OpenMoko codebases with Linux Cross Reference (LXR)

[Joachim Steiger]

Andy Green wrote:
> | currently i would rather like to put that onto the 'todo when we have a
> | bit more time again' list.
> 
> "Never", then :-)

didn't say that. but we have to prioritize stuff, else we will sink that
boat pretty fast by drilling to many holes without proper plugging em. ;)

> | also i think there are some important questions IF we do that:
> |
> | - which of our many source repos do we want in such a thing?
> 
> There's only one "source repo" for kernel, git.  It only makes sense to
> have stable branch I guess, since that is what most people will have on
> their device.
> 
> | - how and when do they get updated/synced?
> 
> Just do it once a day should be fine.  My google method often gets me
> 2.6.17 and even that is OK for many things, like which include file is
> such and such struct defined in... grep is very noisy for popular struct
> names but lxr understands the definition action and lists it separately.

so we basically have only one kernel 'version' in there which is git
'stable' and delete and regenerate that every night?

> | besides that it seems atleast 'do-able' (even if i really do not like
> | perl code anymore ;))
> 
> I read it was ugly to set up, but CPAN is the man for the perl stuff.

still... perl is and will stay ugly ;) (just my own view).. too much
'write only', not intended for reading by people != the author.

i have it on my wishlist ;)

-- 

Joachim Steiger
Openmoko Central Services