Mail Archives: geda-user/2012/12/21/19:05:35
> > > I was just trying but had a problem. Cloning the repository worked but:
> > >
> > > git checkout for_master origin/for_master
> > >
> > > as you said in a previous message failed!
> >
> > What was the error message?
>
> error: pathspec 'for_master' did not match any file(s) known to git.
> error: pathspec 'origin/for_master' did not match any file(s) known to git.
I think I should have said:
git checkout -b for_master origin/master
(create a branch named "for_master" starting with commit
"origin/master").
> > What branches are listed with
> >
> > git branch
>
> * pcb+gl
Ok, I couldn't remember what my repo.or.cz repository defaults to as
HEAD.
pcb+gl should be ok to use, gives you some board-outline rendering
stuff, a completely replaced GL polygon rasteriser (modified from
cairo), and generally should be a little faster. (But cleanly coded
enough to be pushed upstream yet).
pcb+gl_experimental is faster still, but contains even wackier code.
That branch uses GLSL shaders and vertex buffers etc.., so is nice and
quick - but it doesn't fallback gracefully if your OpenGL driver doesn't
support something it relies upon.
(Almost all cards and drivers do support what it needs).
> I'm using git 1.7.10.4, the one from Debian testing updated about 1 month
> ago. Apparently I could simply type git checkout origin/for_master.
Yes, it probably gets you into either a "detached head" state, where
you're not really on a branch.
OR.. into a place where you are on the "origin/for_master" branch, where
you should not actually make commits... or risk messing up your
repository's tracking of my remote repository.
> 41 files changed, 7430 insertions(+), 522 deletions(-)
>
> from diffstat.
Yes, prototype first, refactor later...
The pcb+gl branch has a clone of almost the entire draw.c rendering code
embedded in src/hid/gtk/gtkhid-gl.c
I'm slowly working on figuring out what changes I required in that
branch (for 3D board-stack view etc.), and trying to refactor the git
HEAD draw.c code such that I can make those extensions.)
> > Nice.. I'll have to make it a little bit more configurable, and figure
> > out whether the "FOUND" flag should be applied to every object, or just
> > the connected ones..., then I can push it.
>
> I believe that it should be applied to every object.
That isn't how it is currently implemented, but I am leaning towards
that view - out of risk of breaking anything people liked to do with
that flag in the past.
I know DJ uses "select found objects" after the "f" shortcut, so I need
to fix that to un-break his use-case.
Regards,
--
Peter Clifton <peter DOT clifton AT clifton-electronics DOT co DOT uk>
Clifton Electronics
- Raw text -