Hello Nicholas, On Sun, May 14, 2017 at 10:45:54PM -0400, Nicholas D Steeves wrote: > I thought I'd take a quick look at dh-make-elpa's create_docs, because > I've been using dh-make-elpa a lot recently. What do you think of > iteratively printing (filename\n) from a list generated from *.md and > *.org instead of checking for an *.md and writing "*.md\n"? Then the > file (containing an \n separated list) could be copied/moved to > package1.docs and package2.docs and the items not relevant to > $packageX could simply be cut from its packageX.docs. That's more or > less way I do it from the shell.
Sounds good to me. > Also, if there will be package-name-based heuristics then I'm guessing > any potential-doc-type.files that match a hungry glob for "first" > could go into elpa-first.doc, anything that matches "second" and not > "first" would go into elpa-second.doc, and anything that doesn't match > either can go into something like needs-review.doc. That last one > ought to produce a lintian error and force manual review. I think this would be overly complex and rarely correct. I'd prefer just to dump everything in debian/docs, and let the user split it. -- Sean Whitton
signature.asc
Description: PGP signature
