Hi, On Tue, May 08, 2007 at 09:08:36AM +0900, Charles Plessy wrote: > Dear all, > > In the Debian-Med project, we started to package a bioinformatic > analysis suite - EMBOSS - which exploits local or remote sequence > databases (read-only). When they are small (a few megaoctets), we are > considering pacakging them. But in the case of much bigger ones > (gigabytes), we will use wrappers. I'm curious: What kind of wrappers are you using?
I'm currently packaging some neuro-imaging tools which have source tarballs of several 100 MBs, but the actual code is only a few kilobytes, the rest is datasets. To me it seems to be a waste of space (in the archive) to upload theses huge binary datablobs. I'm currently testing the possibility to download these datasets from the package postinst script. I adapted the scripts from the msttcorefonts package (in a simpler form). This has the advantage that the datasets are only downloaded if it is really necessary (there are modifications checked by md5sum). The is especially usefull as the datasets tend to be the same across releases. Are you using a different approach? Cheers, Michael -- GPG key: 1024D/3144BE0F Michael Hanke http://apsy.gse.uni-magdeburg.de/hanke ICQ: 48230050 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
