T. J. Frazier wrote:
Assuming that you are converting one chapter at a time, if we can figure
out an algorithmic method for naming/numbering the pictures, I should be
able to write some Basic that compares the ODT with the wiki-text, and:
1) saves the pic (always as PNG?) with generated name
2) replaces the [img] filter trash with a real image call (pretty good
assumption, that I can extract the params from the existing pic).
The names might look something like 0302WG3_1, or _F1, or such; captions
should be salvageable, as well. In general, you would have to supply the
code with the leading string (before the underscore), once per chapter.
If I can get the synchronization (between ODF and TXT) working properly,
I might even take a crack at translating the Math output, at least crudely.
If you think this would be useful, I'll start on it, but it is a
non-trivial effort, so DHYB. BTW: please mention any "regular" clean-up
you do; I might be able to do it with code.
Re the pix: not all figures in the user guides have numbers and captions. Some
are icons interspersed in the text or in table cells. Others are figs that at
the time I thought did not need captions. Don't know if that would throw off a
numbering scheme or have some other (unwanted) effect.
Yes, I think it would be useful.
--Jean
