Re: [Fink-devel] How'd I get a completely broken g++-4.0?

[Anthony DeRobertis]

Alexander K. Hansen wrote:

> It could have been Apple's installer.  It loves to miss a file now and
> then to keep one on one's toes.  It's in DevSDK.pkg:
> 

Could be. There wasn't a /usr/include/c++ at all...

Anyway, upgrading to xcode 2.3 seems to have fixed it.


___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] How'd I get a completely broken g++-4.0?

[Anthony DeRobertis]

fermi:/tmp anthony$ g++-3.3 foo.cpp
fermi:/tmp anthony$ echo $?
0

fermi:/tmp anthony$ g++-4.0 foo.cpp
foo.cpp:1:20: error: iostream: No such file or directory
foo.cpp: In function ‘int main()’:
foo.cpp:4: error: ‘cout’ is not a member of ‘std’

fermi:/tmp anthony$ cat foo.cpp
#include 

int main() {
 std::cout << "foo\n";
 return 0;
}

fermi:/tmp anthony$ locate iostream
/usr/include/gcc/darwin/3.3/c++/backward/iostream.h
/usr/include/gcc/darwin/3.3/c++/iostream
/usr/include/php/Zend/zend_istdiostream.h


Well, that's weird. How can a C++ compiler come without the C++ 
library... Any idea how I pulled that off? Sort of stopped my attempt to 
change over from 10.4-transitional to 10.4.

Anyway, I'm downloading a xcode image to upgrade to 2.3 anyway, guess 
I'll try and figure out how I got half of a C++ compiler tomorrow after 
that download is done.



___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Missing dependencies (ethereal-ssl or gtk+2)

[Anthony DeRobertis]

Martin Costabel wrote:
> Anthony DeRobertis wrote:
> []
> 
>> I'm rather curios why you'd want a IneheritedBuildDepends instead of 
>> just fixing the Depends handling --- because, currently, the 
>> installed gtk+2-dev is very much broken[0], and 
>> IneheritedBuildDepends (guessing from the name) won't fix that.
> 
> 
> There is nothing broken if the advice in the description

Unless, of course, you're building software not packaged for fink.

And even for fink, it is broken: What happens when gtk+-2 adds a new
dependency? You have to change a lot of source packages. Forget one, you
get weird bugs (like the ethereal one, which was first noticed over a
year ago).


> 
>>  Any package which BuildDepends on this one must also BuildDepend on:
>>  atk1 (>= 1.6.0-1), glib2-dev (>= 2.4.0-1), pango1-xft2-dev (>= 1.4.0-1) 
> 
> 
> is followed. The only problem is that this has do be done manually and
> is not done automatically. InheritedBuildDepends would be a method to do
>  it automatically.

No, that doesn't solve the non-Fink software problem. You still have a
package installed that, due to non-declared dependencies, is broken.


> 
> I don't know, I have them all on my hard disk :-)
> 

If you happen to have a subject & date, I'll track it down. Message-ID
might help too. Thank you in advance!


---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Missing dependencies (ethereal-ssl or gtk+2)

[Anthony DeRobertis]

On May 7, 2005, at 12:39, Martin Costabel wrote:
glib and glib2 are not good examples; they are compatible. But  
there are many *-dev packages that conflict with and replace other  
*-dev packages and are to be freely swapped in and out during  
package building. If any other package had a Depends on one of  
them, it could not be swapped out.
OK, that makes sense. Well, they could be swapped out (by removing  
the depending packages), but I'm guessing that's not implemented.


There are good reasons for this policy, but it is not complete;  
besides the BuildDependsOnly and BuildConflicts fields, it would  
need some replacement for the forbidden Depends, such as the  
InheritedBuildDepends that has been suggested but not yet implemented.
I'm rather curios why you'd want a IneheritedBuildDepends instead of  
just fixing the Depends handling --- because, currently, the  
installed gtk+2-dev is very much broken[0], and  
IneheritedBuildDepends (guessing from the name) won't fix that.

Is there a good pointer into the list archives?

[0] It, e.g., has #includes to pango, which will fail when only  
Essential and
Depends are installed. Not to mention pkg-config is very upset  
about it.

---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Missing dependencies (ethereal-ssl or gtk+2)

[Anthony DeRobertis]

Peter O'Gorman wrote:

> | I'm sure there is a good reason for this, but that really sounds rather
> | backwards...
> 
> Yes there is a good reason. Think for a second and you'll realize what it
> is Packages which are "Build Depends Only" can not depend on other,
> unrelated, packages.



> It is erfectly fine for them to depend on their own
> shared libraries etc, it is required, in fact. But if it were to depend on,
> for example glib2, then it would be impossible for another package to
> BuildDepend on glib without a BuildConflicts on gtk+2-dev etc,

Why?!

   BuildDepends
pkgA > glib
/|\
  Depends|
gtk+2-dev --/

The isn't anything inconsistent there.


There is the BuildDependsOnly stuff in the Policy Manual (not quite
clear on the reason for that; from reading Policy, it sounds like a
technical limitation in the package manager). And yes, since glib2 is no
doubt BuildDependsOnly: true, gtk+2-dev can't declare a Depends on it by
policy.

> Depends and BuildDepends are totally different things.

Really? Policy says otherwise. Policy says the only difference between
Depends and Build-Depends is that Build-Depends does not apply after the
build has finished. From that, it seems, at least at first glance (and
from experience with how Debian handles it) --- absent BuildDependsOnly
--- that you'd have gtk+2 Depends: on the extra header packages, and
it'd all work.


> Please put some thought into things before postingand accusing others of 
> stupidity.

I have not accused anyone of stupidity.


---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Missing dependencies (ethereal-ssl or gtk+2)

[Anthony DeRobertis]

Martin Costabel wrote:
> 
> gtk+2-dev must not depend on anything.

A quick check didn't show anything in fink packaging policy about -dev
packages not being able to depend on anything, and indeed gtk+2-dev does
depend on things:

Depends: gtk+2-shlibs (= 2.4.9-8), gtk+2 (= 2.4.9-8), darwin (>= 8-1)

> It has, however, in its
> description the lines:
> 
>  Any package which BuildDepends on this one must also BuildDepend on:
>  atk1 (>= 1.6.0-1), glib2-dev (>= 2.4.0-1), pango1-xft2-dev (>= 1.4.0-1)

I'm sure there is a good reason for this, but that really sounds rather
backwards...


---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] Missing dependencies (ethereal-ssl or gtk+2)

[Anthony DeRobertis]

In order for the build of ethereal-ssl to find GTK+2, I had to manually
install:

pango1-xft2-dev
atk1

I suspect gtk+2-dev should depend on these. Otherwise:

pkg-config --cflags gtk+-2.0

fails, complaining about those two packages missing. Which makes
ethereal's configure script decide gtk+2.0 isn't there. I imagine this
affects gtk+2.0-dev users in general, not just ethereal-ssl.


BTW: The symptom of this failure is that in the end you'll wind up with
 only tethereal. The build is still going, I'll report back if there
 turn out to be any additional problems. Googling for the problem,
 it appears that no one really solved it, though back in April
 someone thought their gtk+2.0 was broken; I suspect rebuilding
 gtk+2.0 just managed to install those missing packages.


---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] license question

[Anthony DeRobertis]

On May 2, 2005, at 19:58, Alexander Strange wrote:
That looks functionally identical to the BSD license with the  
advertising clause.

I think you mean WITHOUT the advertising clause. The BSD advertising  
clause says that the notice must appear in all ads, documentation,  
etc. talking about the software, not just the software itself.

---
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4
opportunities to win big! Highest score wins.NEC IT Guy Games. Play to
win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Re: License for .info and .patch files

[Anthony DeRobertis]

Benjamin Reed wrote:
+1
I think public domain is a good idea for such things.
Public domain doesn't do things like disclaim liability. I suggest the 
MIT X11 license instead:

Copyright (c)  
Permission is hereby granted, free of charge, to any person obtaining a 
copy of this software and associated documentation files (the 
"Software"), to deal in the Software without restriction, including 
without limitation the rights to use, copy, modify, merge, publish, 
distribute, sublicense, and/or sell copies of the Software, and to 
permit persons to whom the Software is furnished to do so, subject to 
the following conditions:

The above copyright notice and this permission notice shall be included 
in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS 
OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF 
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. 
IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY 
CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, 
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE 
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.


---
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] License for .info and .patch files

[Anthony DeRobertis]

Trevor Harmon wrote:
Some parts, yes, are facts: 
the name of the package, its dependencies, its home page. Others parts 
are not: the DescPort, for instance, or PatchScript. For this reason, 
one cannot say that the entire work is exempt from copyright law.
I agree --- and that's what I said in 
http://sourceforge.net/mailarchive/message.php?msg_id=11309312

He had created the SplitOffs and SSL handling in a way I never would
have thought of,  and I had added configure parameters that fixed bugs
> he was not able to resolve.
But once you filter out the idea (also not copyrightable, see Title 17, 
Sec. 102(b)), I suspect you'll find that the expression of that idea is 
quite limited by the technical and policy requirements of the info file.

> By
the same token, two programmers implementing a quicksort algorithm may 
end up with very similar code. But does this mean none of these works is 
protected by copyright law?
If the programming language you're working in gives you basically one 
way to write the quicksort algorithm, and you picked that way, then yes.

Now hold on here... Does a metric of sufficient creativity truly exist? 
Yes. See, for example, 
http://www.copyright.gov/help/faq/faq-protect.html#title ("Copyright 
does not protect names, titles, slogans, or /short phrases/" [emphasis 
added]), http://www.copyright.gov/circs/circ34.html (expanded version of 
prior link).

Also, look up the "de minimus" doctrine.
BTW: Copyright law on sound recordings and literary works is different.

Yesterday I went
to the store and bought a glass
of milk and drank it.
Is this poem copyrightable? By your definition, no. It is very short. It 
is also factual.
I'd guess no.
Oh, and watch out for haiku fanatics trying to extract revenge for that 
poem.

And yes, having licensing on info files clearly given by their authors 
would be a very good thing.

---
This SF.net email is sponsored by Demarc:
A global provider of Threat Management Solutions.
Download our HomeAdmin security software for free today!
http://www.demarc.com/Info/Sentarus/hamr30
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] License for .info and .patch files

[Anthony DeRobertis]

Trevor Harmon wrote:
On Mar 30, 2005, at 10:19 AM, Anthony DeRobertis wrote:
But once you filter out the idea (also not copyrightable, see Title 
17, Sec. 102(b)), I suspect you'll find that the expression of that 
idea is quite limited by the technical and policy requirements of the 
info file.

True, but that is not the case here. We are not extracting the ideas 
from .info files and creating new ones; we are copying and distributing 
others' work as-is. This is what I am saying is covered by copyright law.
When there are limited numbers of ways to express something, the idea 
and the expression are the same, and there is no copyright protection 
(as copyright can not protect an idea). This is the "merger doctrine".


And yes, having licensing on info files clearly given by their authors 
would be a very good thing.

Now I'm really confused. If I understand your point correctly, that 
.info files have no protection under copyright law, then what need is 
there for licensing them?
Having the author say "you can use this" is never a bad thing. And (see 
my original message), I believe that portions of some fink files are 
copyrightable.

---
This SF.net email is sponsored by Demarc:
A global provider of Threat Management Solutions.
Download our HomeAdmin security software for free today!
http://www.demarc.com/Info/Sentarus/hamr30
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] License for .info and .patch files

[Anthony DeRobertis]

Trevor Harmon wrote:
On Mar 28, 2005, at 2:23 AM, Anthony DeRobertis wrote:
What about an almanac? A news broadcast? An encyclopedia? These are all 
mere collections of facts. Are you trying to tell me that these cannot 
be copyrighted?
Copyright law in the US covers creative expression, not facts.
The almanac, news broadcast, and encyclopedia all contain a substantial 
amount of creativity in the selection and presentation of those facts. 
That's why they are copyrightable. You are free to lift the facts out of 
any of those --- the facts themselves are not, and can not be protected 
under the copyright act. So, you are free to do whatever you like with 
the sunrise and sunset times in your almanac, or the birth and death 
dates of Thomas Jefferson in your encyclopedia, or the latest 
developments in the Schivo case given on the news report.

I think a much similar case to a Fink info file is a telephone book, 
which can not be copyrighted (see the Supreme Court's decision in Feist 
Publications, Inc. v. Rural Telephone Service Co.[0]).

Take a look at the Fink info file of bzip2, for example. This is one I'd 
argue isn't copyrightable:

Package: bzip2
Version: 1.0.2
Revision: 12
Essential: yes
Depends: %N-shlibs (= %v-%r)
BuildDepends: fink (>= 0.13.0), fink-prebinding
Maintainer: Fink Core Group <[EMAIL PROTECTED]>
Source: mirror:sourceforge:fink/%n-%v.tar.gz
Source-MD5: ee76864958d568677f03db8afad92beb
Patch: %n.patch
CompileScript: make PREFIX=%p
InstallScript: make install PREFIX=%i
DocFiles: LICENSE README CHANGES manual*.html
SplitOff: <<
 Package: %N-shlibs
 Replaces: %N (<= 1.0.2-1)
 Depends: base-files
 Essential: true
 Files: lib/libbz2.*.dylib
 Shlibs: %p/lib/libbz2.1.dylib 1.0.1 %n (>= 1.0.2-2)
 Description: Shared libraries for bzip2 package
 DocFiles: LICENSE README CHANGES manual*.html
<<
SplitOff2: <<
 Package: %N-dev
 Depends: %N-shlibs (= %v-%r)
 Replaces: %N (<= 1.0.2-1)
 BuildDependsOnly: true
 Files: include lib/libbz2.dylib
 Description: Developer files for bzip2 package
 DocFiles: LICENSE README CHANGES manual*.html
<<
Up through here, we have nothing creative at all: Everything is a fact, 
presented in a form mandated by Fink (the software) and Fink policy; if 
you took the raw facts out of this, gave them to a different person, and 
he created a new info file, he'd wind up with the same thing.

Description: Block-sorting file compressor
DescDetail: <<
bzip2 is a portable, lossless data compressor based on the
Burrows-Wheeler transform. It achieves good compression and runs on
practically every (32/64-bit) platform in the known universe.
<<
DescPort: <<
Doesn't use autoconf, but comes with a useful Makefile. Anyway, the
patch modifies it to build a shared library instead of a static one.
<<
This is the only part that could, I think, even concievably be 
copyrighted. However, I very much doubt it; it is a list of facts with 
very little creativity in them. Not only that, its fairly short. If you 
tried to express the same facts, you'd likely wind up with the same thing.

License: OSI-Approved
Homepage: http://sources.redhat.com/bzip2/
More facts which are expressed in a form constrained entirly by Fink policy.

The scripts, if of suffient length and creativity might be. (Ones that 
just invoke install probably aren't. Just another collection of facts.)

The definition of copyright is not based on length nor creativity.
Yes, it is. See above.
[0] http://www.law.cornell.edu/copyright/cases/499_US_340.htm
---
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] License for .info and .patch files

[Anthony DeRobertis]

David R. Morrison wrote:
But what about the .info files?  Should we declare them to be part of fink,
and therefore under the GPL?  Can we do this retroactively, even though
we didn't make it clear to contributers in the past?
Most of a finkinfo file is just a statement of facts, and is nearly 
entirely determined by technical requirements. Those parts are most 
likely not copyrightable at all. No worries there, then.

The scripts, if of suffient length and creativity might be. (Ones that 
just invoke install probably aren't. Just another collection of facts.)

The descriptions also might be.
No idea how likely it is you have an implied license under the GPL. I'd 
guess you have some type of implied license to distribute and modify, as 
anyone who contributed was fairly clear that was going to happen.

IANAL.
---
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] the gpl and openssl

[Anthony DeRobertis]

David R. Morrison wrote:
Anthony,
Thanks very much for this very helpful message.
I'm curious of there is any difference for software released under the
LGPL instead of the GPL.  Can it legally link to openSSL?
LGPL Sec. 6 seems to allow this:
  6. As an exception to the Sections above, you may also combine or
link a "work that uses the Library" with the Library to produce a
work containing portions of the Library, and distribute that work
under terms of your choice, provided that the terms permit
modification of the work for the customer's own use and reverse
engineering for debugging such modifications.
Also, (6) requires you do one of 5 things; in short (read the license 
for details):
	(a) Accompany with source so the user may re-link
	(b) Use shared lib
	(c) 3yr offer of (a)
	(d) Explicitly allows distribution offered from same place
	(e) if you've already done above, you don't have to do it again.

Fink, by distrubuting full source to both OpenSSL and the LGPL'd 
library, seems to easily comply with (a) via (d).

However, do note that the LGPL is not really appropriate for anything 
but a library; see clause 2(a) for example:

(2) You may modify your copy or copies of the Library or any portion
of it, provided that you also meet all of these conditions:
a) The modified work must itself be a software library.
(Note that the LGLP gives different definitions for Library vs. library.)
[FYI: I haven't spent much time considering the LGPL; it doesn't seem to 
come up as much.]

---
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] the gpl and openssl

[Anthony DeRobertis]

As one of the regular participants on debian-legal, and probably one of 
the participants in that thread, I'd like to clarify a few things:

- OpenSSL is not considered 'part of the system libraries', and
  thus does not fall under that excemption in the GPL.
Debian can not ever use the system libraries exception. If you carefuly 
read GPL(3), it's clear why: "need not include anything that is normally 
distributed...with the major components...of the operating 
system...unless that component itself accompanies the executable." On a 
Debian FTP server or cd/dvd set, everything accompanies each other. So 
even if openssl is normally distributed with the major components of the 
Debian OS, it doesn't matter; the executable is being distributed with 
openssl.

If you link against Apple's openssl, then Fink can probably use this 
exception.

- The FSF GPL seems to argue (in their GPL FAQ) that if a (GPL
  licenced) application has specific code to interface with a
  non-GPL package, then you may assume that such an exception is
  implied by the authors of the code. I would then logically
  conclude, that would imply those authors were at fault by just
  distributing that specific code interfacing with OpenSSL. However,
  I am not a lawyer, but had the impression that the legal people
  did not agree with my logic here. So I gave up.
There are two problems with this.
1. Debian is very conservative on licensing issues. So we never
   allow this argument.
2. This would really only apply if OpenSSL support was in the
   software from the start and no code has been borrowed from
   other GPL projects.
   If the OpenSSL code was added later then contributers before
   the addition of OpenSSL certainly can't be said to have
   intended their code to be used with gpl-incompatible OpenSSL.
   If the project has used GPL code from other projects then
   there is no reason to expect those authors are OK with OpenSSL
   either.
---
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] dpkg-deb failed, exit code 10 or 11

[Anthony DeRobertis]

On Thu, 2004-01-08 at 02:40, Martin Costabel wrote:
> (in particular, you cannot reinstall dpkg when 
> dpkg is not working).

You most certainly can. .deb's are "ar" files. You can use ar x to
unpack:

[EMAIL PROTECTED]:tmp$ mkdir /tmp/foo
[EMAIL PROTECTED]:tmp$ cd /tmp/foo
[EMAIL PROTECTED]:foo$ ar x /var/cache/apt/archives/apt_0.6.9_i386.deb 
[EMAIL PROTECTED]:foo$ ls
control.tar.gz  data.tar.gz  debian-binary

At this point, you have tar files.

[EMAIL PROTECTED]:foo$ tar tzf data.tar.gz | head
./
./usr/
./usr/bin/
./usr/bin/apt-cache
./usr/bin/apt-cdrom
./usr/bin/apt-config
./usr/bin/apt-get
./usr/bin/apt-key
./usr/lib/
./usr/lib/apt/

So, you can certainly unpack that into / thereby replacing it. 

OTOH, if dpkg-deb is broken, then you should still have, e.g.,
debian/tmp/ or whatever fink builds into, and you can copy that into /


signature.asc
Description: This is a digitally signed message part

[Fink-devel] Ethereal-ssl out of date?

[Anthony DeRobertis]

http://fink.sourceforge.net/pdb/package.php/ethereal lists there being 
a 0.9.16-11 ethereal but only 0.9.14-1 ethereal-ssl. Any reason for 
that?



---
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] selfupdate-cvs broken?

[Anthony DeRobertis]

On Nov 20, 2003, at 17:41, Darian Lanx wrote:

Please note. If you are using the "unstable" package manager versioned 
0.17.1 and you are NOT behind a firewall or a developer,
Well, I have several packages I maintain locally, but they're in local, 
so I doubt rsync would blow them away (I hope!). And my firewall is OK 
with rsync.

I tried rsync and it had various brokenesses, so I switched back to 
cvs, which looking at recent threads, seems like it might still be a 
good idea.

That made no sense.
/me goes to take the previous sentence out and shoot it.


---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] selfupdate-cvs broken?

[Anthony DeRobertis]

On Nov 20, 2003, at 16:46, TheSin wrote:

yes x11-wm isn't the problem, those are all Us look for an other 
letter higher up
Hmmm, you're right --- there were some "files in the way" further up. 
Thanks.



---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] sources for xc-20031117.tar.bz2

[Anthony DeRobertis]

On Nov 20, 2003, at 14:47, Alexander K. Hansen wrote:

This can happen if the upstream site (and mirrors) removes the tarball  
before it gets mirrored onto the master (opendarwin) site.  I got  
bitten this way by pkg-order, which has apparently been purged by  
Debian.
http://archive.debian.org/dists/potato/main/source/misc/pkg- 
order_1.12.dsc
http://archive.debian.org/dists/potato/main/source/misc/pkg- 
order_1.12.tar.gz

And http://cvs.debian.org/debian-cd/Attic/pkg-order makes interesting  
reading.
http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=pkg-order&archive=yes  
as well.

"Hi,

I have asked pkg-order to be removed from the distribution.
 The reasons for this were many:
   a) This no longer scratches an itch that I have. This is reflected
  in the fact that there have been grave bugs reported before
  releasing potato that have not been solved; and there has been
  no updates since July 2000.
   b) This package was meant top be a library package; and that seems
  to have been a dismal failure. Apart from the few example
  scripts distributed with the libs, nothing was ever built on top
  of it
   c) Reading the Status file is a kludgey way to do what this package
  does
   d) dpkg and apt have evolved since the package was written; and,
  really, in order to not have to track stuff like that this
  package needs be rewritten in terms of libapt
   e) This needs to be redone in C++ or something.
   f) Some of the motivation of this package is already addressed by
  modern apt capabilities.
	Thanks for your interest in pkg order

	manoj"



---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] selfupdate-cvs broken?

[Anthony DeRobertis]

U 10.3/unstable/main/finkinfo/x11-wm/fvwm2-2.4.15-11.patch
U 10.3/unstable/main/finkinfo/x11-wm/fvwm2-no-gnome-2.4.15-11.info
U 10.3/unstable/main/finkinfo/x11-wm/icewm-1.2.2-1.info
U 10.3/unstable/main/finkinfo/x11-wm/icewm-1.2.2-1.patch
U 10.3/unstable/main/finkinfo/x11-wm/icewm-basic-1.2.2-1.info
U 10.3/unstable/main/finkinfo/x11-wm/icewm-basic-1.2.2-1.patch
U 10.3/unstable/main/finkinfo/x11-wm/obconf.info
U 10.3/unstable/main/finkinfo/x11-wm/openbox.info
U 10.3/unstable/main/finkinfo/x11-wm/sawfish-1.1-16.info
U 10.3/unstable/main/finkinfo/x11-wm/startup-notification.info
U 10.3/unstable/main/finkinfo/x11-wm/windowmaker-0.80.1-3.info
U 10.3/unstable/main/finkinfo/x11-wm/windowmaker-0.80.1-3.patch
### execution of su failed, exit code 1
Failed: Updating using CVS failed. Check the error messages above.
Fermi:~ anthony$
This is after I already did an sudo rm * in 
10.3/unstable/main/finkinfo/x11-wm/; any idea what's up (or not, as the 
case may be)?



---
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] /man directory?

[Anthony DeRobertis]

It seems that at some point I aquired a /man directory.

Fermi:/ anthony$ ls -lR /man
total 0
drwxr-xr-x  9 root  admin  306 27 Oct 19:33 man3
/man/man3:
total 160
-r--r--r--  1 root  admin   4710 27 Oct 19:22 NetSNMP::ASN.3pm
-r--r--r--  1 root  admin   5463 27 Oct 19:22 NetSNMP::OID.3pm
-r--r--r--  1 root  admin   4586 27 Oct 19:23 NetSNMP::agent.3pm
-r--r--r--  1 root  admin   5744 27 Oct 19:22 
NetSNMP::agent::default_store.3pm
-r--r--r--  1 root  admin   9099 27 Oct 19:21 NetSNMP::default_store.3pm
-r--r--r--  1 root  admin   4081 27 Oct 19:23 
NetSNMP::netsnmp_request_infoPtr.3pm
-r--r--r--  1 root  admin  32720 27 Oct 19:24 SNMP.3pm
Fermi:/ anthony$

dpkg denies all knowledge of these files, so I'm just rm -Rf'ing them. 
These all have a data of 27 Oct, so I guess its from an older net-snmp.



---
This SF.Net email sponsored by: ApacheCon 2003,
16-19 November in Las Vegas. Learn firsthand the latest
developments in Apache, PHP, Perl, XML, Java, MySQL,
WebDAV, and more! http://www.apachecon.com/
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Dpkg...

[Anthony DeRobertis]

On Fri, 2003-10-10 at 20:19, John Davidorff Pell wrote:
> Aren't there ways  
> of simply extracting the files within a deb? Then we could parse the  
> pre/post scripts ourselves... anyone up for a dpkg replacement? ;-)

try 'man dpkg' before re-inventing the wheel:

dpkg --force-not-root

btw, .deb files can be extracted with either dpkg-deb or the standard
utilities 'ar', 'tar' and 'gzip'.


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Dpkg...

[Anthony DeRobertis]

On Sun, 2003-10-19 at 02:04, John Davidorff Pell wrote:
> But this does not work as intended. Dpkg still tries to set group and 
> user IDs even tho they are already correct, resulting in permission 
> denied errors. :-(

--force-all seems to make it continue anyway. Actually, on my Debian
system when I did that, all the errors I saw were from preinst and
postinst. 

This is with dpkg 1.10.10.


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] ABI Changes: 3.1 -> 3.3

[Anthony DeRobertis]

On Sunday, Jun 29, 2003, at 12:12 US/Eastern, David R. Morrison wrote:

What does this imply for
our binary distribution?
fyi, Debian renamed almost every C++ library package to deal with this 
one.



---
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa0016ave/direct;at.asp_061203_01/01
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] essential splitoffs

[Anthony DeRobertis]

On Friday, Jun 13, 2003, at 12:56 US/Eastern, David R. Morrison wrote:
And
in fact, we've had a policy (unstated, I believe) that there was no
need to specify any dependencies on an essential package.
Debian's policy on the matter is that versioned dependencies on 
essential packages must be declared; unversioned ones do not need to be.



---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] essential splitoffs

[Anthony DeRobertis]

On Thursday, Jun 12, 2003, at 16:12 US/Eastern, David R. Morrison wrote:
At that point, it might be possible to switch which package is  
essential
from gettext to gettext2.  However, if other packages have *assumed*
that every essential package is present and doesn't need to be  
declared,
then it won't be possible to make this switch.
My point is there should already be a versioned dependency on gettext  
(because of libintl) created by dpkg-shlibdeps (or the Fink equiv.)

For example, apt-cache showpkg libc6 (an essential package, of course)  
shows:

$ apt-cache showpkg libc6
Package: libc6
Versions:
2.2.5-11.5(/var/lib/apt/lists/ 
security.debian.org_dists_stable_updates_main_binary-i386_Packages)(/ 
var/lib/dpkg/status)
2.2.5-11.2(/var/lib/apt/lists/ 
http.us.debian.org_debian_dists_stable_main_binary-i386_Packages)

Reverse Depends:
  nessus,libc6 2.2.4-4
  smake,libc6 2.2.4-4
  libber0,libc6 2.2.4-4
  libnasl1,libc6 2.2.4-4
  bidwatcher,libc6 2.2.4-4
  smartmontools,libc6 2.2.4-4
  cdrdao,libc6 2.2.4-4
  libconfig0,libc6 2.2.4-4
  libgd-tools,libc6 2.2.4-4
  nessus-plugins,libc6 2.2.4-4
  amap,libc6 2.2.4-4
  libgd2-xpm,libc6 2.2.4-4
  gtk-gnutella,libc6 2.2.4-4
  nessusd,libc6 2.2.4-4
  tuxtime,libc6 2.2.4-4
  ... (there are 5427 lines total)


---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] essential splitoffs

[Anthony DeRobertis]

On Saturday, Jun 7, 2003, at 11:16 US/Eastern, David R. Morrison wrote:
1) Proposed New Policy:  You must declare any needed Depends or 
BuildDepends
of your package on essential packages.
If you do this for even unversioned dependencies, what's the point of 
having essential packages anymore?



---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] essential splitoffs

[Anthony DeRobertis]

On Monday, Jun 9, 2003, at 13:11 US/Eastern, David R. Morrison wrote:

If other packages
have not been required to specify their dependency on gettext-shlibs, 
things
will break at that stage.
Ummm, if packages depend on a specific version of gettext (like a 
library dependency does), shouldn't there already be a dependency 
there? It shouldn't matter if gettext is essential or not.

Of course, apt-cache showpkg gettext shows troubling results.

What happens if:

1) User installs essential library liba-shlibs-1.0
2) New liba-shlibs-1.1 is released. Is backwards compatible, so the
   soname need not be changed. However, it adds a new feature
3) Package foo-0.9 is released, linked against liba-shlibs-1.1
4) user does "apt-get install foo"
Either foo has a depends on liba-shlibs >= 1.1 or problems result. 
Whether liba-shlibs is Essential: yes is irrelevant.

(Try an apt-cache showpkg libc6 on a Debian system sometime; a LOT of 
reverse dependencies will show up, despite libc6 being Essential: yes. 
At least in Debian policy, if a package needs a specific version of an 
essential package, it must declare a dependency.)



---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Re: Competition for Fink: gentoo

[Anthony DeRobertis]

Later, when dependencies change or packages are removed, portage can 
use 'world' to figure out what packages were installed as dependencies 
but are no longer needed, and 'clean' them out.
At least for the bindist, aptitude would do this for you.



---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: glut license (was Re: [Fink-devel] MD5)

[Anthony DeRobertis]

On Sun, 2003-03-30 at 17:43, David R. Morrison wrote:

> I looked at glut, and I guess there is some ambiguity in the sentence
> "The programs are not in the
> public domain, but they are freely distributable without licensing
> fees."

>From /usr/share/doc/libglut3/copyright on my Debian system:

Regarding the right to modify and distribute this library:

  From: Mark Kilgard <[EMAIL PROTECTED]>
  Subject: Bug#131997: GLUT license
  
  Anthony,
  
  > Would it be possible for you to just make it absolutely clear that
  > everyone's allowed to use, copy, and modify (and distribute their
  > modifications) of libglut?
  > 
  > Just quoting the above and replying "Yes, that's fine" or similar
  > would be okay.
  
  Yes, that's fine.

[If you think Fink is strict with licenses, you've never seen Debian]


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Proposal: move all info-docs into splitoffs

[Anthony DeRobertis]

On Mon, 2003-03-10 at 19:03, Carsten Klapp wrote:

> The info files for the various Fink packages seem to take up a lot of 
> space. I just rm /sw/share/info/* which is fine for my personal 
> machine, but I had this splitoff idea and wonder how others feel.

It's really the wrong solution. The right solution is to implement the
long-wanted dpkg feature to not populate certain directories. You could
do it for usr/share/doc as well to save some more space.


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Fink distfiles mirror now up

[Anthony DeRobertis]

On Mon, 2003-03-10 at 17:19, Randal L. Schwartz wrote:

> I always get a red flag when people say "ls $SOMEPATTERN".

Hmmm, starting a Useless use of ls award?

Find can be useful there, btw:

find . -type f -print0 | xargs -0 ...

not only does it recurse, but it ignores directories! And gets past the
kernel argument length limit, too!


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Epoch

[Anthony DeRobertis]

On Wed, 2003-03-05 at 09:39, Max Horn wrote:
> the 5.0-rc -> 4.999-rc sounds like shit

5.0alpha < 5.0beta < 5.0rc < 5.0rel

Someone on debian-devel came up with that once. I think I've remembered
it right.


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Advice needed: manpage filename conflicts

[Anthony DeRobertis]

On Sat, 2003-03-01 at 21:08, David R. Morrison wrote:
> We gotta get you on the Fink team, Anthony... you understand the Debian
> tools better than we do!

Thanks ;-)

> 
> So how does Debian handle the situation where two different pacakges want
> to provide executables and/or man pages with the same name?

Debian handles it as specified in (Debian) Policy, 11.1:

Two different packages must not install programs with different
functionality but with the same filenames. (The case of two programs 
having the same functionality but different implementations is handled
via `alternatives' or the `Conflicts' mechanism. See Maintainer
Scripts, Section 2.3.9 and Conflicting binary packages - Conflicts,
Section 7.3 respectively.) If this case happens, one of the programs
must be renamed. The maintainers should report this to the
debian-devel mailing list and try to find a consensus about which
program will have to be renamed. If a consensus cannot be reached,
both programs must be renamed.

On my Debian system, btw, screensaver modules are installed in
/usr/lib/xscreensaver and manpages for them in section 6 (in the normal
directories).

Which screensaver module is this? I can check what its called on Debian.
That might provide some good ideas. Since screensavers are in their own
directory not normally in $PATH, there is a chance that Policy 11.1
isn't followed, of course ;-)



signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Advice needed: manpage filename conflicts

[Anthony DeRobertis]

On Fri, 2003-02-28 at 07:03, David R. Morrison wrote:
> Fink has borrowed from Debian something called the "alternatives" system.
> If there are two packages which want to install files with the same name,

Noo! That's not what alternatives is for. Its for two packages that
provide the same (or nearly the same) functionality. For example, two
implementations of awk (on Debian, we have gawk and mawk, for example).
We also have things like "editor" and "x-terminal-emulator"

This sounds like two completely different programs, so they are not
alternatives.

Which xscreensaver module is this? 


signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

On Sun, 2003-02-23 at 21:26, David R. Morrison wrote:

> Hmmm... Yeah, I could do it in tetex-texmf... It's tricky, though; I only
> want to make the tetex-base script fail if its an old version of tetex-base
> that's installed.  OK, I will try some more...

Some suggestions on how to do this:
1) md5sum. Problematic if there were a bunch of different broken
   prerm's. Great if there was only one.
2) grep for the mistaken rm -Rf. Probably the best solution.
3) You might be able to use dpkg -s --- I'm not sure. If so, you
   can use dpkg to compare the version. Problematic if you aren't
   careful and people built local versions of tetex-base.
4) Break all versions, trust that the new prerm can handle it.
   Probably a bad idea; may lead to future problems.

You may want to do something like (2) and sed out the bad line. Then let
the (new) tetex-base preinst do the cleaup the right way.



signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

On Sun, 2003-02-23 at 17:46, David R. Morrison wrote:

> Sorry, I said it wrong before: the error is in the old PreRemove script,
> not the PostRemove script.
> 
> The problem is, the very first thing which dpkg does upon updating a
> package is to execute the *old* prerm script.  I would never have a
> chance to modify that before it has done its damage.

Hmmm, sounds like you'd have to do it in the tetex-texmf maintainer
scripts, then. I think at least the pre-inst for tetex-texmf should be 
called before tetex-base's prerm.

[ Yes, I understand that this is a one-time thing. But its also a
  one-time thing that effects upgrades from the stable distribution,
  correct? Even if not, I guess I'm used to Debian where if
  apt-get upgrade doesn't work, it's a grave bug... ]




signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

On Friday, February 21, 2003, at 04:40 PM, David R. Morrison wrote:
The problem is that the removal of files is happening in PostRemove, 
outside
of dpkg's control.  But if you can think of an alternate way to solve 
this
problem, I would love to hear it.
Have one of your scripts edit the var/lib/dpkg/info/tetex-base.postrm. 
I don't have a copy of Debian policy in front of me to check where you 
could do this (probably even in tetex-base pre-inst, but I'd have to 
check that), but I'm sure there are several places.

I think you can force dpkg to try the postrm from the new package by 
just replacing the old postrm with something that will fail, like "exit 
1"

Yes, this method is evil. But I'd argue less evil than telling people 
to reinstall a package.

PS: Please don't demolish configuration files without asking the user 
--- check tetex-base.preinst.



---
This SF.net email is sponsored by: SlickEdit Inc. Develop an edge.
The most comprehensive and flexible code editor you can use.
Code faster. C/C++, C#, Java, HTML, XML, many more. FREE 30-Day Trial.
www.slickedit.com/sourceforge
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

On Mon, 2003-02-17 at 05:14, Martin Costabel wrote:
> On lundi, fév 17, 2003, at 08:48 Europe/Paris, Anthony DeRobertis wrote:
> []
> > [0] Reading the postinst, this seems to be when a certain file is 
> > missing. What ever happened to this file? And there are certainly 
> > better ways to deal with this than asking for a manual re-install...
> 
> I don' think so. This is a neat trick, in fact. The reinstall costs you 
> nothing, since you already built the deb file.

Not really. Remove/reinstall takes many minutes on my iBook 500 ---
certainly not as long as the build, but it's not free. And unlike the
build, it requires manual action.

Some suggestions:

1) Add the file back. Depending on how it went missing, it
   should still be in dpkg's database. Otherwise things like
   dpkg-divert should make it possible.
2) Rename the package, give the new one Provides/Replaces/Conflicts
   with the old one. Depend on the new one.

Probably some other courses of action as well --- but I'd have to hear
details before thinking of them.

> 
> Don't underestimate the difficulty of upgrading a bunch of heavily 
> interdependent packages to a new version that has many new files and 
> files moved around to different places.

I don't --- I've been around Debian long enough to appreciate the
difficulty of making things work right. 



---
This SF.net email is sponsored by: SlickEdit Inc. Develop an edge.
The most comprehensive and flexible code editor you can use.
Code faster. C/C++, C#, Java, HTML, XML, many more. FREE 30-Day Trial.
www.slickedit.com/sourceforge
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

On Monday, February 17, 2003, at 11:00 AM, Ben Hines wrote:


Hmm? dpkg -r huh? When it says "reinstall", do that: "fink reinstall 
tetex-texmf"

 dpkg -r, fink install or dpkg -r, dpkg -i re-installs too.

Actually, come to thing of it, just dpkg -i would, too.

(I probably did it the dpkg way because I'm more familiar with Debian.)



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

[Fink-devel] tetex upgrade problems

[Anthony DeRobertis]

[Sorry, I didn't keep build logs...]

I just did a fink update-all, and had a much less-than-pleasant 
experience with tetex. Well, actually, the lyx build is still running, 
so I'm still having it. Hopefully it all works now...

This experience involved nice things like having to use dpkg --force a 
time or two (files existing in multiple packages). Having to manually 
remove dvipdf (or was it dvips, sorry, don't remember...) [and thus 
lyx] because it conflicted with the new version of tetex.

Then, to top it all off, tetex-base postinst asked me to re-install 
tetex-texmf[0]

A dpkg -r later, it's time for fink install lyx. Hope it works...

What's going on with tetex? I finally found a note in the mailing list 
archives about some of this, but now I'm curious: Why?

And, is there anything I can do to help kill this evil mess?

[0] Reading the postinst, this seems to be when a certain file is 
missing. What ever happened to this file? And there are certainly 
better ways to deal with this than asking for a manual re-install...



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Who should Provide X11 ?

[Anthony DeRobertis]

[ Please forgive me if the tone is a little harsh; I'm stuck in front of
  a Windoze box for the moment... ]

> Sure you can think of scenarios where someone might want the X
> libraries installed and not the X server, but these are largely
> artificial.

Erm, what?! I have at least 20 servers, of varios OS's, which have the X11
base clients and X11 shared libraries in stalled, but no X11 server. No X
fonts, font servers, etc. either.

Why would I? I don't even have a monitor on some of them; many are at
various data centers; etc. ssh -X forwards X11 to my workstation.

> As people have pointed out you are only saving 5-10% of
> the size of the distribution by separating out the server and there

For those of us with 10GB HDs, XFree takes some very real disk space! And
don't say that's silly --- it's what my slightly over a year old iBook has.
I have many servers with far less.




---
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Re: [Fink-users] debianutils_1.23.tar.gz not foundfor selfupdate-cvs

[Anthony DeRobertis]

FYI, you can grab the old sources from http://snapshot.debian.net/



signature.asc
Description: This is a digitally signed message part

Re: [OT] Re: [Fink-devel] Sourceforge project of the month

[Anthony DeRobertis]

On Mon, 2002-11-11 at 23:04, Carsten Klapp wrote:
> 
> Hi Anthony,
> 
> Funny you should mention this I just managed to get Debian's Potato 
> installed on my 6100/66 yesterday.

Ugh. Why potato? Woody is out now, and I think you'll find it much
better. New XFree86, for example.

And, oh yeah, twice as many packages.

> Of course I don't know all the 
> details about apt-get and dselect, maybe there is some way to do these 
> already.

assuming you have deb-src lines in your /etc/apt/sources.list, you can
do:

   apt-get source package_name
   apt-get source --compile package_name

Also, "dpkg-buildpackage -rfakeroot" is your friend.

(if these want root, look at the fakeroot package)




signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] dpkg 'available' file - sections?

[Anthony DeRobertis]

On Saturday, November 2, 2002, at 10:43 PM, Ben Hines wrote:


This probably also explains why i have no sections in dselect... Does 
anyone?

I do, for some packages (both available and installed). On the Debian 
system, the section comes from the Packages and Packages.gz files which 
are maintained by the ftp masters.

scanpackages *should* generate these files in Fink; after that, "sudo 
deselect update" or "sudo apt-get update" may be required to make 
dselect see them.

Now, no guarantee it isn't broken. 



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Sourceforge project of the month

[Anthony DeRobertis]

On Saturday, November 2, 2002, at 06:41 PM, Carsten Klapp wrote:


I'm going to try to "port" fink to MkLinux so I don't have to use rpm 
anymore (of course it will have to always build from source and not 
use the binary packages).

An easier (and more package-complete) alternative would be Debian's PPC 
port, if you can run it. May even be able to run it on Mach.



---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] FWD: tetex-base

[Anthony DeRobertis]

On Fri, 2002-10-11 at 20:10, David R. Morrison wrote:

> Anyone have suggestions about how to do this?

Remove dialog from tetex-base, make tetex-base Depend: on dialog.
(Assuming the dialog package works with tetex)



signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] Localizing fink

[Anthony DeRobertis]

On Sat, 2002-06-22 at 02:57, Ben Hines wrote:
> Has any thought gone into localizing fink? 

I haven't heard any. Much thought (and flamewars), however, has gone
into localizing apt and dpkg. You can find it all at the debian-devel
mailing list.

http://lists.debian.org/debian-devel/

Also, a lot of package descriptions in Debian have already been
translated --- see http://ddtp.debian.org/

Also, for general translation, http://www.debian.org/international/l10n/

A lot of the work has been done by Debian, in other words.



signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] cleaner Fink environment?

[Anthony DeRobertis]

On Thu, 2002-06-20 at 06:16, David R. Morrison wrote:

> So I'm wondering if we should get Fink to compile in a cleaner environment.
> There should be some way to fork a shell with none of the environment variables
> inherited by the new shell, I would think?

Yes, see the env command. At least there is an env command on my Debian
box here.



signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] dselect

[Anthony DeRobertis]

On Mon, 2002-06-10 at 22:21, Ward W. Vuillemot wrote:
> I have a very simple question.
> 
> ... What is dselect written in?

C++. Parts of dpkg are writtin in C. 

> If dselect is 
> written in Perl, I would _love_ to know how you got the interface...what 
> modules, anything to get me pointed in the right direction.

I trust you know of CPAN? Look at the ReadLine module. With that, you
can make a shell complete with completion, etc.

If you want to go further, look at ncurses. Browsing the Debian package
list, I see a couple of curses/ncurses packages for perl.

ncurses is the defacto way of doing text-mode GUI.

There are also perl bindings for, e.g., gtk. There might even be aqua
ones; not sure.

> I noticed that dselect is in binary...which 
> has me wondering that it might be something other than perl.  But the 
> web-site has me wondering. . . .

Wonder no more! Use the source --- http://cvs.debian.org/?cvsroot=dpkg
or apt-get source dpkg (not sure if that works with Fink...)



signature.asc
Description: This is a digitally signed message part

[Fink-devel] unstable ethereal not buildable?

[Anthony DeRobertis]

(cd .libs && cc -c -fno-builtin -fno-rtti -fno-exceptions "etherealS.c")
/sw/src/ethereal-0.9.4-2/ethereal-0.9.4/.libs
etherealS.c:1972: syntax error, found `@'
etherealS.c:1972: illegal external declaration, missing `;' after `,'
etherealS.c:1973: illegal external declaration, missing `;' after `*'
etherealS.c:1974: illegal external declaration, missing `;' after `*'

[ And then another several thousand of those ]

I've tried both bash and tcsh.


___

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm

___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Encouragement and Constructive Criticism

[Anthony DeRobertis]

[ I know this is rather late, but... ]

On Thursday, May 16, 2002, at 06:08 , Max Horn wrote:

> It's like testing in debian, it will contains broken packages! 
> You gotta expect it. Not that we are not trying to avoid this 
> and to fix it, but our resources are limited.

Testing (currently "woody") in Debian does not contain 
non-installable packages. The archive maintainence scripts don't 
move non-installable packages to testing. You're thinking of 
unstable ("sid").

There are even rumors that Debian's testing might be of 
releasable quality ;-)

> One more thing: my focus is definitly not on end users. My 
> focus is first on me, and then on users.

Get your priorities strait! There ought to be several more 
things before "end users."

> Sure, I would love that. Give me a DP-800 or faster machine, 
> and pay me a salary, and I will do it immediatly. :-)

LOL.

>
> 1) Asking SF for permission to host 4-6 GB instead of the 2 GB 
> they already endure.

Maybe more space than this. But if this is the biggest hurdle, 
we'd be 95% there.

>
> 2) The tool chain for maintaining a bindist must be updated to 
> support bindist. It must be improved to allow easy use, and be 
> made "fool proof"

Nothing is fool proof. But there is already freely-available 
software that would do most of it, courtesy our friends at 
Debian who do it for over 9,000 packages.

>
> 3) The people that are doing this must be trustworthy.

No more trustworthy than the people who commit the .info and 
.patch files, though.

However, with autobuilders and fakeroot, we could use the 
sourceforge compile farm. [ Note: I'm attempting to port 
fakeroot... not much free time lately, though ]

>
> 4) We need a QA team that checks if this unstable bindist is 
> actually usable

Not really. It would be no worse than today, when we don't have 
checks if source is buildable.

Though if we want it, Debian has code to do it.


___

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm

___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] /sw/include/apt-pkg docs?

[Anthony DeRobertis]

On Saturday, June 1, 2002, at 11:32 , Kyle Moffett wrote:

> Does anybody know where I can find docs or reference stuff or 
> file descriptions of the apt headers in /sw/include/apt-pkg?

The only thing I can find is 



___

Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm

___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] FYI: Porting fakeroot

[Anthony DeRobertis]

On Mon, 2002-05-13 at 10:23, Justin Hallett wrote:
> it doesn't appear to, do you know what headers normally provide this on
> linux?

sys/stat.h

There is some magic in there to make stat64 be called stat if
__USE_FILE_OFFSET64 is defined, too.



signature.asc
Description: This is a digitally signed message part

[Fink-devel] FYI: Porting fakeroot

[Anthony DeRobertis]

Well, I'm trying to port fakeroot over. I've figured out how to do the
LD_PRELOAD equivelant, and I'm trying to get the actual software to
compile.

Then I'll worry about the dyld stuff.

PS: Anyone know if Darwin has stat64 and friends?





signature.asc
Description: This is a digitally signed message part

Re: [Fink-devel] fink vs apt-get

[Anthony DeRobertis]

On Wednesday, January 16, 2002, at 04:54 PM, Finlay Dobbie wrote:

> And compare the number of Debian build servers, Debian 
> donations, and just resources in general.
>
SF provides a compile farm. Could those be used as build servers?


___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel

Re: [Fink-devel] Fink - dpkg - annoying lack of sections in info

[Anthony DeRobertis]

On Friday, January 25, 2002, at 06:11 PM, Max Horn wrote:

> You mean, it is a reported bug in dpkg? Can you point me at 
> some bugreport/old mail/whatever on this?
>

#76100: install-info: ignores section/title from .info file.

It is marked pending upload in the debian bts.

BTW: Take a look at 59425.


(BTW: For anyone who doesn't know how to look at debian bugs, 
see )


___
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel