From: Marco Atzeri <marco.atzeri@gmail.com>
To: cygwin-apps@cygwin.com
Subject: Re: [ITP] FUSE 2.8
Date: Sat, 23 Jul 2016 06:02:00 -0000 [thread overview]
Message-ID: <4279d649-4e3f-a295-46d0-3d39bee15428@gmail.com> (raw)
In-Reply-To: <D3B806F7.A69F%billziss@navimatics.com>
On 23/07/2016 02:31, Bill Zissimopoulos wrote:
> On 7/22/16, 12:57 PM, Marco Atzeri wrote:
>
>
>> On 22/07/2016 19:58, Bill Zissimopoulos wrote:
>>>> winfsp-fuse is a reasonable name.
>>>> dokan-fuse also (IMHO)
>>>
>>> In the interest of moving things forward, I am happy to rename the
>>> package. Is it possible for a package with a name winfsp-fuse to
>>> satisfy a
>>> âfuseâ dependency?
>>
>> It is not clear to me what you mean.
>>
>> If some package depends on winfsp-fuse, the dependency will be
>> winfsp-fuse.
>>
>> winfsp-fuse will require the external package winfsp.
>
> My apologies for not making it clearer. What I meant to say is this:
>
> Suppose I have a package XYZ that requires FUSE. Is it possible that the
> âFUSEâ dependency can be satisfied by either winfsp-fuse or dokan-fuse?
>
> If that is not possible it looks like we would have to have winfsp-sshfs
> or dokan-sshfs, etc. which IMO is less desirable.
>
> Bill
It depends on the type of dependency
- The hard coded dependency is based on dll's
example : octave dll requires the readline dll provided by
the package libreadline7
$ cygcheck /usr/bin/cygoctave-3.dll |grep readline
E:\cygwin64\bin\cygreadline7.dll
$ cygcheck -f $(cygpath -u 'E:\cygwin64\bin\cygreadline7.dll')
libreadline7-6.3.8-1
In this case the dependency is hard coded in the octave binary itself.
We don't have a way (yet) to provide the same dll from two different
packages. In theory it can be done with pre/prost install scripts
that copy and remove dll's in "/usr/bin" but it will be tricky.
We can not use links with dll's on cygwin for windows loader constrain.
- The dependency is based on command
ls -l /usr/bin/automake
lrwxrwxrwx 1 marco Administrators 34 Aug 3 2013 /usr/bin/automake ->
/usr/share/autotools/am-wrapper.sh
64 $ cygcheck -f /usr/share/autotools/am-wrapper.sh
automake-9-1
$ cygcheck -cd |grep automake
automake 9-1
automake1.10 1.10.3-2
automake1.11 1.11.6-2
automake1.12 1.12.6-2
automake1.13 1.13.4-1
automake1.14 1.14.1-2
automake1.15 1.15-1
automake1.4 1.4p6-11
automake1.5 1.5-11
automake1.6 1.6.3-12
automake1.7 1.7.9-11
automake1.8 1.8.5-11
automake1.9 1.9.6-11
automake is a wrapper package that can be defined
to match the needed version and pull all the versions
$ automake --version
automake (GNU automake) 1.14.1
$ cygcheck-dep -r automake
automake: requires ( automake1.10 automake1.11 automake1.12
automake1.13 automake1.14 automake1.15 automake1.4 automake1.5
automake1.6 automake1.7 automake1.8 automake1.9 bash gawk )
Alternatives can also be used to manage command dependency
$ alternatives --display unison
unison - status is auto.
link currently points to /usr/bin/unison-2.48
/usr/bin/unison-2.48 - priority 2048
Current `best' version is /usr/bin/unison-2.48.
specially when you need only one variant installed,
but several are potentiall available:
$ cygcheck -cd |grep unison
unison2.48 2.48.3-2
$ cygcheck -p "bin/unison-"
Found 7 matches for bin/unison-
unison2.27-2.27.157-4 - unison2.27: Synchronize ...
unison2.32-2.32.52-4 - unison2.32: Synchronize ...
unison2.40-2.40.102-1 - unison2.40: Synchronize ..
unison2.45-2.45.28-1 - unison2.45: Synchronize ..
unison2.48-2.48.3-1 - unison2.48: Synchronize ..
unison2.48-2.48.3-2 - unison2.48: Synchronize ..
unison2.49-2.49.543-1 - unison2.49: Synchronize ..
Regards
Marco
PS: today there is only one case of packages providing same name dll's
and the solution is sub-optimal and working only as
- there is a preferred order
openblas cygblas-0.dll is preferred versus lapack one.
- the maintainer is the same: me
More packages with the same dll collision will require to work on
pre/prost install scripts general solution.
next prev parent reply other threads:[~2016-07-23 6:02 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-17 1:03 Bill Zissimopoulos
2016-07-17 15:43 ` David Stacey
2016-07-17 21:18 ` Marco Atzeri
2016-07-18 19:15 ` Bill Zissimopoulos
2016-07-18 8:19 ` Corinna Vinschen
2016-07-18 19:43 ` Bill Zissimopoulos
2016-07-18 19:52 ` Bill Zissimopoulos
2016-07-19 9:41 ` Corinna Vinschen
2016-07-19 17:27 ` Bill Zissimopoulos
2016-07-20 8:52 ` Corinna Vinschen
2016-07-21 22:12 ` Bill Zissimopoulos
2016-07-22 7:59 ` Corinna Vinschen
2016-07-22 12:30 ` Adrien JUND
2016-07-22 12:53 ` Marco Atzeri
2016-07-22 13:41 ` Corinna Vinschen
2016-07-22 17:58 ` Bill Zissimopoulos
2016-07-22 19:58 ` Marco Atzeri
2016-07-23 0:31 ` Bill Zissimopoulos
2016-07-23 6:02 ` Marco Atzeri [this message]
2016-07-23 6:33 ` Bill Zissimopoulos
2016-07-22 17:55 ` Bill Zissimopoulos
2016-07-22 11:59 Adrien JUND
2016-07-22 17:55 ` Bill Zissimopoulos
2016-07-22 19:56 ` Adrien JUND
2016-07-23 0:23 ` Bill Zissimopoulos
2016-07-23 10:40 ` Corinna Vinschen
2016-07-23 16:44 ` Bill Zissimopoulos
2016-07-23 17:48 ` Corinna Vinschen
2016-07-23 18:34 ` Bill Zissimopoulos
2016-07-25 7:35 ` Corinna Vinschen
2016-07-25 21:20 ` Bill Zissimopoulos
2016-07-26 6:28 ` Mark Geisert
2016-07-26 8:45 ` Herbert Stocker
2016-07-26 12:40 ` Adrien JUND
2016-07-26 18:13 ` Mark Geisert
2016-07-26 18:41 ` Bill Zissimopoulos
2016-07-26 17:23 ` Bill Zissimopoulos
2016-07-26 18:38 ` Bill Zissimopoulos
2016-07-26 19:02 ` Mark Geisert
2016-07-26 19:10 ` Bill Zissimopoulos
2016-07-26 23:22 ` Bill Zissimopoulos
2016-07-27 9:03 ` Mark Geisert
2016-07-27 17:45 ` Bill Zissimopoulos
2016-07-28 9:17 ` Mark Geisert
2016-07-28 16:36 ` Bill Zissimopoulos
2016-07-28 17:29 ` Mark Geisert
2016-07-28 17:43 ` Bill Zissimopoulos
2016-07-28 19:14 ` Bill Zissimopoulos
2016-07-28 19:21 ` Bill Zissimopoulos
2016-07-28 19:59 ` Mark Geisert
2016-07-28 20:08 ` Corinna Vinschen
2016-07-28 20:04 ` Corinna Vinschen
2016-07-28 21:18 ` Bill Zissimopoulos
2016-07-29 0:05 ` Mark Geisert
2016-07-29 0:17 ` Bill Zissimopoulos
2016-07-29 0:27 ` Bill Zissimopoulos
2016-07-29 8:19 ` Mark Geisert
2016-07-29 8:51 ` Corinna Vinschen
2016-07-29 9:15 ` Mark Geisert
2016-07-29 9:48 ` Corinna Vinschen
2016-08-16 17:32 ` Corinna Vinschen
2016-08-17 8:27 ` Mark Geisert
2016-08-17 9:14 ` Corinna Vinschen
2016-08-22 9:43 ` Mark Geisert
2016-08-22 11:54 ` Corinna Vinschen
2016-08-25 8:49 ` Bill Zissimopoulos
2016-09-05 9:36 ` Mark Geisert
2016-09-05 11:20 ` Adrien JUND
2016-09-05 20:16 ` Mark Geisert
2016-09-06 21:15 ` Bill Zissimopoulos
2016-09-07 5:22 ` Herbert Stocker
2016-09-08 8:13 ` Mark Geisert
2016-09-06 22:07 ` Mark Geisert
2016-09-06 21:13 ` Bill Zissimopoulos
2016-09-08 12:01 ` Corinna Vinschen
2016-09-08 16:58 ` Bill Zissimopoulos
2016-08-25 8:58 ` Bill Zissimopoulos
2016-07-29 15:35 ` Bill Zissimopoulos
2016-07-29 18:35 ` Mark Geisert
2016-07-29 7:50 ` Bill Zissimopoulos
2016-07-29 8:02 ` Mark Geisert
2016-07-26 20:07 ` Adrien JUND
2016-07-26 20:56 ` Bill Zissimopoulos
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4279d649-4e3f-a295-46d0-3d39bee15428@gmail.com \
--to=marco.atzeri@gmail.com \
--cc=cygwin-apps@cygwin.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).