Suggested way to make markdown image with crossref label.

[arbipher]

Hi developers,

I was using hybrid mode following some overleaf tutorials, but I can't figure out the correct way to cross-ref them.

\setkeys{Gin}{width=\linewidth}
![dvn1](Paul_Gauguin_-_D'ou_venons-nous.jpg "Where Do We Come From? What Are We? Where Are We Going?")

I tried figuring it out locally and here is the snippet from the lualatex output

{}\markdownRendererImage{dvn1}{asset/Paul_Gauguin_-_D'ou_venons-nous.jpg}{asset/Paul_Gauguin_-_D'ou_venons-nous.jpg}{Where Do We Come From? What Are We? Where Are We Going?}\markdownRendererInterblockSeparator

The manuals show \markdownRendererImage takes four arguments and the 1st is the label, however, I cannot use \ref{dvn1} to correct cross-ref it.

Meanwhile, the heavy-tex way make it correct

\begin{figure}[H]
    \setkeys{Gin}{width=\linewidth}
    ![Where Do We Come From? What Are We? Where Are We Going?](Paul_Gauguin_-_D'ou_venons-nous.jpg)
    \caption{Where Do We Come From? What Are We? Where Are We Going?}
    \label{fig:dvn0}
\end{figure}

p.s. as for the soft-deprecated hybrid, I also didn't figure out what should be the correct option to use my above first snippet.

[Witiko]

Hi and thanks for using the Markdown package!

You may use the option linkAttributes, which would allow you to specify an image identifier as follows:

\setkeys{Gin}{width=\linewidth}
![dvn1](Paul_Gauguin_-_D'ou_venons-nous.jpg "Where Do We Come From? What Are We? Where Are We Going?"){#fig:dvn0}

p.s. as for the soft-deprecated hybrid, I also didn't figure out what should be the correct option to use my above first snippet.

You may use the option rawAttribute in as follows:

`\setkeys{Gin}{width=\linewidth}`{=tex}
![dvn1](Paul_Gauguin_-_D'ou_venons-nous.jpg "Where Do We Come From? What Are We? Where Are We Going?"){#fig:dvn0}

or:

``` {=tex}
\setkeys{Gin}{width=\linewidth}
```
![dvn1](Paul_Gauguin_-_D'ou_venons-nous.jpg "Where Do We Come From? What Are We? Where Are We Going?"){#fig:dvn0}

Compared to the hybrid option, this has the advantage of cleanly separating the Markdown document from LaTeX commands.

However, a cleaner approach is to place the \setkeys command outside the Markdown document in the LaTeX preamble to avoid embedding raw LaTeX altogether.

Let me know if you need further clarification or examples!

[arbipher]

Hi, thanks for your timely and detailed answer.

For my first question, it doesn't solve the label problem. With your fix, now the generated markdown is

\markdownRendererAttributeIdentifier{fig:dvn1}\markdownRendererImage{dvn1}{dvn.jpg}{dvn.jpg}{Where Do We Come From? What Are We? Where Are We Going?}\markdownRendererImageAttributeContextEnd{}\markdownRendererParagraphSeparator

However, I tried fig:dvn1 as many variants as the label in \ref{}, not none of them worked. Is there any way I can see the compiled tex file output for my markdown input.