[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20170908160255.217d0e14@vento.lan>
Date: Fri, 8 Sep 2017 16:02:55 -0300
From: Mauro Carvalho Chehab <mchehab@...pensource.com>
To: Jonathan Corbet <corbet@....net>
Cc: Linux Doc Mailing List <linux-doc@...r.kernel.org>,
Mauro Carvalho Chehab <mchehab@...radead.org>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/2] Two cleanups for Sphinx conf.py
Em Fri, 8 Sep 2017 10:10:26 -0600
Jonathan Corbet <corbet@....net> escreveu:
> On Sun, 3 Sep 2017 16:12:44 -0300
> Mauro Carvalho Chehab <mchehab@...pensource.com> wrote:
>
> > Those patches are just cleanups for Documentation/conf.py.
> >
> > The first one removes a package usage directive for a LateX package
> > that it is not used.
> >
> > The second one moves some LaTeX preamble code that it is used only
> > on Sphinx 1.4 and 1.5 to a separate if. On Sphinx 1.6, the macros there
> > are silently ignored. While the conf.py becomes a little bit more complex,
> > it clearly marks the backward-compatible code, with IMHO, it is a
> > good idea, specially as some day, we may drop support for older Sphinx
> > versions.
> >
> > None of those patches are mandatory. Feel free to apply any time you
> > want (or discard them, if you don't thing it is worth).
>
> OK, these are applied (and the adjustbox one too).
Thanks!
>
> PDF generation seems to work pretty well at the moment, with one
> exception: I don't get a table of contents in any of the files (just a
> blank page saying "Contents"). Are you seeing something different?
That depends on the version. With Sphinx 1.4, the CONTENTS page is
filled with a LaTeX 1-depth ToC (e. g. 2 levels).
With Sphinx 1.6, the first time make pdfdocs is called, it doesn't
produce a CONTENTS page. It does if it is called twice.
I suspect that the difference here is that, with Sphinx up to
version 1.5, the Makefile created under Documentation/output/latex
calls xelatex several times:
%.pdf: %.tex
$(PDFLATEX) $(LATEXOPTS) '$<'
$(PDFLATEX) $(LATEXOPTS) '$<'
$(PDFLATEX) $(LATEXOPTS) '$<'
-$(MAKEINDEX) -s python.ist '$(basename $<).idx'
$(PDFLATEX) $(LATEXOPTS) '$<'
$(PDFLATEX) $(LATEXOPTS) '$<'
With Sphinx 1.6, it calls it only once:
%.pdf: %.tex FORCE_MAKE
$(PDFLATEX) $(LATEXMKOPTS) '$<'
That seems another regression on Sphinx 1.6.
Regards,
Mauro
Powered by blists - more mailing lists