1 |
On Monday, 19. September 2011 10:58:52 Allan Gottlieb wrote: |
2 |
> On Mon, Sep 19 2011, Michael Schreckenbauer wrote: |
3 |
> > On Monday, 19. September 2011 10:20:25 Allan Gottlieb wrote: |
4 |
> >> On Mon, Sep 19 2011, Alan McKinnon wrote: |
5 |
> >> >> > revdep-rebuild checks everything, revdep-rebuild --library |
6 |
> >> >> > checks just some things. |
7 |
> >> >> > |
8 |
> >> >> > ebuilds sometimes issue messages to check just the libraries |
9 |
> >> >> > known |
10 |
> >> >> > to have been updated, but a full revdep-rebuild after an |
11 |
> >> >> > update |
12 |
> >> >> > will catch those anyway. |
13 |
> >> >> |
14 |
> >> >> Until recently I skipped the "--library" step exactly because I |
15 |
> >> >> knew |
16 |
> >> >> revdep-rebuild will find and fix the broken packages after I |
17 |
> >> >> delete |
18 |
> >> >> the old library. So, why bother with the --library step, right? |
19 |
> >> >> |
20 |
> >> >> However. A few weeks ago I got caught when I deleted one of |
21 |
> >> >> those |
22 |
> >> >> obsolete libraries and only then did I find out that gcc is one |
23 |
> >> >> of |
24 |
> >> >> the packages that depend on it :( |
25 |
> >> >> |
26 |
> >> >> I don't skip the --library step any more. |
27 |
> >> > |
28 |
> >> > That's odd behaviour, I wonder what caused the difference. |
29 |
> >> > |
30 |
> >> > Surely revdep-rebuild itself can't do this different just because |
31 |
> >> > you |
32 |
> >> > specified a library to compare? I wonder if that lib was maybe in |
33 |
> >> > the |
34 |
> >> > revdep-rebuild exclude list. |
35 |
> >> > |
36 |
> >> > I'd be interested to track it down for reference, do you remember |
37 |
> >> > the |
38 |
> >> > library involved? |
39 |
> >> |
40 |
> >> It occurs exactly in the case we are discussing libpng |
41 |
> >> |
42 |
> >> ajglap gottlieb # revdep-rebuild; revdep-rebuild --library |
43 |
> >> '/usr/lib64/libpng14.so.14' * Configuring search environment for |
44 |
> >> revdep-rebuild |
45 |
> >> |
46 |
> >> * Checking reverse dependencies |
47 |
> >> * Packages containing binaries and libraries broken by a package |
48 |
> >> update |
49 |
> >> * will be emerged. |
50 |
> >> |
51 |
> >> ... |
52 |
> >> |
53 |
> >> * Checking reverse dependencies |
54 |
> >> * Packages containing binaries and libraries using |
55 |
> >> |
56 |
> >> /usr/lib64/libpng14.so.14 * will be emerged. |
57 |
> > |
58 |
> > First one emerges *broken* packages. |
59 |
> > Second one emerge packages *using* png14 (not necessarily broken) |
60 |
> |
61 |
> OK. But the claim was that: if |
62 |
> revdep-rebuild |
63 |
> with no argument found nothing to build, then |
64 |
> revdep-rebuild --library <some-library> |
65 |
> will find nothing. |
66 |
|
67 |
This is not true. |
68 |
revdep-rebuild without --library argument checks for packages with *broken* |
69 |
linking, eg a library changed it's name and the original one was removed by an |
70 |
update. |
71 |
revdep-rebuild --library checks for packages using that library. |
72 |
|
73 |
If the first one returns nothing, that means, your linking is all ok. |
74 |
If the second one returns nothing, it tells you, that nothing is using that |
75 |
library at all. |
76 |
|
77 |
> This guarantee is apparently no long true as my example in another msg |
78 |
> illustrated. |
79 |
|
80 |
I doubt it was true in the past. |
81 |
|
82 |
Best, |
83 |
Michael |