This is the mail archive of the
mailing list for the Cygwin project.
Re: Obsolete dependency report, 2015-May-15 [GOLDSTAR]
- From: Marco Atzeri <marco dot atzeri at gmail dot com>
- To: cygwin-apps at cygwin dot com
- Date: Sun, 17 May 2015 11:52:13 +0200
- Subject: Re: Obsolete dependency report, 2015-May-15 [GOLDSTAR]
- Authentication-results: sourceware.org; auth=none
- References: <1431716634 dot 7012 dot 35 dot camel at cygwin dot com> <55570642 dot 3030809 at gmail dot com> <1431851905 dot 5092 dot 3 dot camel at cygwin dot com>
On 5/17/2015 10:38 AM, Yaakov Selkowitz wrote:
On Sat, 2015-05-16 at 10:56 +0200, Marco Atzeri wrote:
I am uploading the dependencies of singular now and testing Yue's last
monolithic proposal. If it passes the test as expected I will also
upload both archs during weekend.
Wonderful! Thank you very much for helping with this.
As singular-surf is replaced by surf (and not singular), I moved the
former under the latter instead. Because surf has a lower version
though, the usual cygport OBSOLETES mechanism doesn't help, so I added
an empty 9999-1 upgrade helper manually.
I missed this your mail, but I saw the change in the structure.
WRT HDF5 in particular, as the primary consumer thereof, you way want to
consider only shipping an SONAME version bump in coordination with a new
release or rebuild of netcdf and octave. After all, a "fixed" newer
version doesn't really help anyone if an SONAME change means nobody is
using it yet.
I am upgrading hdf5 and octave this time but is a mere
coincidence of release date.
I will postpone any further hdf5 release
OTOH you're probably not the only one complaining about this. Is there
an upstream forum for packagers where you could mention the difficulties
this involves, or chime in on others' who have already mentioned the
Unfortunately they seems to NOT understand the concept of stable API.