Thangalin via ntg-context
The source comes from Markdown, which is converted to XML, then typeset using ConTeXt. There's no Markdown-specific mechanism to relate images to a particular external figure definition, unfortunately. All images use the same syntax and are treated the same way. The only part that differs is the file name extension (and header within the file).
The below is likely overkill and a good bit of work to setup, but it is also very flexible. Maybe it suits your needs. The Markdown-to-XML conversion can be tuned to include all the relevant information. E.g., you could use pandoc with a custom writer to produce the XML format of your choice. The link below is an example XML writer that can be tuned as needed. https://gist.github.com/tarleb/634b409be0af62ca210cc9e96d41ca8c With that setup, one can modify the `Image` function to produce the output you need. E.g., match on the file name extension use the library function `pandoc.path.split_extension` and then change the element type depending on the extension. This way you'd have all the important information in your XML. More info: https://pandoc.org/custom-writers and https://pandoc.org/lua-filters. Of course, you could also use pandoc to go directly to ConTeXt; that is currently my preferred way of producing PDF from Markdown. Cheers, Albert -- Albert Krewinkel GPG: 8eed e3e2 e8c5 6f18 81fe e836 388d c0b2 1f63 1124