| 1 |
On Mon, 01 Aug 2005 19:23:37 -0400 Alec Warner <warnera6@×××××××.edu> |
| 2 |
wrote: |
| 3 |
| > Hrmmmmm. Is this going to be sanely doable by your average dev? How |
| 4 |
| > long a dep string would we be having in typical cases? How about in |
| 5 |
| > bad cases? |
| 6 |
| > |
| 7 |
| The more formal the depstring, the quicker the packages build ( |
| 8 |
| using |
| 9 |
| only needed packages instead of lumping them in one group ). This is |
| 10 |
| essentially what the DEPEND should be, just what the packages needs to |
| 11 |
| compile and run. This especially benefits embedded targets who need a |
| 12 |
| bare-bones set of libraries and nothing else. |
| 13 |
|
| 14 |
The problem is... By hard-coding a bunch of xorg packages, you're making |
| 15 |
your DEPEND *less* accurate. Most packages will build just fine with |
| 16 |
other X implementations. |
| 17 |
|
| 18 |
Mmm, but for now this is pretty much a pointless argument. We're not a |
| 19 |
real metadistribution yet :) |
| 20 |
|
| 21 |
| I think concepts are important and abstract complexity from a |
| 22 |
| packages |
| 23 |
| DEPEND. However, having the DEPEND be accurate is important as well. |
| 24 |
| This almost reeks of the USE group GLEP being applied here as well. |
| 25 |
| Setting up DEPEND-set would be great for this, although it is |
| 26 |
| difficult to imagine sets that can be shared between many packages. |
| 27 |
| eclasses are marginally decent for this right now anyway. |
| 28 |
|
| 29 |
GLEP 37 allows that, in effect. |
| 30 |
|
| 31 |
Speaking of GLEP 37... Jason -- I'm assuming that virtual-x11/ (say) |
| 32 |
would be possible? |
| 33 |
|
| 34 |
| The problem with the individual approach is if I wanted to run |
| 35 |
| XFree, |
| 36 |
| or a competing X implementation, I have to add about 200+ packages to |
| 37 |
| /etc/portage/package.provided. This is not clean at all; it's |
| 38 |
| hideous. If I add the meta-build to /etc/portage/package.provided it |
| 39 |
| just means that I am managing the meta-ebuild and it is valid to count |
| 40 |
| it as installed for dep calculation. This is not true of any of the |
| 41 |
| dependencies of the meta-ebuild however. Thats just what I remember |
| 42 |
| fro m discussion about it, so correct me if it's wrong ;) |
| 43 |
|
| 44 |
Providing a specific metapackage is a bad idea. What if a package really |
| 45 |
does depend upon xorg? Providing a specific concept would be better. |
| 46 |
Whether such a thing is implementable currently is up for debate... |
| 47 |
|
| 48 |
-- |
| 49 |
Ciaran McCreesh |
| 50 |
-- |
| 51 |
gentoo-dev@g.o mailing list |
Archives