DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org EB3A238319CF
Message-ID: <a61a7379-a470-4bb8-9a93-fef3fa25f2c3@kircheis.it>
Date: Fri, 21 Jun 2024 19:35:32 +0200
MIME-Version: 1.0
Subject: Re: Analyze dependencies of a cygwin package
To: cygwin@cygwin.com
References: <07082870-22a4-47a6-8fec-4ef954d181d1@kircheis.it>
 <ff59d014-94ec-490d-980c-538177daf658@maxrnd.com>
In-Reply-To: <ff59d014-94ec-490d-980c-538177daf658@maxrnd.com>
From: Federico Kircheis via Cygwin <cygwin@cygwin.com>
Reply-To: Federico Kircheis <federico@kircheis.it>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="us-ascii"; Format="flowed"
Sender: "Cygwin" <cygwin-bounces+archive-cygwin=delorie.com@cygwin.com>

On 21/06/2024 09.52, Mark Geisert via Cygwin wrote:
> On 6/20/2024 11:21 PM, Federico Kircheis via Cygwin wrote:
>> Hello to everyone,
>>
>> yesterday I've uploaded a new cmus package, with some optional 
>> features enabled.
>>
>> Today, through setup.exe, I noticed that now cmus brings clang as 
>> dependency.
>>
>> Is there something I can use for analyzing transitive dependencies of 
>> a package?
>> It does not make sense for an audio player to bring a compiler as 
>> dependency, and would like to remove the offending dependency.
> 
> Try the cygcheck-dep tool, downloadable through setup.exe. It was 
> created to answer questions just like this. Recommended :-).
> 
> ..mark
> 
> 

After some investigation, it seems that ffmeg hash clang (which has gcc) 
as dependency in its chain.

I would consider it a bug, although not critical.

It seems that most libraries have bash as a dependency, for example

  libmad0: recursively requires ( bash cygwin libgcc1 libiconv2 libintl8 
libncursesw10 libreadline7 libzstd1 terminfo zlib0 _windows(>=6.3) )
  libvorbis: recursively requires ( )
  flac: recursively requires ( bash cygwin libFLAC12 libgcc1 libiconv2 
libintl8 libncursesw10 libogg0 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )
  libopusfile0: recursively requires ( bash ca-certificates 
crypto-policies cygwin libffi6 libgcc1 libiconv2 libintl8 libncursesw10 
libogg0 libopus0 libp11-kit0 librea
dline7 libssl1.1 libtasn1_6 libzstd1 p11-kit p11-kit-trust terminfo 
zlib0 _windows(>=6.3) )
  libsamplerate0: recursively requires ( bash cygwin libgcc1 libiconv2 
libintl8 libncursesw10 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )
  libwavpack1: recursively requires ( bash cygwin libgcc1 libiconv2 
libintl8 libncursesw10 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )
  libcdio_paranoia2: recursively requires ( bash cygwin libcdio18 
libcdio_cdda2 libgcc1 libiconv2 libintl8 libncursesw10 libreadline7 
libzstd1 terminfo zlib0 _windows(
 >=6.3) )
  libmpcdec7: recursively requires ( bash cygwin libgcc1 libiconv2 
libintl8 libncursesw10 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )
  libcddb: recursively requires ( bash cygwin libcddb2 libcdio18 libgcc1 
libiconv2 libintl8 libncursesw10 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )
  libdiscid: recursively requires ( bash cygwin libdiscid0 libgcc1 
libiconv2 libintl8 libncursesw10 libreadline7 libzstd1 terminfo zlib0 
_windows(>=6.3) )

Again, it seems like a bug, but since so many libraries have bash as a 
dependency, could it be that cygcheck-dep incorrectly adds it?


-- 
Problem reports:      https://cygwin.com/problems.html
FAQ:                  https://cygwin.com/faq/
Documentation:        https://cygwin.com/docs.html
Unsubscribe info:     https://cygwin.com/ml/#unsubscribe-simple