1 |
On 03/10/2014 1:27 PM, Michał Górny wrote: |
2 |
[snip] |
3 |
> |
4 |
> I'm sorry but I don't get what you are complaining about. |
5 |
> |
6 |
> Bluez -- that is the package that aims to bring bluetooth support -- |
7 |
> requires udev to support most of bluetooth hardware, and properly |
8 |
> depends on it. You are complaining about that because it collides with |
9 |
> your fancy device manager that doesn't have any API that could bring |
10 |
> hardware support to bluez. |
11 |
> |
12 |
> So now, do you request that we should provide bluez *without* bluetooth |
13 |
> support? Wouldn't the correct solution be to, say, make bluetooth |
14 |
> support in your package optional if you don't need it? |
15 |
|
16 |
|
17 |
That's not what I am saying. I proposed a patch to make udev optional in |
18 |
bluez because it apparently allows for this support to be optional: |
19 |
|
20 |
> bluez-5.15$ ./configure --help |
21 |
> `configure' configures bluez 5.15 to adapt to many kinds of systems. |
22 |
> |
23 |
> Usage: ./configure [OPTION]... [VAR=VALUE]... |
24 |
[snip] |
25 |
> Optional Features: |
26 |
> --disable-option-checking ignore unrecognized --enable/--with options |
27 |
> --disable-FEATURE do not include FEATURE (same as --enable-FEATURE=no) |
28 |
> --enable-FEATURE[=ARG] include FEATURE [ARG=yes] |
29 |
[snip] |
30 |
> --disable-udev disable udev device support |
31 |
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |
32 |
|
33 |
I don't know what the side-effects are of disabling udev support are. I |
34 |
simply saw that switch in the configure script's --help output, and so I |
35 |
thought I'd modify my local ebuild copy to depend on the 'udev' USE flag to |
36 |
make udev optional so I could try and get omphalos' configure script to |
37 |
complete successfully. Which it didn't for reasons I've already stated. |
38 |
|
39 |
I have to do further experiments and probably get a hold of omphalos' |
40 |
upstream team to better understand why they require bluetooth support in the |
41 |
current codebase. Maybe it can be made optional -- I don't know just yet. |
42 |
|
43 |
Anyways, you say bluez requires udev, but the bluez configure script |
44 |
suggests otherwise. I don't know (or care, really) the reasons why. |
45 |
That's a question for the bluez developers I guess. Alex's patch looks to |
46 |
be better than my proposed solution, so if there is desire, that's probably |
47 |
what should be used. However, if our bluetooth experts think that allowing |
48 |
udev to be optional does more harm than good, then I guess it needs to |
49 |
remain a harddep. |
50 |
|
51 |
In that case, the better solution probably is either to make bluetooth |
52 |
optional in omphalos or create a libbluetooth package that provides the |
53 |
header/libs that'd satisfy the build condition. We've done it before, but |
54 |
I'll have to look into it at another time. |
55 |
|
56 |
-- |
57 |
Joshua Kinard |
58 |
Gentoo/MIPS |
59 |
kumba@g.o |
60 |
4096R/D25D95E3 2011-03-28 |
61 |
|
62 |
"The past tempts us, the present confuses us, the future frightens us. And |
63 |
our lives slip away, moment by moment, lost in that vast, terrible in-between." |
64 |
|
65 |
--Emperor Turhan, Centauri Republic |