1 |
On Mon, 23 Mar 2009 11:51:28 +0100 |
2 |
Fabian Groffen <grobian@g.o> wrote: |
3 |
|
4 |
> On 23-03-2009 11:41:08 +0100, Sebastian Pipping wrote: |
5 |
> > People split into three groups: |
6 |
> > |
7 |
> > - Friends of ${P}-fix-issue.patch naming |
8 |
> > - Friends of ${PN}-fix-issue.patch naming |
9 |
> > - Friends of ${PN}-1.2.3-fix-issue.patch naming |
10 |
> > |
11 |
> > Qualities |
12 |
> |
13 |
> [snip] |
14 |
> |
15 |
> I think what's missing is the following observation: |
16 |
> |
17 |
> ${PN}-fix-issue.patch naming is bad if you patch code that is (likely) |
18 |
> to change in newer releases. This is almost always the case. |
19 |
> Ultimate example, patch something in ffmpeg or mplayer, and the next |
20 |
> snapshot will break the patch. (i.e. doesn't apply any more.) Using |
21 |
> ${PN}-fix-issue.patch in this case gets you into |
22 |
> ${PN}-fix-issue-2.patch, which IMO is ugly. |
23 |
> |
24 |
> If patches are named this way, they probably fall in the case where |
25 |
> the code it patches is unlikely to change. (assumption) |
26 |
> |
27 |
> > Possible solutions |
28 |
> > |
29 |
> > - *Communicating* your likes to all co-maintainers |
30 |
> > in hope the will respect and remember your agreement |
31 |
> > |
32 |
> > - Add a related local comment (*documenting*) to ebuilds |
33 |
> > and expect other developers to act accordingly on a bump |
34 |
> |
35 |
> probably best solution |
36 |
> |
37 |
> > - Making a GLEP *enforcing* on of these and make people |
38 |
> > vote on which |
39 |
> |
40 |
> very bad one. |
41 |
> |
42 |
> |
43 |
|
44 |
Thanks for this. And apologies to Alin. I was in a very bad mood |
45 |
yesterday. |
46 |
|
47 |
|
48 |
-- |
49 |
gcc-porting, by design, by neglect |
50 |
treecleaner, for a fact or just for effect |
51 |
wxwidgets @ gentoo EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662 |