Re: [office] normalizing / cleaning up ODF files for use in revision control

[Jos van den Oever <jos@vandenoever.info>]

Here is a progress report on normalizing the past ODF publications for 
revision control.

The diffs between different versions of the documents are now quite clean. 
Alphabetial marks and toc marks are removed from the documents.  The 
identifiers associated with those markers were quite random and I have not come 
up with a good way to stabilize them. Only ODF 1.1 had a significant amount of 
text:alphabetical-index-mark-start and text:alphabetical-index-mark-end tags. 
For the next errata of v1.1, we should retain the alphabetical marks.

The Relax NG files are now also normalized. This works nicely. A diff between 
the Relax NG for 1.1 and 1.2 is very readable. I've attached the (partially) 
normalized versions so you can look at them in a diff viewer.

The history of the Relax NG for v 1.2 can be explored here:
  https://gitlab.com/odfplugfest/odfspecifcationhistory/commits/master/v1.2/OpenDocument-v1.2-os-schema.rng

The code to do all this work is here:
  https://gitlab.com/odfplugfest/odfhistory/

Now I've approached the point where I think I can start creating commits for 
accepted jira issues.

Cheers,
Jos

Attachment: rngdiff.zip
Description: Zip archive