On 25 June 2011 00:51, Nathan Phillip Brink <bi...@gentoo.org> wrote: > On Wed, Jun 22, 2011 at 11:20:40PM -0400, Wyatt Epp wrote: >> I bring this up because there are several packages with the same name >> and different qualification. Obviously, they'll have different tags >> because they're not the same thing, but neither should they share the >> same directory. So the simple solution is to just change the package >> names so we avoid collision and preserve our flat ontology (I've >> forgotten the objection to doing this; please forgive). > > I believe that the objection is that it is better to follow upstreams' > package names as directly as possible. This would look better and be > less confusing than having a package named git and git-core, like I've > seen elsewhere. Having categories would also prevent changing an > ebuild's name from upstream's name only for the sake of giving it a > unique name in Gentoo.
You could possibly abuse the tag feature to solve this issue in interesting ways. Both could have: <tag class="common-name">git</tag> and thus people looking for 'git' would be more likely to find it. You could also perhaps extend this idea to other forms of metadata that users are likely to want to be able to search, perhaps: <tag class="provides-binary">git</tag> But that specific usage would probably be deemed slightly abusive. ( as for instance, you may wish to also list what the binary is usually called, and what gentoo ships it as to prevent collisions ) > > I think that in most cases, when package name collisions happen, the > colliding packages differ enough that they'd conceivably be in > different portage categories, letting them be uniquely identified in > Gentoo. If someone is planning on writing a new program, he likely > knows about already-existing alternatives to this package. The author > of a new sound editing suite would not name his suite `sox' because > the author cannot help but to know that media-sound/sox exists. But > someone writing some new sax thing might play off of `sax' and name it > `sox', though this is hypothetical ;-). But with this, you could store one as media-sound/sox-translator and the other as media-sound/sox-saxophone or something equally unique but arbitrary and tag them both with <tag class="common-name">sox</tag> Also, with this in mind, it may be better to have some types of tags that are only aggregated in an index of sorts, and others which are also perhaps made available as a tree of symlinks. > -- > binki > > Look out for missing or extraneous apostrophes! > -- Kent perl -e "print substr( \"edrgmaM SPA NOcomil.ic\\@tfrken\", \$_ * 3, 3 ) for ( 9,8,0,7,1,6,5,4,3,2 );" http://kent-fredric.fox.geek.nz
