1 |
Hi all, |
2 |
|
3 |
Inspired by the recent poppler move from autoconf to cmake for its |
4 |
build system, the following. |
5 |
|
6 |
Given that poppler didn't compile on at least two arches, I found that |
7 |
cmake is pretty much terse in its output, especially when errors are |
8 |
encountered. Often it is important to know how the compiler or linker |
9 |
was invoked, to be able to (quickly) resolve the issue at hand. Cmake |
10 |
doesn't seem to report such call by default, it needs VERBOSE=1 to be |
11 |
set in the environment in order to do so. |
12 |
|
13 |
I recently proposed to enable this by default for cmake, but got some |
14 |
negative feedback for that. Hence, I'd like to know the opinion of more |
15 |
people on the issue. |
16 |
|
17 |
In the past we have had verbose build systems, that printed a lot of |
18 |
messages. Portage even analyses this output to look for common |
19 |
problems. Newer buildsystems (like cmake), or just newer insights (like |
20 |
gnustep makefiles, linux kernel, git, ...) suppress more messages |
21 |
leading to reduced output. |
22 |
|
23 |
- should we leave defaults of build systems as is, keeping some very |
24 |
verbose and others very terse? |
25 |
- should we always enable verbosity such that we can analyse logs, both |
26 |
by Portage as well as in bugs when something apparently went wrong? |
27 |
- should make the output level consistent for all build systems? |
28 |
|
29 |
I think verbosity is useful when debugging problems. Portage's --jobs |
30 |
feature nicely allows to hide the "ugly" output (even with --jobs=1), |
31 |
still storing the log for when something goes wrong, while eliminating |
32 |
the need to look at it all the time. |
33 |
|
34 |
So what do you think? Pros, cons? |
35 |
|
36 |
|
37 |
-- |
38 |
Fabian Groffen |
39 |
Gentoo on a different level |