search

ctargett / refguide-asciidoc-poc

Proof of concept of Solr Ref Guide converted to asciidoc format & using Asciidoctor for publishing
2 stars 4 forks source link

Some images aren't getting converted properly #20

Closed ctargett closed 7 years ago

ctargett commented 7 years ago

Kevin Risden noted in the mailing list that images are missing from: http://home.apache.org/~ctargett/RefGuidePOC/jekyll-full/solr-jdbc-dbvisualizer.html. There appears to be a code block with the references.

risdenk commented 7 years ago

@ctargett would you like me to open issues for each item I find instead of mailing list responses?

ctargett commented 7 years ago

From your most recent mail, I'll make them for the ones that are problems (I'm checking them as I reply to your list) - once I'm done with that, you can open new issues if you find new stuff.

ctargett commented 7 years ago

The reason the image refs are getting interpreted as a code block is because they are all indented one space from the start of the line. This is an alternate way of defining a code block with Asciidoctor (http://asciidoctor.org/docs/user-manual/#literal-text-and-blocks).

So, the conversion process needs to make sure that the image:::<path> strings are at the start of the line.

See the list of images in the "Add Driver Files to Classpath" section of https://raw.githubusercontent.com/ctargett/refguide-asciidoc-poc/master/confluence-export/converted-asciidoc/solr-jdbc-dbvisualizer.adoc for an example. The first one is fine, but the next 4 are indented.

hossman commented 7 years ago

This will be easy to clean up in the existing post processing piece of the conversion code

I'll double check, but i don't think pandoc should ever be using that alternate way of specifying code blocks, so I'll have the code remove any leading spaces it finds from any lines.

hossman commented 7 years ago

FWIW: my "easy to clean up in post processing" comment was naive -- it only took a few moments to realize that if we tried to remove any leading spaces from lines we'd break indenting on actual code formatting blocks

Post processing would have worked if we were only/explicitly looking just for the images: markup, but once i started down that road i discovered all the other various places that paragraphs of text were being indented in this way and i went down the rabbit hole of getting to the bottom of where this leading whitespace was coming from

(running-your-analyzer.adoc being a really good example of a file that's greatly improved by these fixes: https://github.com/ctargett/refguide-asciidoc-poc/commit/3a669a003af0ea93cc62e11c56c1f9e2bf7ce03f?short_path=362f349#diff-362f34970898283f4731c6922b783fb7 )