Brian Nelson <pyro@debian.org> writes: > Jan Alonzo <jmalonzo@unpluggable.com> writes: > >> Frank Küster <frank@debian.org> writes: >> >> Hi, >> >> (on somewhat related topic) >> >> I recently filed an RFP on itagalog and aspell-tl and one of the developers >> said that I should consider merging both sources since they use the same word >> list. >> >> <snip> >> >>>> way. I'd like to receive advices about the matter. >>> >>> I didn't check how big the individual upstream data files are, but yes, >>> if they are rather small, you can consider creating a merged >>> "orig".tar.gz file from all of them. >> >> Would this be ok in my case? Get all the aspell files, put in >> ispell-tl, and build the binary packages for aspell and ispell from >> there? But that will make my orig.tar.gz lose its pristineness? > > What aspell files are you putting in, precisely? The dictionary > packages available from aspell.net are contain wordlists that are > pre-compiled to an extent. You don't want to include those; rather, you > should generate those from the common wordlists (the plain, uncompiled, > newline-delimited ones). tl.dat (word list) and the 'info' file. > > If you have an upstream tarball containing the plain wordlists, you > should be able to keep that pristine and add the rest of the stuff to > the diff.gz. If you don't, then there's probably no reason to attempt > to combine them. That's the problem. The aspell.net (and ispell) tarball _is_ the 'upstream'. I already talked to the upstream maintainer of the word list and we're planning to setup a single source for the word lists (just waiting for alioth's approval). So I guess, I'll stick with separate sources for the meantime. Thanks again. Cheers, > -- > For every sprinkle I find, I shall kill you! -- Jan Alonzo <jmalonzo@unpluggable.com>
Attachment:
pgpDGq2AGALmN.pgp
Description: PGP signature