Re: [ITP] libgpg-error-1.0-2

2004-10-01 Thread Corinna Vinschen
On Sep 30 10:00, Volker Quetschke wrote:
 http://www.scytek.de/cygwin/release/libgpg-error/setup.hint
 http://www.scytek.de/cygwin/release/libgpg-error/libgpg-error-1.0-2.tar.bz2
 http://www.scytek.de/cygwin/release/libgpg-error/libgpg-error-1.0-2-src.tar.bz2

3 votes, a GTG review.

Uploaded.


Thanks,
Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader  mailto:[EMAIL PROTECTED]
Red Hat, Inc.


Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Corinna Vinschen
On Sep 30 13:25, Volker Quetschke wrote:
 http://www.scytek.de/cygwin/release/libgcrypt/setup.hint
 http://www.scytek.de/cygwin/release/libgcrypt/libgcrypt-1.2.0-2.tar.bz2
 http://www.scytek.de/cygwin/release/libgcrypt/libgcrypt-1.2.0-2-src.tar.bz2

And another 3 votes + GTG review.

Uploaded with Volker's patch to setup.hint.


Thanks,
Corinna



-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader  mailto:[EMAIL PROTECTED]
Red Hat, Inc.


Re: setup: current setup version crashes on XP

2004-10-01 Thread Gerrit P. Haase
Gerrit schrieb:

 Gerrit schrieb:

 Gary R. Van Sickle wrote:
I'm running XP SP2, setup.exe crashes, I cannot install the
latest packages.

So currently I cannot use my XP box to work on new packages.

 I've been running SP2 for quite a while now at work and home and haven't had
 any trouble with setup.

 I ran it under dependency walker and it succeeds to install the packages
 I'm currently missing.

 I have Visual Studio .NET installed at the XP PC, maybe that makes a
 difference?

Nope, it is getting weird, I got the same error on the server (NT4 same
executable), maybe it has s.th. to do with the .NET framework?  I have
updated it recently at the XP box.  I have version 1.1 and the update
was the latest SP for the framework.  Does XP or some other MS
'services' modify executables?  Since it fails also at the NT4 it must
be s.th. different with the executable.  I deleted the  executable and
refetched it and now it runs again on the NT4 server.  Unfortunately
setup.exe doesn't even start anymore at the XP box.

I also noted that I cannot run another setup.exe named file, but I can
run 'normal' executables. 


Gerrit
-- 
=^..^=



Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Gerrit P. Haase
Reini schrieb:

 monster patch (gerrit-style)

Hey, all the great packages outside are including libtool-1.4 and
require automake-1.4 and autoconf-2.13, why should we depend on five
years old packages which are buggy and unusable most of the time?

 but +1

 I'd rather run ./autogen.sh style scripts in the conf step.

That would give the same patch size (usually it runs also aclocal,
libtoolize, autoconf, automake).


Gerrit
-- 
=^..^=



Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Gerrit P. Haase
Reini schrieb:

 monster patch (gerrit-style)

BTW, do you want to review my kaffe patch (4 MB bzip2 compressed)?

Gerrit
-- 
=^..^=



Re: upset errors for libopenldap2

2004-10-01 Thread Reini Urban
Corinna Vinschen schrieb:
On Sep 21 10:06, Christopher Faylor wrote:
Gerrit please fix this ASAP:
upset: *** warning package libopenldap2 refers to non-existent external-source: openldap

Sorry, my bad.  I removed version 2.1.whatever of openldap but I didn't
know that libopenldap2 has this external-source ref.  Fixed.
still errors in setup.ini for libopenldap2-2-15
@ libopenldap2-2-15
sdesc: Lightweight Directory Access Protocol runtime
ldesc: Lightweight Directory Access Protocol runtime
category: Libs Net
requires: cygwin minires openssl
@ libopenldap2_2_7
and the Current version line in setup.exe has a nice endless 
recursion. at least it looks like so.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/


libtool-devel-1.5.10-1-src download/install fails

2004-10-01 Thread Reini Urban
just a short notice, that you apparently cannot install/download
src packages for the test version of libtool-devel-1.5.10-1-src
trying to get libtool-devel-1.5.10-1-src via setup.exe
always gets me libtool-devel-1.5.6-3-src
but I couldn't verify that it fails generally:
libungif test src was possible.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/


Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Reini Urban
Gerrit P. Haase schrieb:
Reini schrieb:
monster patch (gerrit-style)
Hey, all the great packages outside are including libtool-1.4 and
require automake-1.4 and autoconf-2.13, why should we depend on five
years old packages which are buggy and unusable most of the time?
by using our own autotool 1.5 scripts in the prep or conf step.
Lapo does in prep (see libungif), I do it in conf.
(prep might be better as I think of it)
I'd rather run ./autogen.sh style scripts in the conf step.
That would give the same patch size (usually it runs also aclocal,
libtoolize, autoconf, automake).
no, do it in the .sh script. That keeps the patch to the bare and 
readable minimum,
and you only have to persuade the maintainers upstream to use
the newer 1.5 autotools. (just for our dll's, but what the heck.)
with the monster patch (new libtool, .in files, ...) they might get 
frightened.
And our src dependencies in the README must list the -devel autotool 1.5 
requirements of course, otherwise it will fail, and your explicit patch 
must be used.
--
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/


Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Gerrit P. Haase
Hi Reini,

I'd rather run ./autogen.sh style scripts in the conf step.
 
 That would give the same patch size (usually it runs also aclocal,
 libtoolize, autoconf, automake).

 no, do it in the .sh script. That keeps the patch to the bare and 
 readable minimum,

I cannot confirm that it reduces the patch sizes.  It will always be
huge, even if you use automake-1.8 where the upstream package includes
automak-1.79 generated files.

 and you only have to persuade the maintainers upstream to use
 the newer 1.5 autotools. (just for our dll's, but what the heck.)
 with the monster patch (new libtool, .in files, ...) they might get 
 frightened.

I don't submit those patches upstream, the only patches I submit
upstream are against Makefile.am and configure.in files and of course
source fixes.

 And our src dependencies in the README must list the -devel autotool 1.5
 requirements of course, otherwise it will fail, and your explicit patch
 must be used.

Yes, that is the reason why I include the patch, otherwise, Iwe could
add lines like: /usr/autotool/devel/bin/autoreconf -fiv to the script,
but then it will fail as soon as newer autotool version are avaiable or
the user has older versions installed.  Additionally I need to use a
pacthed libtool which is also included with the patch, if I would not
patch libtool I would have to patch at least ten Makefile.am of GLib 
Co. and probably other packages too.


Gerrit
-- 
=^..^=



Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Gerrit P. Haase
Gerrit schrieb:

 Hi Reini,

I'd rather run ./autogen.sh style scripts in the conf step.
 
 That would give the same patch size (usually it runs also aclocal,
 libtoolize, autoconf, automake).

 no, do it in the .sh script. That keeps the patch to the bare and 
 readable minimum,

 I cannot confirm that it reduces the patch sizes.  It will always be
 huge, even if you use automake-1.8 where the upstream package includes
 automak-1.79 generated files.

Sorry, my fault, misread your repliy.  I already described why I don't
do it dynamically.  Anyway, what is the problem with the great patches?

You can open an editor and search for +++ and you'll see that the
interesting files are always Makefile.am  configure.in.

Patchfiles are not for reading, though.

What you can do is: open the buildscript of any package and add the
following to the mkpatch function: -x 'Makefile.am' -x 'configure' \
-x 'ltmain.sh' -x 'aclocal.m4' -x 'config.sub' -x 'missing' and so on,
then you'll get a patchfile with only the relevant changes.  What about
including a function in the g-b-s to create such a small patch in
CYGWIN-PATCHES and deliver this minimal patch too?


Gerrit
-- 
=^..^=



Re: netpbm?

2004-10-01 Thread Igor Pechtchanski
On Fri, 1 Oct 2004, Charles Wilson wrote:

 Igor Pechtchanski wrote:

  Chuck, if you could dig it up, that'd be great.  Did you adapt it to use
  the generic-build-script?  If so, how did you deal with the weird
  configure?

 It uses a variant of the gbs, IIRC.  I'm on dailup right now, so I'll let you
 download it...

 http://users.ece.gatech.edu/~cwilson/cygutils/testing/index.php?dir=release%2Fnetpbm/

Thanks.  I was hoping for a less invasive set of changes to the GBS, but
most of them seem unavoidable.  I'll see what I can do, though.

The division into libnetpbm* and netpbm is good -- I hadn't thought of
that.

I'll get it working for myself first, and then see if I can find the time
to ITP it properly.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

Happiness lies in being privileged to work hard for long hours in doing
whatever you think is worth doing.  -- Dr. Jubal Harshaw


Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Gerrit P. Haase
Hi all,

 What you can do is: open the buildscript of any package and add the
 following to the mkpatch function: -x 'Makefile.am' -x 'configure' \
 -x 'ltmain.sh' -x 'aclocal.m4' -x 'config.sub' -x 'missing' and so on,
 then you'll get a patchfile with only the relevant changes.  What about
 including a function in the g-b-s to create such a small patch in
 CYGWIN-PATCHES and deliver this minimal patch too?

And don't miss to add -x 'CYGWIN-PATCHES' in the new mkrawpatch()
function.

Gerrit
-- 
=^..^=



Re: [GTG Review] Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Christopher Faylor
On Fri, Oct 01, 2004 at 08:33:28AM +0200, Dr. Volker Zell wrote:
 Volker Quetschke writes:

 As advertised, I would like to maintain the cygwin-port of the
 libgcrypt library.

This one is GTG with the following patch to setup.hint.

 URLs:
 http://www.scytek.de/cygwin/release/libgcrypt/setup.hint

--- setup.hint.orig 2004-10-01 08:30:45.494182400 +0200
+++ setup.hint  2004-10-01 08:30:58.082283200 +0200
@@ -2,4 +2,4 @@
 ldesc: Libgcrypt is a general purpose crypto library based on the code
 used in GnuPG.
 category: Libs
-requires: cygwin libgpg-error
+requires: cygwin libgpg-error _update-info-dir

You don't need to do this.  I went to great effort to make sure that this
gets added automatically to the setup.ini entry of anything which uses
info files.

I see that the usage has crept into a bunch of setup.hint files.  I've
nuked these instances.

cgf


Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Volker Quetschke
Hi!
Sorry, my fault, misread your repliy.  I already described why I don't
do it dynamically.  Anyway, what is the problem with the great patches?
You can open an editor and search for +++ and you'll see that the
interesting files are always Makefile.am  configure.in.
Patchfiles are not for reading, though.
What you can do is: open the buildscript of any package and add the
following to the mkpatch function: -x 'Makefile.am' -x 'configure' \
-x 'ltmain.sh' -x 'aclocal.m4' -x 'config.sub' -x 'missing' and so on,
then you'll get a patchfile with only the relevant changes.  What about
including a function in the g-b-s to create such a small patch in
CYGWIN-PATCHES and deliver this minimal patch too?
At least for libgcrypt, libgpg-error and gnupg you can do:
$ ./package-name.sh prep
$ ./package-name.sh conf
$ ./package-name.sh devspkg
In package-name-src.dev.tar.bz2 you will then find the bare,
short patch without the autotool stuff.
Volker
--
PGP/GPG key  (ID: 0x9F8A785D)  available  from  wwwkeys.de.pgp.net
key-fingerprint 550D F17E B082 A3E9 F913  9E53 3D35 C9BA 9F8A 785D


signature.asc
Description: OpenPGP digital signature


Re: [GTG Review] Re: [ITP] libgcrypt-1.2.0-2

2004-10-01 Thread Dr. Volker Zell
 Christopher Faylor writes:

 You don't need to do this.  I went to great effort to make sure that this
 gets added automatically to the setup.ini entry of anything which uses
 info files.

Well, gmp-4.1.3-3 for example has info files, but they definitly do not
show up in my dir file after installation.

 I see that the usage has crept into a bunch of setup.hint files.  I've
 nuked these instances.

 cgf

Ciao
  Volker



Re: [ITP] cdrtools-2.01

2004-10-01 Thread Ross Smith II
Quoting Corinna Vinschen:

 The author provides a ProDVD edition, which is based on the same source
 as cdrecord, but contains proprietary code to access DVD drives.  By
 providing a closed source application in binary only form linked against
 Cygwin, he's infringing the GPL.  I've contacted Joerg by mail to solve
 this issue but until then, I'm not happy with providing the tool in the
 net distro.  It feels wrong.

Corinna,

Any response from Joerg? It would make my life a little easier if cdrtools
was an offical package.

-Ross



rsync news (upstream patch + please upload) [3rd try]

2004-10-01 Thread Lapo Luchini
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Last 2 messages seems not to have reached the ML.. retrying for the
second time... (it was HTML by error, I guess that's why it was
filtered out?)

Wayne Davison wrote:
 On Fri, Oct 01, 2004 at 12:31:55AM +0200, Lapo Luchini wrote:
 I don't think there is any reason not to ask for binary mode on
 non-Windows hosts, anyway (it's already done also in
 do_open(...), also).
 It has to be configured around the presence of the setmode()
 function and only done if O_BINARY isn't 0 (which it is on systems
 that don't need it). So, I'll add it with the appropriate
 conditional compilation code wrapping it up. Thanks!

OK, so in next version the patch won't be needed anymore on our side.

 (I still wonder why it didn't show in earlier build, really):
 Probably because something recently changed in the cygwin support
 functions. (That code-path wouldn't have been used if HAVE_FCHMOD
 or HAVE_SECURE_MKSTEMP was zero.)

Is that so? Did cygwin just add/change support of one of those two
functions?
Anyway... now it's patches so that's not a real problem anymore.

What to say?

Ready for next release, I'd say!

URL:
http://www.lapo.it/cygwin/rsync-2.6.3-1-src.tar.bz2
http://www.lapo.it/cygwin/rsync-2.6.3-1.tar.bz2

Size:
594393
146864

MD5:
3b114d672f3e0fc727a3157a2368c404
3fd17049658428ec923ddc0e66aea7d4

SHA-1:
52aa37d24e4aef4590a408efa3a86652349556b7
2b2bf8773b102fd72da0bc7aef5a37c7b9f6d80d

Ah, some past version (by error) did use the internal popt instead of
the shared one, now it uses the shared one again.

Lapo

- --
Lapo Luchini
[EMAIL PROTECTED] (PGP  X.509 keys available)
http://www.lapo.it (ICQ UIN: 529796)
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (Cygwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iEYEARECAAYFAkFdf+4ACgkQaJiCLMjyUvuzyQCgv+o/FWNS377Tofb0Mj+I4BCz
e80AoLgFGSZ4Xq4w3SnrXrEBaZ7e9vwC
=AB1g
-END PGP SIGNATURE-



Re: upset errors for libopenldap2

2004-10-01 Thread Christopher Faylor
On Fri, Oct 01, 2004 at 02:41:08PM +0200, Reini Urban wrote:
Corinna Vinschen schrieb:

On Sep 21 10:06, Christopher Faylor wrote:

Gerrit please fix this ASAP:

upset: *** warning package libopenldap2 refers to non-existent 
external-source: openldap


Sorry, my bad.  I removed version 2.1.whatever of openldap but I didn't
know that libopenldap2 has this external-source ref.  Fixed.

still errors in setup.ini for libopenldap2-2-15

@ libopenldap2-2-15
sdesc: Lightweight Directory Access Protocol runtime
ldesc: Lightweight Directory Access Protocol runtime
category: Libs Net
requires: cygwin minires openssl

@ libopenldap2_2_7

and the Current version line in setup.exe has a nice endless 
recursion. at least it looks like so.

This should be fixed now.

cgf


Re: [ITP] cdrtools-2.01

2004-10-01 Thread Corinna Vinschen
On Oct  1 08:59, Ross Smith II wrote:
 Quoting Corinna Vinschen:
 
  The author provides a ProDVD edition, which is based on the same source
  as cdrecord, but contains proprietary code to access DVD drives.  By
  providing a closed source application in binary only form linked against
  Cygwin, he's infringing the GPL.  I've contacted Joerg by mail to solve
  this issue but until then, I'm not happy with providing the tool in the
  net distro.  It feels wrong.
 
 Corinna,
 
 Any response from Joerg?

No, I'm sorry.  I didn't get a reply so far.  It's two weeks now. 
There's a good chance that Joerg is just on vacation or so, so I'll
repeat my posting on Monday.


Corinna

P.S: Please don't Cc me.  I'm reading the mailing list.  I'm setting the
 Reply-To for a reason.

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader  mailto:[EMAIL PROTECTED]
Red Hat, Inc.