| 1 |
On Tuesday 03 December 2013 11:52:47 Robin H. Johnson wrote: |
| 2 |
> On Tue, Dec 03, 2013 at 11:05:51AM -0500, Mike Frysinger wrote: |
| 3 |
> > as for the patch, i'm of the opinion that make.conf is not for |
| 4 |
> > documenting random USE_EXPAND-ed variables. |
| 5 |
> |
| 6 |
> ... |
| 7 |
> |
| 8 |
> > there is the matter of visibility ... we could add a generic pointer to |
| 9 |
> > the make.conf man page discussing that there are many more variables |
| 10 |
> > that might impact the build and to look at the man page for the |
| 11 |
> > eclasses. |
| 12 |
> |
| 13 |
> Can I make a possible additional suggestion? |
| 14 |
> |
| 15 |
> We do already have profiles/${VAR}.desc, describing what the settings |
| 16 |
> themselves do. Can we augment those files to describe the variable |
| 17 |
> itself? |
| 18 |
> |
| 19 |
> Then put a generic note in make.conf about reading those (or something |
| 20 |
> to build docs from them like the eclass docs). |
| 21 |
|
| 22 |
we have profiles/desc/python_targets.desc too. how about we update the eclass- |
| 23 |
manpages to parse those files and produce man pages ? it'd use the existing |
| 24 |
doc format that has been adopted for eclasses. |
| 25 |
|
| 26 |
then the make.conf man page would point you to a main landing page (like |
| 27 |
use.desc), and that would bounce you to all the other generated man pages. |
| 28 |
-mike |
Archives