Skip to content

Latest commit

 

History

History
75 lines (55 loc) · 2.36 KB

uplift_prep.md

File metadata and controls

75 lines (55 loc) · 2.36 KB
author publisher
Larry Kollar
Oje Media

Preparing a Book for Uplift {#upliftprep .task}

Converting an MDITA-based LwDITA project to full DITA requires some preparation. Much, but not all, of the preparation can be automated. Be ready to override mistakes when necessary.

  1. Do a global search and replace in the bookmap:

    • Search: format="mdita"
    • Replace: format="markdown"
  2. Move the ID in MDITA files to the topic title.

    The awk script below is an example of how this can be done automatically. Wrap it in a shell script to loop over all the Markdown files in the book.

    NR == 1 && /^---/ { print; heading = 1; next; } # test for a heading
    
    /^ID:/ && heading {
      id = $2; next; # save the ID and eliminate it from the heading
    }
    
    /^---/ { heading = 0; print; next; } # in case a content line starts with "ID:"
    
    /^# / && id { # level 1 heading gets the ID, skip if no ID was found in metadata
      print $0, "{#" id "}"; next;
    }
    
    { print } # everything else gets passed through
  3. Categorize each topic as one of:

    • concept (information about an object or process)
    • task (how to do something)
    • reference (information referred to regularly)
    • topic (introductory content that doesn't fit in other categories)

    The primary difference between concept and reference is that you might only look at a concept once, but return often to reference topics.

    Categorizing topics can be at least partially automated. See Tips for Automating Information Typing.

  4. Assign each topic a topic type on the topic title line:

    • {.concept}
    • {.task}
    • {.reference}

    The topic ID shares the brackets with the topic type, so your topic title might look like this:

    # Doing the Thing {#dothing .task}
  5. Use the DITA-OT to convert the topics:

    dita --input=book.ditamap --output=out/dita --format=dita --args.rellinks=nofamily

    The converted files can be found in the subdirectory out/dita. Note that the converted topics still have an .md extension (more precisely, DITA-OT preserves the extension you used in the original files).

  6. Inspect the DITA files to make sure the conversion worked. If needed, rename extensions (and update references accordingly).