Thread: [Docutils-users] LaTeX and html vector graphics from same rst

I wish to create both epub and pdf (for printing) versions of my book 
from the same rst source. I create html for the epub and latex for the 
pdf. The formats of the figures (vector graphics) need to be different 
for each: svg for html; eps for latex.

I have created the source:

.. figure:: book_figs/decoderRing
    :align: center
    :class: latex

.. figure:: book_figs/decoderRing.svg
    :align: center
    :class: epub

When I use rst2html.py --strip-elements-with-class=latex to get html I get:

<div class="figure align-center">
</div>
<div class="figure align-center">
<object class="epub" data="book_figs/decoderRing.svg" type="image/svg+xml">
book_figs/decoderRing.svg</object>
</div>

And when I use rst2latex.py --strip-elements-with-class=epub to get html 
I get:

\begin{figure}
\noindent\makebox[\textwidth][c]{\includegraphics{book_figs/decoderRing}}
\end{figure}
\begin{figure}\end{figure}

(For those unfamiliar with latex, it assumes the .eps extension if none 
is specified.)

In both cases I have to remove the "null" figures for each file.

This is acceptable; I will have to do other tweaking anyway. But if 
there's a better way to achieve this "if-else" construct, I would like 
to learn it.

--Bob

On 2011-11-22, Bob Plantz wrote:
> I wish to create both epub and pdf (for printing) versions of my book
> from the same rst source. I create html for the epub and latex for the
> pdf. The formats of the figures (vector graphics) need to be different
> for each: svg for html; eps for latex.

> I have created the source:

> .. figure:: book_figs/decoderRing
>     :align: center
>     :class: latex

> .. figure:: book_figs/decoderRing.svg
>     :align: center
>     :class: epub

> When I use rst2html.py --strip-elements-with-class=latex to get html I get:

...
><div class="figure align-center">
></div>
...

> In both cases I have to remove the "null" figures for each file.

> This is acceptable; I will have to do other tweaking anyway. But if
> there's a better way to achieve this "if-else" construct, I would like
> to learn it.

You choose a difficult element to start with. The "figure" directive passes
most arguments to the included image object. Use

figclass : text
    Set a "classes" attribute value on the figure element.

See the "directives" doc at
docutils/docs/ref/rst/directives.html#figure

Günter

On 11/22/2011 11:53 PM, Guenter Milde wrote:
> On 2011-11-22, Bob Plantz wrote:
>> I wish to create both epub and pdf (for printing) versions of my book
>> from the same rst source. I create html for the epub and latex for the
>> pdf. The formats of the figures (vector graphics) need to be different
>> for each: svg for html; eps for latex.
>> I have created the source:
>> .. figure:: book_figs/decoderRing
>>      :align: center
>>      :class: latex
>> .. figure:: book_figs/decoderRing.svg
>>      :align: center
>>      :class: epub
>> When I use rst2html.py --strip-elements-with-class=latex to get html I get:
> ...
>> <div class="figure align-center">
>> </div>
> ...
>
>> In both cases I have to remove the "null" figures for each file.
>> This is acceptable; I will have to do other tweaking anyway. But if
>> there's a better way to achieve this "if-else" construct, I would like
>> to learn it.
> You choose a difficult element to start with. The "figure" directive passes
> most arguments to the included image object. Use
>
> figclass : text
>      Set a "classes" attribute value on the figure element.
>
> See the "directives" doc at
> docutils/docs/ref/rst/directives.html#figure
>
> Günter
>
Thank you. I should have read the docs more carefully. Using the 
following works quite well:

.. figure:: book_figs/decoderRing
    :align: center
    :figclass: latex

.. figure:: book_figs/decoderRing.svg
    :align: center
    :figclass: epub

--Bob