adding output specific annotations to headings in revealjs mode

[psteinb]

Hi - pandocfilters is a great tool to help me making slides based on markdown but producing any output. I am currently struggling to write a pandocfilter in python that allows me to do the following

# Any Title {.data-background="path/to/image.png"}
## Sub-slide 1
## Sub-slide 2

which then results in

<section data-background="path/to/image.png">
<h1>Any Title</h1>
<section><h2>Sub-slide 1</h2></section>
<section><h2>Sub-slide 2</h2></section>
</section>

inside the reveal.js based output. Should I just look for the Header and replace it with raw HTML? How is the closing `' taken care of then? The whole idea behind this would be similar to the {.fragile} annotation in latex beamer. (see http://pandoc.org/demo/example9/producing-slide-shows-with-pandoc.html at the bottom)

Best, P

[psteinb]

Digging a bit further, the following seems to do the job

from pandocfilters import toJSONFilter
from pandocfilters import RawBlock
from pandocfilters import Header
from pandocfilters import stringify

import re 
marked_re = re.compile('\{\.[a-z]+.*=.*\}')

def html(x):
    return RawBlock('html', x)

def mk_data_background(key, value, fmat, meta):
    if key == "Header":
        #print stringify(value)
        [ident, brief, contents] = value

        str_content = stringify(contents)

        if not marked_re.search(str_content):
            return Header(ident, brief, contents)

        new_content = str_content[:str_content.find("{.")]
        marked_content = str_content[str_content.find("{.")+2:str_content.rfind("}")]

        if marked_content:
            html_text = r'<section %s><h1>%s</h1>' % (marked_content, new_content)
            return html(html_text)

if __name__ == "__main__":
    toJSONFilter(mk_data_background)

But the html has exactly the problem I expected:

<section>
    <h1 class="title">Testing data-background</h1>
    <h2 class="author">Peter Steinbach, steinbac@mpi-cbg</h2>
    <h3 class="date">April 30, 2015</h3>
</section>

<section class="slide level6">

<section data-background="../images/PrincipleFusionDeconvolution.png"><h1>Fusion vs Deconvolution </h1>
</section>
<section><section id="slide-1" class="titleslide slide level2"><h1>slide 1</h1></section></section>
<section><section id="slide-2" class="titleslide slide level2"><h1>slide 2</h1></section></section>

Any ideas?

[jgm]

I suggest you post questions like this on pandoc-discuss at googlegroups. There are lots of people there who can help with filter questions; very few people look at this tracker and it's really for bug reports, not questions.

+++ Peter Steinbach [Jun 09 15 01:58 ]:

Hi - pandocfilters is a great tool to help me making slides based on markdown but producing any output. I am currently struggling to write a pandocfilter in python that allows me to do the following

# Any Title {.data-background="path/to/image.png"}
## Sub-slide 1
## Sub-slide 2

which then results in

<section data-background="path/to/image.png">
<h1>Any Title</h1>
<section><h2>Sub-slide 1</h2></section>
<section><h2>Sub-slide 2</h2></section>
</section>

inside the reveal.js based output. Should I just look for the Header and replace it with raw HTML? How is the closing `' taken care of then? The whole idea behind this would be similar to the {.fragile} annotation in latex beamer. (see http://pandoc.org/demo/example9/producing-slide-shows-with-pandoc.html at the bottom)

Best, P

---

Reply to this email directly or view it on GitHub: https://github.com/jgm/pandocfilters/issues/18

[psteinb]

fair enough, I wasn't aware of this. Thanks. Keep going! P

On 06/09/2015 06:42 PM, John MacFarlane wrote:

I suggest you post questions like this on pandoc-discuss at googlegroups. There are lots of people there who can help with filter questions; very few people look at this tracker and it's really for bug reports, not questions.

+++ Peter Steinbach [Jun 09 15 01:58 ]:

Hi - pandocfilters is a great tool to help me making slides based on markdown but producing any output. I am currently struggling to write a pandocfilter in python that allows me to do the following

# Any Title {.data-background="path/to/image.png"}
## Sub-slide 1
## Sub-slide 2

which then results in

<section data-background="path/to/image.png">
<h1>Any Title</h1>
<section><h2>Sub-slide 1</h2></section>
<section><h2>Sub-slide 2</h2></section>
</section>

inside the reveal.js based output. Should I just look for the Header and replace it with raw HTML? How is the closing `' taken care of then? The whole idea behind this would be similar to the {.fragile} annotation in latex beamer. (see http://pandoc.org/demo/example9/producing-slide-shows-with-pandoc.html at the bottom)

Best, P

---

Reply to this email directly or view it on GitHub: https://github.com/jgm/pandocfilters/issues/18

---

Reply to this email directly or view it on GitHub: https://github.com/jgm/pandocfilters/issues/18#issuecomment-110427582

[psteinb]

Just to resolve this, I stumbled over a hint that the functionality I'm interested in already works with plain pandoc:

# Any Title {data-background="path/to/image.png"}
## Sub-slide 1
## Sub-slide 2