Den mån 26 juni 2023 21:12Ioan Muntean <imuntean-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> skrev:

Dear BPJ
I read this post and thank you for illuminating some aspects of walking a DIV element in pandoc. My story is different but I tried to achieve something simpler.
I have a docx document and I want to convert it to Latex to include some special styles. . I adapted your code and all works well except that I get extra lines between the environment definition and content.

Sorry for the delay. At first I didn't notice your post and then I was very busy.

The reason for the "extra" whitespace is that the raw text before and after the div are separate blocks, and Pandoc inserts a blank space between blocks because usually it is both wanted and necessary. The workaround for that is to convert the div to latex while running the filter and concatenate it with the prefix and postfix into a single raw latex block. You can see how to do that in the attached filter, which maybe will work for you as is. It is a slimmed-down version of an undocumented filter which does the same for either classes or arbitrary attributes and for multiple output formats, but with the whitespace-avoiding stuff added.

Here is my adapted code:

function Div(divclaims)

local preclaims = pandoc.RawInline('latex', '\\begin{claims}')
local postclaims = pandoc.RawInline('latex', '\\end{claims}')
local preissues = pandoc.RawInline('latex', '\\begin{issues}')
local postissues = pandoc.RawInline('latex', '\\end{issues}')

divbe=tostring(divclaims.t)
styletobe=tostring(divclaims.attr)

if (string.find(styletobe, "Claims") or string.find(styletobe, "Issues")) then
if (string.find(styletobe, "Claims")) then pre=preclaims post=postclaims print("Claim found") end
if (string.find(styletobe, "Issues")) then pre=preissues post=postissues print("Issue found") end

local content = divclaims.content

table.insert(content, 1, pre)
table.insert(content, post)

return content
end
return nil
end

The Latex looks cool, except that the table.insert adds some empty lines :

\begin{issues}

text with formatting....

\end{issues}

That baffles the Latex interpreter . My question is: how can I modify the code such that my latex output will have no extra empty lines?

\begin{issues} text text \end{issues}

Thank you in advance!
Ioan M.

On Monday, January 10, 2022 at 12:46:37 PM UTC-6 BPJ wrote:
It is neither possible nor needed to convert the whole block to HTML within the filter; rather you should just inject the start and end tags:

``````lua
-- Create these only once, for speed and resources saving!
local pre = pandoc.RawBlock('html', '<note>')
local post = pandoc.RawBlock('html', '</note>')

function Div (div)
-- The order of the classes shouldn't matter!
if div.classes:includes('replace-me') then
local content = div.content
table.insert(content, 1, pre)
table.insert(content, post)
return content
end
return nil
end
``````

Den mån 10 jan. 2022 15:33Tomáš Kruliš <tomas....-hR4I4Euo7CWBik42HM7KXg@public.gmane.org> skrev:
Hello,

I would like to ask how you should, in general, detect and manipulate with Pandoc `block` elements. Currently, I am trying to replace `<div class='replace-me'>` tag with `<note>` tag in similar (highly simplified) HTML file:

```.{html}
<html>
<body>
<p> First line. </p>
<div class="replace-me another-class"> This should carry on to converted document. </div>
<p>End.</p>
</body>
</html>
```

I have tried to detect the `<div>` tag, use `walk_block` to get the `<div>` content and put it in `<note>` tag, I also found a code using `:walk` method. Lastly, I tried to convert `<div>` content to simple string and concatenate that in `RawInline` type:

```.{lua}
if elem.t == 'Div' and elem.classes[1] == "replace-me" then
content = pandoc.utils.stringify(elem.content)
return pandoc.RawInline('html', '<note>' .. content.. '</note>')
else
return elem
end
```

But none of that is working. I would like to ask you, how to work in general with `pandoc_walk` or `:walk` (are they the same?) and how to deal with my specific situation?
Thank you very much for any help, I ope that afterwards I will be able to help myself a little bit more :)
Regards Tomas

--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/590abdf0-6bc5-4f37-a978-a46ad5cff5a8n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "pandoc-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pandoc-discuss+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pandoc-discuss/ae032a8d-4d0a-4608-b479-61965cee2793n%40googlegroups.com.