1 |
Am Mittwoch, den 18.02.2009, 03:22 -0500 schrieb Michael Sterrett: |
2 |
> Patches welcome. |
3 |
> |
4 |
> On Wed, Feb 18, 2009 at 3:13 AM, Torsten Veller <ml-en@××××××.net> wrote: |
5 |
> > * Michael Sterrett <mr_bones_@g.o>: |
6 |
> >> I added a prepalldocs function to eutils.eclass to provide the |
7 |
> >> functionality. It implements the |
8 |
> >> behavior of the current stable sys-apps/portage-2.1.6.4. |
9 |
> > |
10 |
> > ecompressdir is more portage internal than prepalldocs ever was. |
11 |
> > This must be fixed. |
12 |
> > |
13 |
> > |
14 |
Instead of breaking things you could have tried to come up with a real |
15 |
solution. |
16 |
|
17 |
Even though I share the opinion that compressing docs is unnecessary I |
18 |
started looking through the use cases. |
19 |
I'm not done yet, but I think I already identified a couple of common |
20 |
cases: |
21 |
|
22 |
[1] |
23 |
src_install() { |
24 |
emake DESTDIR="${D}" docdir="/usr/share/doc/${PF}" |
25 |
prepalldocs |
26 |
} |
27 |
(or similar) |
28 |
will change to: |
29 |
src_install() { |
30 |
emake DESTDIR="${D}" docdir="${T}/doc" |
31 |
dodoc "${T}/doc" |
32 |
} |
33 |
|
34 |
[2] |
35 |
Not needed at all since every doc gets installed using dodoc or newdoc |
36 |
|
37 |
[3] |
38 |
Package installs doc to wrong location. |
39 |
Solution: dodoc <OLDLOCATION>/* ; rm -rf <OLDLOCATION> |
40 |
|
41 |
[4] |
42 |
Assuming that "make install" installs something in /usr/share/doc/${PF}. |
43 |
Assumption is wrong since "make install" will install |
44 |
to /usr/share/doc/${P} if at all (since the package doesn't know |
45 |
anything about -rX). So it should be changed anyway and then case [1] or |
46 |
[3] applies. |
47 |
|
48 |
Example packages for case [1]: app-arch/gtk-splitter, app-cdr/cdcover, |
49 |
app-crypt/steghide, app-editors/lpe, app-emacs/ess |
50 |
Example packages for case [2]: app-arch/rar, app-backup/amanda |
51 |
Example packages for case [3]: app-crypt/gnupg |
52 |
Example packages for case [4]: app-arch/xdms, app-crypt/gnupg-pkcs11-scd |
53 |
|
54 |
The only problem I see here is that either <OLDLOCATION> or ${T}/doc |
55 |
contains subdirs. So my proposal for the next EAPI is to allow dodoc and |
56 |
newdoc to operate on dirs. Which also gives the benefit to reduce this |
57 |
idiom: |
58 |
insinto /usr/share/doc/${PF} |
59 |
doins -r examples |
60 |
to: |
61 |
dodoc examples |
62 |
|
63 |
|
64 |
Your comments? |
65 |
|
66 |
Cheers, |
67 |
Tiziano |
68 |
|
69 |
|
70 |
|
71 |
-- |
72 |
------------------------------------------------------- |
73 |
Tiziano Müller |
74 |
Gentoo Linux Developer, Council Member |
75 |
Areas of responsibility: |
76 |
Samba, PostgreSQL, CPP, Python, sysadmin |
77 |
E-Mail : dev-zero@g.o |
78 |
GnuPG FP : F327 283A E769 2E36 18D5 4DE2 1B05 6A63 AE9C 1E30 |