| 1 |
On Mon, 08 Sep 2008 22:40:37 +0100 |
| 2 |
Steve Long <slong@××××××××××××××××××.uk> wrote: |
| 3 |
> >> * should be treated as being very quickly installable |
| 4 |
> >> * should be treated as having zero cost for installs |
| 5 |
> >> |
| 6 |
> Both of which follow from "installs nothing." Or would you disagree? |
| 7 |
|
| 8 |
No, they're separate properties, with different implications. |
| 9 |
|
| 10 |
Consider, for example, a split baselayout-style package. There could be |
| 11 |
a skeleton-filesystem-layout package that does all its work in pkg_* |
| 12 |
functions (to avoid permission and empty directory problems that come |
| 13 |
from installing directories via the normal methods). It would install |
| 14 |
nothing, but should not be considered for either zero-cost property. |
| 15 |
|
| 16 |
Or, for the reverse: a package that merely installs a simple control |
| 17 |
file that enables functionality in another package may well be best |
| 18 |
considered as zero-cost for package selection. If a package depends |
| 19 |
upon || ( big-scary/processing-package simple-little/plugin-for-foo ) |
| 20 |
and you already have foo but not plugin-for-foo installed, the right |
| 21 |
thing for the resolver to do would be to suggest plugin-for-foo. |
| 22 |
|
| 23 |
As for the quickly installable property, plugin-for-foo might not |
| 24 |
possess it -- for example, vim plugins generally do a vim tag |
| 25 |
regeneration upon pkg_postinst, so they're not 'quick' to install even |
| 26 |
if all they do is provide one text file. |
| 27 |
|
| 28 |
-- |
| 29 |
Ciaran McCreesh |
Archives