Re: [Article] Migrating Unstructured Adobe FrameMaker Content to DITA #Oxygen


Chris Papademetrious
 

Hi Chris,

Your recollection is correct! The feature is named "FrameMaker conversion tables", as it is a table of rules applied to infer structure from unstructured content.

FrameMaker > Structured authoring > Converted unstructured documents

We have successfully used this feature to migrate content to DITA (unstructured FrameMaker .fm -> structured FrameMaker .fm -> DITA). It's a bit tedious to get the rules defined correctly (we required some iteration), but it works well after that if everyone followed similar styling conventions.

For the structured .fm to DITA conversion step, we used custom perl/Make scripts that handled keyreffing, topic chunking, graphics conversion, equation conversion to MathML, and various other things I can't remember now. We used the following tool to create specialized DITA elements that replicated our FrameMaker elements, so the writers felt more comfortable during the transition:

DITA-plugin-utilities > make_dita_grammar.pl

The perl script was complicated and filled with spaghetti rules to handle various special cases. At that time, I was not familiar with the Batch Documents Converter plugin. If I were to start over, I would see if I could make the Batch Documents Converter work. Maybe there is some way I could have passed attribute values into the HTML generated by FrameMaker to control the element inference in the Batch Documents Converter, also perhaps with some XSLT refactoring on the back end to handle final cleanup.

 - Chris

Join main@dita-users.groups.io to automatically receive all group messages.