R: Trouble with it-pdf generation ?

Federico Nebiacolombo federico.nebiacolombo at softeco.it
Thu Nov 9 01:29:13 CST 2006


> The next step would be to work out precisely which bits of the XML are
> triggering the weirdness, but it's getting late here, so I probably
> won't get around to that today.
Yes, I look in log files sent on mail list by build process, but it seems
that we have only info
about pdf missing.
Of course this is a good information but no details about what is the cause.
I agree about your idea: it could be simply a bug in PDF generator.
I will check diff in xml source taking HEAD vs r2498 if I will have any idea
I will
let you known.
Of course if we can do something to help you, ask us.

best regards
Nebiac

> -----Messaggio originale-----
> Da: Max Bowsher [mailto:maxb1 at ukf.net]
> Inviato: Wednesday, November 08, 2006 23:57
> A: svnbook-dev at red-bean.com
> Cc: Federico Nebiacolombo; Svn-It-mail List
> Oggetto: Re: Trouble with it-pdf generation ?
>
>
> Max Bowsher wrote:
> > Federico Nebiacolombo wrote:
> >> Dear all,
> >>         The link http://svnbook.red-bean.com/nightly/it/svn-book.pdf is
> >> actually broken.
> >> We did not received any report from automatic build system.
> >
> > Erm, oops. Looking back at the script I see it sends mail only if the
> > entire process fails. But the HTML generation succeeded, so it decided
> > things were OK :-/
>
> Fixed now to detect make errors.
>
> >> If this can help you: the problem is only related to it-pdf.
> >> I think it could be related to some xml validation problem but
> it seems book
> >> src are ok,
> >> If we can help you in some way let us known.
> >
> > Looking at:
> > http://svnbook.red-bean.com/nightly/nightly-build.it.log
> >
> > it seems that FOP just died without giving any reason at all :-(
>
> After some further investigation, I have found that rolling back
> ch08.xml and appb.xml to r2498 avoids the problem.
>
> Since the XML validation does not complain, I am inclined to suspect
> that FOP (the PDF generator) simply has a bug causing the generation to
> fail.
>
> The next step would be to work out precisely which bits of the XML are
> triggering the weirdness, but it's getting late here, so I probably
> won't get around to that today.
>
> Max.
>
>







More information about the svn-it mailing list