Uttered James Laska <jlaska@xxxxxxxxxx>, spake thus:
> Are you already set on this format? Something about leaving the image
> filenames open to the authors appeals to me. What are your thoughts on
> a directory structure instead of filename rules? Perhaps something
> like:
> figs/$lang/
> figs/common/ or just figs/
> If we force all images to be in the figs/ subdirectory, this will make
> it difficult for large documents to remain sane.
> Thoughts/concerns?
Yeah, I'm going to be pretty stubborn on the filename convention.
This format mimics the decision we have for the various document
translations themselves. We have a base filename ${DOCBASE}, a
${LANG} specifier, and an .xml extention. However, with the images
we also allow generic images to omit the ${LANG} component.
I don't really care how the "figs/" files are organized. Make a flat
directory, make tons of subdirs, it all works for me. The
"docs-common/bin/copy-figs" script traverses the "figs/" tree,
copying individual files into corresponding subdirs in the
destination.
I have added the "figs/Manifest-${LANG}" file, mentioned in my
previous email, so you can name your files using whatever convention
you choose; the cost is having to explicity identify the files.
Sound OK, now?
Attachment:
pgpxvmyKWqI86.pgp
Description: PGP signature
-- fedora-docs-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-docs-list
