PDF Generation broken.

[Gavin Henry]

Dear all,

I just wanted to bring this to everyones attention. I have been chatting
to Kwade about this over on the #fedora-docs irc channel and he is aware
that this is still broken. It has been since Fedora was announced.

I have testing as much as I can, and xmlto seem to be fine, it's the tex
part that breaks whenever there are itemizedlist's

I am quite happily generating PDF when these are not in the xml doc.

What can I do to fix this?

Also, page headers i.e. the documents name don't span the to of the PDF
properly. Is this a xslt issue?

Thanks,

Gavin.

-- 
Kind Regards,

Gavin Henry.
Managing Director.

T [phone]
M [phone]
F [phone]
E ghenry(a)suretecsystems.com

Open Source. Open Solutions(tm).

http://www.suretecsystems.com/

[Karsten Wade]

On Tue, 2005-01-25 at 14:54 +0000, Gavin Henry wrote:
>  Dear all,
>  
>  I just wanted to bring this to everyones attention. I have been chatting
>  to Kwade about this over on the #fedora-docs irc channel and he is aware
>  that this is still broken. It has been since Fedora was announced.
>  
>  I have testing as much as I can, and xmlto seem to be fine, it's the tex
>  part that breaks whenever there are itemizedlist's
>  
>  I am quite happily generating PDF when these are not in the xml doc.
>  
>  What can I do to fix this? 
I'm poking Tim for some input.  

Do you have some example XML?

>  Also, page headers i.e. the documents name don't span the to of
> the PDF
>  properly. Is this a xslt issue? 
Dunno, haven't had the time to peel back the layers, but I want to Real
Soon Now.  This doesn't help you now, though.

I don't fully understand the issues.  IIRC, we were discussing using fop
instead of the tex backend for pdf generation.  Is one of our barriers
that there is no fop package for Fedora?  And we want it compiled with
gcj?

- Karsten
-- 
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint:  2680 DBFD D968 3141 0115    5F1B D992 0E06 AD0E 0C41   
               IT executives rate Red Hat #1 for value              
                 http://www.redhat.com/promo/vendor/

[Tim Waugh]

On Wed, Jan 26, 2005 at 06:32:34PM -0800, Karsten Wade wrote:

>  I'm poking Tim for some input.   
The xmltex and passivetex applications are no longer supported
upstream.  I think the new way is to use some sort of ConTeXt-based
parser, but don't know the details.

>  I don't fully understand the issues.  IIRC, we were discussing
> using fop
>  instead of the tex backend for pdf generation.  Is one of our barriers
>  that there is no fop package for Fedora?  And we want it compiled with
>  gcj? 
If we had a fop package for Fedora, and someone was willing to make
the necessary adjustments to xmlto, I feel sure that things would be
bound to improve.

Tim.
*/

[Gavin Henry]

<quote who="Tim Waugh">
>  On Wed, Jan 26, 2005 at 06:32:34PM -0800, Karsten Wade wrote:
>
> > I'm poking Tim for some input.
>
>  The xmltex and passivetex applications are no longer supported
>  upstream.  I think the new way is to use some sort of ConTeXt-based
>  parser, but don't know the details.
>
> > I don't fully understand the issues.  IIRC, we were discussing using fop
> > instead of the tex backend for pdf generation.  Is one of our barriers
> > that there is no fop package for Fedora?  And we want it compiled with
> > gcj?
>
>  If we had a fop package for Fedora, and someone was willing to make
>  the necessary adjustments to xmlto, I feel sure that things would be
>  bound to improve. 
When do we start?

>  Tim.
>  */
>  --
>  fedora-docs-list mailing list
>  fedora-docs-list(a)redhat.com
>  To unsubscribe:
>  http://www.redhat.com/mailman/listinfo/fedora-docs-list

[Stuart Ellis]

On Thu, 27 Jan 2005 11:20:19 -0000 (GMT), "Gavin Henry"
<ghenry(a)suretecsystems.com&gt; said:
>  <quote who="Tim Waugh">
>  > On Wed, Jan 26, 2005 at 06:32:34PM -0800, Karsten Wade wrote: 
>  > If we had a fop package for Fedora, and someone was willing to
> make
>  > the necessary adjustments to xmlto, I feel sure that things would be
>  > bound to improve. 
As a practical matter, what should document authors be doing at this
point ?

I'm going to spend some more time this weekend on putting the
screenshots into the Installation Guide, so I can try to figure out
correct dimensions on the EPS files, build PDFs (albeit with broken
headings) and guinea pig anything else that people would like.

On the other hand, if PDF output is Officially Broken then should I be
leaving out the EPS stuff from the DocBook source that I push out ?
--

Stuart Ellis
s.ellis(a)fastmail.co.uk

[Karsten Wade]

On Thu, 2005-01-27 at 12:51 +0000, Stuart Ellis wrote:
>  On Thu, 27 Jan 2005 11:20:19 -0000 (GMT), "Gavin Henry"
>  <ghenry(a)suretecsystems.com&gt; said:
>  > <quote who="Tim Waugh">
>  > > On Wed, Jan 26, 2005 at 06:32:34PM -0800, Karsten Wade wrote:
>  
>  > > If we had a fop package for Fedora, and someone was willing to make
>  > > the necessary adjustments to xmlto, I feel sure that things would be
>  > > bound to improve.
>  
>  As a practical matter, what should document authors be doing at this
>  point ?
>  
>  I'm going to spend some more time this weekend on putting the
>  screenshots into the Installation Guide, so I can try to figure out
>  correct dimensions on the EPS files, build PDFs (albeit with broken
>  headings) and guinea pig anything else that people would like.
>  
>  On the other hand, if PDF output is Officially Broken then should I be
>  leaving out the EPS stuff from the DocBook source that I push out ? 
Personally, I'd make the doc PDF capable while it was fresh and in front
of me, rather then going in to backfill that work later.  Unless time
required otherwise.  In the long run, the time won't be wasted.  PDF
will eventually work.

OTOH, if you want dispensation to drop the EPS work because PDF is
broken, by all means, go ahead. ;-)  It won't be _that_ much harder to
do it later instead.

- Karsten
-- 
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint:  2680 DBFD D968 3141 0115    5F1B D992 0E06 AD0E 0C41   
               IT executives rate Red Hat #1 for value              
                 http://www.redhat.com/promo/vendor/

[Stuart Ellis]

On Thu, 27 Jan 2005 14:34:27 -0800, "Karsten Wade" <kwade(a)redhat.com&gt;
said:
>  Personally, I'd make the doc PDF capable while it was fresh and in front
>  of me, rather then going in to backfill that work later.  Unless time
>  required otherwise.  In the long run, the time won't be wasted.  PDF
>  will eventually work.
>  
>  OTOH, if you want dispensation to drop the EPS work because PDF is
>  broken, by all means, go ahead. ;-)  It won't be _that_ much harder to
>  do it later instead. 
You've pinpointed my inner conflict :-)  I would like this to work and I
don't mind putting some time in, but it seems a waste to prep and ship a
bunch of files that don't do anything.  I'll do them and make sure the
PDF builds, and will perhaps ship the test release without them if the
functionality is still broken.  Those of us poor souls who are still on
dial-up are very conscious about file sizes :-)
--

Stuart Ellis
s.ellis(a)fastmail.co.uk