dec 2013 12:36:48 +0000 |
User-agent: |
Mozilla/5.0 (X11; Linux i686 on x86_64; rv:17.0) Gecko/20130806 Thunderbird/17.0.8 |
Hi,
I've noticed that the version of libiso9660 produced by building libcdio-0.91
is less than the version produced by
building libcdio-0.83. Whilst I can't claim to fully understand the libtool
versioning scheme, this does seem odd.
Installing the rpm I have built fails because other applications depend on
libiso9660.so.8. Consequently, I looked at
the related guidance notes in lib/iso9660/Makefile.am (and the libtool info
page) and am a bit puzzled about the values
I see in the libiso9660_la_{CURRENT,REVISION,AGE} variables.
In v0.83 the values are CURRENT = 8, REVISION = 0 and AGE= 0. In v0.91 CURRENT
and REVISION are unchanged, but AGE has
been incremented to 1. According to the notes, the only reason to increment AGE
is that interfaces have been added since
the last public release - i.e. the condition in note 5 is true. If note 5 is
true, then note 4 must also have been true,
which means that CURRENT should now have the value 9.
Of course, I'm assuming here that one works through notes 3 to 6 in order,
amending the variables according to the truth
of the condition. That may be an incorrect assumption, but, having read the
notes in the info page, I can't see how the
dynamic linker could work if any other method of arriving at the version
numbering was used.
I'm not subscribed, so please cc me on any reply.
Thanks
Chris
- [Libcdio-devel] libcdio-0.91 - libiso9660 version,
Chris Clayton <=