Bugzilla Bug 51992

Hello,

this is readseq, a small utility that converts between many file formats for nucleotide/amino acid sequences. It has a dirty compilation process (made by a biologist I guess :-) ), which I left unchanged as it does work. I had to change the emake into a simple make though, since it doesn't cope well with -j2 or other.

I suggest it to be added to the app-sci category.

Regards,
koen

Created an attachment (id=31988) [details]
readseq-1.ebuild

Created an attachment (id=32554) [details]
readseq-1.ebuild - Revised ebuild

- Removed "inherit flag-o-matic gnuconfig" (no inherited function was used).
- Changed KEYWORDS (x86 -> ~x86)
- Corrected license (GPL-2 -> public-domain. I don't see GPL-2 mentionned
anywhere on the home page; correct me if I am wrong.)
- Removed unnecessary src_unpack().
- Compile with user's CFLAGS.
- Install documentation.
- Made SLOTable since we'll probably want to add the more recent (readseq-2)
implementation, programmed in Java, and which offers a Web interface.
(readseq-1 is symlinked to readseq. Users can define their prefered readseq
implementation, if both are installed, using an alias. The last installed
readseq implementation is the default.)

Created an attachment (id=32555) [details]
metadata.xml

- Assigns to sci herd.
- Provides long description.

Created an attachment (id=33140) [details]
readseq-1.ebuild - Revised ebuild

- Dropped the SLOT in favor of two different packages. Although both readseq
implementations offer a command-line interface with the same options, they vary
greatly for the rest and do not benefit from SLOTing.
- Dropped the "default readseq implementation" since it could broke many
existing readseq scripts.

Commited.