headers
Tassilo Horn | 19 Jul 2012 19:42
Picon
Gravatar

First try at the Biber support merge

Hi all,

I've just done my first try at merging Philips Biber support changes,
simply by ediffing his files agains the current CVS trunk and merging
everything that looked Biber-related.  I might have missed something,
and maybe I even destroyed existing functionality, but I tried to be as
conservative as possible.

At least, it seems that I don't have broken the RefTeX support.  And
even better, now I also have completion at the citation prompt! :-)

So Biber users, could you please try applying the attached patch against
AUCTeX from CVS and check if the Biber stuff also works as expected?

Bye,
Tassilo
Attachment (biber-merge-1st-try.patch): text/x-patch, 12 KiB
_______________________________________________
auctex mailing list
auctex <at> gnu.org
https://lists.gnu.org/mailman/listinfo/auctex
Permalink | Reply |
headers
Marcus Frings | 20 Jul 2012 10:47
Face

Re: First try at the Biber support merge

* Tassilo Horn <tassilo <at> member.fsf.org> wrote:

> I've just done my first try at merging Philips Biber support changes,
> simply by ediffing his files agains the current CVS trunk and merging
> everything that looked Biber-related.

Great, thank you very much for adding the patches.

> At least, it seems that I don't have broken the RefTeX support.  And
> even better, now I also have completion at the citation prompt! :-)

:-)

> So Biber users, could you please try applying the attached patch against
> AUCTeX from CVS and check if the Biber stuff also works as expected?

The patch file applied cleanly to my CVS version. At the moment I'm
working with it and everything seems to be fine. If I found a weird
behaviour or a bug I'll report it here!

Gruß,
Marcus
--

-- 
Fry:   "I have an idea."
Leela: "I have a better one."

_______________________________________________
auctex mailing list
auctex <at> gnu.org
https://lists.gnu.org/mailman/listinfo/auctex
(Continue reading)

Permalink | Reply |
headers
Tassilo Horn | 20 Jul 2012 11:42
Picon
Gravatar

Re: First try at the Biber support merge

Marcus Frings <iam-est-hora-surgere <at> despammed.com> writes:

>> So Biber users, could you please try applying the attached patch
>> against AUCTeX from CVS and check if the Biber stuff also works as
>> expected?
>
> The patch file applied cleanly to my CVS version. At the moment I'm
> working with it and everything seems to be fine. If I found a weird
> behaviour or a bug I'll report it here!

Great.

My main problem when compiling that patch was that there were a lot of
differences between Philips files and the AUCTeX CVS version that didn't
look Biber-related.  Maybe that were general improvements and fixes of
the AUCTeX code, or maybe it was just an older AUCTeX version he worked
on.  Since I couldn't judge that, I preferred the stock AUCTeX versions
in those cases.

Bye,
Tassilo
Permalink | Reply |
headers
Ken Brown | 20 Jul 2012 14:17
Picon
Favicon

Re: First try at the Biber support merge

On 7/20/2012 5:42 AM, Tassilo Horn wrote:
> Marcus Frings <iam-est-hora-surgere <at> despammed.com> writes:
>
>>> So Biber users, could you please try applying the attached patch
>>> against AUCTeX from CVS and check if the Biber stuff also works as
>>> expected?
>>
>> The patch file applied cleanly to my CVS version. At the moment I'm
>> working with it and everything seems to be fine. If I found a weird
>> behaviour or a bug I'll report it here!
>
> Great.
>
> My main problem when compiling that patch was that there were a lot of
> differences between Philips files and the AUCTeX CVS version that didn't
> look Biber-related.  Maybe that were general improvements and fixes of
> the AUCTeX code, or maybe it was just an older AUCTeX version he worked
> on.  Since I couldn't judge that, I preferred the stock AUCTeX versions
> in those cases.

Tassilo,

My impression is that Philip's files were based on auctex-11.86, but he 
could confirm.  Based on that assumption, I did a 3-way merge (using 
diff3) of Philip's files and CVS head, using auctex-11.86 as common 
ancestor.  My patches are different from yours in a few places.

I don't want to waste everyone's time, so I'll do a little testing 
before sending my patches to the list.
(Continue reading)

Permalink | Reply |
headers
Tassilo Horn | 21 Jul 2012 01:26
Picon
Gravatar

Re: First try at the Biber support merge

Ken Brown <kbrown <at> cornell.edu> writes:

Hi Ken,

>> My main problem when compiling that patch was that there were a lot
>> of differences between Philips files and the AUCTeX CVS version that
>> didn't look Biber-related.  Maybe that were general improvements and
>> fixes of the AUCTeX code, or maybe it was just an older AUCTeX
>> version he worked on.  Since I couldn't judge that, I preferred the
>> stock AUCTeX versions in those cases.
>
> My impression is that Philip's files were based on auctex-11.86, but
> he could confirm.  Based on that assumption, I did a 3-way merge
> (using diff3) of Philip's files and CVS head, using auctex-11.86 as
> common ancestor.  My patches are different from yours in a few places.

Awesome, I guess your results will probably be more accurate than mine.

> I don't want to waste everyone's time, so I'll do a little testing
> before sending my patches to the list.

Sure.

Bye,
Tassilo
Permalink | Reply |
headers
Ken Brown | 21 Jul 2012 12:36
Picon
Favicon

Re: First try at the Biber support merge

Hi Tassilo,

On 7/20/2012 7:26 PM, Tassilo Horn wrote:
> Ken Brown <kbrown <at> cornell.edu> writes:
>
> Hi Ken,
>
>>> My main problem when compiling that patch was that there were a lot
>>> of differences between Philips files and the AUCTeX CVS version that
>>> didn't look Biber-related.  Maybe that were general improvements and
>>> fixes of the AUCTeX code, or maybe it was just an older AUCTeX
>>> version he worked on.  Since I couldn't judge that, I preferred the
>>> stock AUCTeX versions in those cases.
>>
>> My impression is that Philip's files were based on auctex-11.86, but
>> he could confirm.  Based on that assumption, I did a 3-way merge
>> (using diff3) of Philip's files and CVS head, using auctex-11.86 as
>> common ancestor.  My patches are different from yours in a few places.
>
> Awesome, I guess your results will probably be more accurate than mine.
>
>> I don't want to waste everyone's time, so I'll do a little testing
>> before sending my patches to the list.

My version works for me, so I'm attaching it in two forms; one is a 
unified diff, which I find easier to read, and the other is in the same 
format as yours, so you can more easily see where it differs.

Philip did confirm in private mail that his patches were based on 11.86. 
  I hope he can take a quick look and see if what I did seems right.
(Continue reading)

Permalink | Reply |
headers
Philip Kime | 21 Jul 2012 19:07
Picon

Re: First try at the Biber support merge

Many thanks for looking at this. It's hard to tell from the patches if they are right - some people will have to
install and test it. Can I just pull this from the CVS head? One thing to note is that many users of biblatex
2.0 will want to set:

(setq LaTeX-always-use-Biber t)

Since biber is the default backend as of version 2.0 of biblatex and most of the really important features
are biber only in 2.0.

PK

On 21 Jul 2012, at 12:36 PM, Ken Brown wrote:

> Hi Tassilo,
> 
> On 7/20/2012 7:26 PM, Tassilo Horn wrote:
>> Ken Brown <kbrown <at> cornell.edu> writes:
>> 
>> Hi Ken,
>> 
>>>> My main problem when compiling that patch was that there were a lot
>>>> of differences between Philips files and the AUCTeX CVS version that
>>>> didn't look Biber-related.  Maybe that were general improvements and
>>>> fixes of the AUCTeX code, or maybe it was just an older AUCTeX
>>>> version he worked on.  Since I couldn't judge that, I preferred the
>>>> stock AUCTeX versions in those cases.
>>> 
>>> My impression is that Philip's files were based on auctex-11.86, but
>>> he could confirm.  Based on that assumption, I did a 3-way merge
>>> (using diff3) of Philip's files and CVS head, using auctex-11.86 as
(Continue reading)

Permalink | Reply |
headers
Ken Brown | 21 Jul 2012 19:53
Picon
Favicon

Re: First try at the Biber support merge

On 7/21/2012 1:07 PM, Philip Kime wrote:
> Many thanks for looking at this. It's hard to tell from the patches if they are right - some people will have
to install and test it. Can I just pull this from the CVS head?

Just pull CVS head and apply the patches.

> One thing to note is that many users of biblatex 2.0 will want to set:
>
> (setq LaTeX-always-use-Biber t)
>
> Since biber is the default backend as of version 2.0 of biblatex and most of the really important features
are biber only in 2.0.

So I guess that advice should go into the auctex manual, as well as the 
release notes for the next release.

Ken
Permalink | Reply |
headers
Tassilo Horn | 22 Jul 2012 11:01
Picon
Gravatar

Re: First try at the Biber support merge

Ken Brown <kbrown <at> cornell.edu> writes:

>> Many thanks for looking at this. It's hard to tell from the patches
>> if they are right - some people will have to install and test it. Can
>> I just pull this from the CVS head?
>
> Just pull CVS head and apply the patches.
>
>> One thing to note is that many users of biblatex 2.0 will want to set:
>>
>> (setq LaTeX-always-use-Biber t)
>>
>> Since biber is the default backend as of version 2.0 of biblatex and most of
>> the really important features are biber only in 2.0.
>
> So I guess that advice should go into the auctex manual, as well as
> the release notes for the next release.

Yes, the release notes are in doc/changes.texi which is both manual, and
the text only CHANGES is generated from it.  But probably there should
only be a statement that biber's now supported, and all usage docs
should be an own section in the manual.

Bye,
Tassilo
Permalink | Reply |
headers
Tassilo Horn | 22 Jul 2012 11:07
Picon
Gravatar

Re: First try at the Biber support merge

Ken Brown <kbrown <at> cornell.edu> writes:

>>> I don't want to waste everyone's time, so I'll do a little testing
>>> before sending my patches to the list.
>
> My version works for me, so I'm attaching it in two forms; one is a
> unified diff, which I find easier to read, and the other is in the
> same format as yours, so you can more easily see where it differs.

Thanks.  I'm using your patch now for a few days and see if it works for
me.

Bye,
Tassilo
Permalink | Reply |
headers
Ettore Aldrovandi | 24 Jul 2012 22:59
Picon
Favicon

Re: [AUCTeX] First try at the Biber support merge

On Sun, Jul 22, 2012 at 11:07:58AM +0200, Tassilo Horn wrote:

> Thanks.  I'm using your patch now for a few days and see if it works for
> me.

Hello,

I have been following this, and wanted to try the patches on OS X. 
After rebuilding auctex, they work well with a stock Emacs.app.

With Aquamacs.app, however, while the LaTeX-Biber-LaTeX-LaTeX cycle
works (as above), the View command breaks. There is some code designed to
configure auctex in order to use Skim as a viewer application, and that
now fails.  Specifically, it's the aquamacs-call-viewer function, but I
need guidance to understand why.

I'm posting this to the macosx-emacs list as well.

--Ettore

--

-- 
Ettore Aldrovandi
Department of Mathematics, Florida State University
1017 Academic Way		 *    http://www.math.fsu.edu/~ealdrov
Tallahassee, FL 32306-4510, USA * * aldrovandi at math dot fsu dot edu

_____________________________________________________________
MacOSX-Emacs mailing list
MacOSX-Emacs <at> email.esm.psu.edu
(Continue reading)

Permalink | Reply |
headers
David Reitter | 25 Jul 2012 00:36
Picon
Gravatar

Re: [AUCTeX] First try at the Biber support merge

On Jul 24, 2012, at 4:59 PM, Ettore Aldrovandi wrote:
> 
> I have been following this, and wanted to try the patches on OS X. 
> After rebuilding auctex, they work well with a stock Emacs.app.
> 
> With Aquamacs.app, however, while the LaTeX-Biber-LaTeX-LaTeX cycle
> works (as above), the View command breaks. There is some code designed to
> configure auctex in order to use Skim as a viewer application, and that
> now fails.  Specifically, it's the aquamacs-call-viewer function, but I
> need guidance to understand why.

Where is this patch?

If you trace this function, you may be able to find out where (and why) it fails.

- David_____________________________________________________________
MacOSX-Emacs mailing list
MacOSX-Emacs <at> email.esm.psu.edu
http://email.esm.psu.edu/mailman/listinfo/macosx-emacs
List Archives: http://dir.gmane.org/gmane.emacs.macintosh.osx
Permalink | Reply |
headers
Ettore Aldrovandi | 25 Jul 2012 01:29
Picon
Favicon

Re: [AUCTeX] First try at the Biber support merge

On Tue, Jul 24, 2012 at 06:36:23PM -0400, David Reitter wrote:

> Where is this patch?
> 
> If you trace this function, you may be able to find out where (and why) it fails.
> 
> - David

Oh, sorry: it's on the auctex <at> gnu.org mailing list.  See:

http://lists.gnu.org/archive/html/auctex/2012-07/msg00025.html

(I used the mbox format, the text from the web browser causes a bad patch.)

It appears the patch is going to go into the auctex CVS, since the compiling cycle seems to work well.

--Ettore

--

-- 
Ettore Aldrovandi
Department of Mathematics, Florida State University
1017 Academic Way		 *    http://www.math.fsu.edu/~ealdrov
Tallahassee, FL 32306-4510, USA * * aldrovandi at math dot fsu dot edu

_____________________________________________________________
MacOSX-Emacs mailing list
MacOSX-Emacs <at> email.esm.psu.edu
http://email.esm.psu.edu/mailman/listinfo/macosx-emacs
List Archives: http://dir.gmane.org/gmane.emacs.macintosh.osx
(Continue reading)

Permalink | Reply |
headers
Ettore Aldrovandi | 26 Jul 2012 01:28
Picon
Favicon

Re: [AUCTeX] First try at the Biber support merge

On Tue, Jul 24, 2012 at 06:36:23PM -0400, David Reitter wrote:

> If you trace this function, you may be able to find out where (and why) it fails.

I'm still not  getting anywhere (see the other messages to the list).

I have one question: short of rebuilding the whole app, what is the proper way to get a patched AUCTeX into the
app bundle?

Thanks,
--

-- 
Ettore

_____________________________________________________________
MacOSX-Emacs mailing list
MacOSX-Emacs <at> email.esm.psu.edu
http://email.esm.psu.edu/mailman/listinfo/macosx-emacs
List Archives: http://dir.gmane.org/gmane.emacs.macintosh.osx
Permalink | Reply |
headers
Tassilo Horn | 25 Jul 2012 08:50
Picon
Gravatar

Re: First try at the Biber support merge

Ettore Aldrovandi <ealdrov <at> math.fsu.edu> writes:

Hi Ettore,

>> Thanks.  I'm using your patch now for a few days and see if it works
>> for me.
>
> I have been following this, and wanted to try the patches on OS X.
> After rebuilding auctex, they work well with a stock Emacs.app.
>
> With Aquamacs.app, however, while the LaTeX-Biber-LaTeX-LaTeX cycle
> works (as above), the View command breaks. There is some code designed
> to configure auctex in order to use Skim as a viewer application, and
> that now fails.  Specifically, it's the aquamacs-call-viewer function,
> but I need guidance to understand why.

Do you have a backtrace, or in what way does it fail?

Looking at the patch, the only function in that procedure that was
touched is `TeX-command-query', but to me it looks unlikely to be the
cause.  Are you sure that the problem doesn't also occur in aquamacs
without Ken's biber patch?

Bye,
Tassilo
Permalink | Reply |
headers
Ettore Aldrovandi | 25 Jul 2012 20:12
Picon
Favicon

Re: First try at the Biber support merge

On Wed, Jul 25, 2012 at 08:50:36AM +0200, Tassilo Horn wrote:

> Do you have a backtrace, or in what way does it fail?
> 
> Looking at the patch, the only function in that procedure that was
> touched is `TeX-command-query', but to me it looks unlikely to be the

Here's the trace on TeX-command-query ("ringstacks" is the file I'm using) if Skim is already running:

======================================================================
1 -> TeX-command-master: override-confirm=nil
| 2 -> TeX-command-query: name="ringstacks"
| 2 <- TeX-command-query: "View"
| 2 -> TeX-command: name="View" file=TeX-master-file override-confirm=nil
| 2 <- TeX-command: nil
1 <- TeX-command-master: nil
======================================================================

At the 2 -> Tex-command: stage, under normal behavior, I should have a
Skim window pop up. What I get is:

View command: (aquamacs-call-viewer 815 "ringstacks.tex")

in the mini-buffer. Then it returns nil.

If Skim is not running, then I get:

======================================================================
1 -> TeX-command-master: override-confirm=nil
| 2 -> TeX-command-query: name="ringstacks"
(Continue reading)

Permalink | Reply |
headers
Tassilo Horn | 25 Jul 2012 20:28
Picon
Gravatar

Re: First try at the Biber support merge

Ettore Aldrovandi <ealdrov <at> math.fsu.edu> writes:

Hi Ettore,

> View command: (aquamacs-call-viewer 815 "ringstacks.tex")
>
> in the mini-buffer. Then it returns nil.

What view command do you get with the unpatched CVS version?

Bye,
Tassilo
Permalink | Reply |
headers
Ettore Aldrovandi | 25 Jul 2012 22:53
Picon
Favicon

Re: First try at the Biber support merge

On Wed, Jul 25, 2012 at 08:28:08PM +0200, Tassilo Horn wrote:

> Ettore Aldrovandi <ealdrov <at> math.fsu.edu> writes:
> 
> > View command: (aquamacs-call-viewer 815 "ringstacks.tex")
> >
> > in the mini-buffer. Then it returns nil.
> 
> What view command do you get with the unpatched CVS version?

AHA! I get the same behavior!  So the problem is with aquamacs init
code for auctex vs. CVS version.  Good catch, I should have thought of
running this check.  I'll try to apply the patches to 11.86 and see
what happens.

Sorry for the (partial) noise.

--Ettore

--

-- 
Ettore Aldrovandi
Department of Mathematics, Florida State University
1017 Academic Way		 *    http://www.math.fsu.edu/~ealdrov
Tallahassee, FL 32306-4510, USA * * aldrovandi at math dot fsu dot edu
Permalink | Reply |
headers
Ken Brown | 26 Jul 2012 00:24
Picon
Favicon

Re: First try at the Biber support merge

On 7/25/2012 4:53 PM, Ettore Aldrovandi wrote:
> On Wed, Jul 25, 2012 at 08:28:08PM +0200, Tassilo Horn wrote:
>
>> Ettore Aldrovandi <ealdrov <at> math.fsu.edu> writes:
>>
>>> View command: (aquamacs-call-viewer 815 "ringstacks.tex")
>>>
>>> in the mini-buffer. Then it returns nil.
>>
>> What view command do you get with the unpatched CVS version?
>
> AHA! I get the same behavior!  So the problem is with aquamacs init
> code for auctex vs. CVS version.  Good catch, I should have thought of
> running this check.  I'll try to apply the patches to 11.86 and see
> what happens.

The patches won't apply.  But you can just use Philip's files directly:

   http://sourceforge.net/projects/biblatex-biber/files/auctex-biber.zip

Ken
Permalink | Reply |
headers
Ettore Aldrovandi | 26 Jul 2012 00:36
Picon
Favicon

Re: First try at the Biber support merge

On Wed, Jul 25, 2012 at 06:24:25PM -0400, Ken Brown wrote:
> 
> The patches won't apply.  But you can just use Philip's files directly:
> 
>   http://sourceforge.net/projects/biblatex-biber/files/auctex-biber.zip

Yes, that's what I have. Thanks,

--Ettore 

--

-- 
Ettore Aldrovandi
Department of Mathematics, Florida State University
1017 Academic Way		 *    http://www.math.fsu.edu/~ealdrov
Tallahassee, FL 32306-4510, USA * * aldrovandi at math dot fsu dot edu
Permalink | Reply |
headers
Ken Brown | 25 Jul 2012 20:49
Picon
Favicon

Re: First try at the Biber support merge

On 7/25/2012 2:50 AM, Tassilo Horn wrote:
> Looking at the patch, the only function in that procedure that was
> touched is `TeX-command-query', but to me it looks unlikely to be the
> cause.

I agree that the changes to TeX-command-query look harmless.  But 
TeX-command-query calls TeX-check-files, and Philip made quite a few 
changes to the latter.  Could the problem be in that function?

Ken
Permalink | Reply |
headers
Tassilo Horn | 25 Jul 2012 21:23
Picon
Gravatar

Re: First try at the Biber support merge

Ken Brown <kbrown <at> cornell.edu> writes:

>> Looking at the patch, the only function in that procedure that was
>> touched is `TeX-command-query', but to me it looks unlikely to be the
>> cause.
>
> I agree that the changes to TeX-command-query look harmless.  But
> TeX-command-query calls TeX-check-files, and Philip made quite a few
> changes to the latter.  Could the problem be in that function?

I don't think so.  It seems to be used only to figure out the next TeX
command that would make sense.  If the user selects View explicitly,
that shouldn't have an effect.

Bye,
Tassilo
Permalink | Reply |
headers
Ken Brown | 25 Jul 2012 21:33
Picon
Favicon

Re: First try at the Biber support merge

On 7/25/2012 3:23 PM, Tassilo Horn wrote:
> Ken Brown<kbrown <at> cornell.edu>  writes:
>
>>> Looking at the patch, the only function in that procedure that was
>>> touched is `TeX-command-query', but to me it looks unlikely to be the
>>> cause.
>>
>> I agree that the changes to TeX-command-query look harmless.  But
>> TeX-command-query calls TeX-check-files, and Philip made quite a few
>> changes to the latter.  Could the problem be in that function?
>
> I don't think so.  It seems to be used only to figure out the next TeX
> command that would make sense.  If the user selects View explicitly,
> that shouldn't have an effect.

OK.

BTW, it seems surprising that Philip rewrote TeX-check-files, unless he 
thought there was a bug in it.  I don't see anything specific to Biber 
or BibTeX either in the original version or in his rewrite.  (That 
explains why you didn't include it in your version of the biber-merge 
patch.)

Philip, am I missing something?

Ken
Permalink | Reply |
headers
Ettore Aldrovandi | 26 Jul 2012 20:19
Picon
Favicon

Re: [OS X Emacs] Re: First try at the Biber support merge

On Tue, Jul 24, 2012 at 04:59:54PM -0400, Ettore Aldrovandi wrote:

> With Aquamacs.app, however, while the LaTeX-Biber-LaTeX-LaTeX cycle
> works (as above), the View command breaks. There is some code designed to
> configure auctex in order to use Skim as a viewer application, and that
> now fails. 

OK, all seems to be well now. I still don't understand some of the
details, but it turns out the auctex-11.86 in aquamacs are (slightly)
patched. Thus I diff3'ed  the aquamacs vs. stock 11.86 vs. Philip's
files and generated a new set which appears to work. In fact there were
no overlaps, so they merged cleanly.  The (now aquamacs specific) files
are at:

http://www.math.fsu.edu/~ealdrov/files/auctex-aquamacs-biber.zip

Thanks to all,

--Ettore

--

-- 
Ettore Aldrovandi
Department of Mathematics, Florida State University
1017 Academic Way		 *    http://www.math.fsu.edu/~ealdrov
Tallahassee, FL 32306-4510, USA * * aldrovandi at math dot fsu dot edu
Permalink | Reply |

Gmane