Re: [E-devel] patch for edje

[ZigsMcKenzie <zig...@gm...>]

Try #2, thanks for the comments.

Fixed the ugly indentation.  Vim tabstop confilcts.

On Mon, 12 Jun 2006 18:42:03 +0200
Tilman Sauerbeck <ti...@co...> wrote:

> > +      case EDJE_STATE_PARAM_COLOR_CLASS:
> > +         CHKPARAM(3);
> > +
> > +	 char *cc;
> 
> Declaring variables in the middle of a block isn't allowed in C89, and
> IMO we shouldn't break C89 support for no good reason
> (same problem in other places, too).
> 

I used the variable declaration in this way because it was used all over the place in the GET/SET macros that were allready there.  I just wanted the code to look similar to the rest of it.  I've changed all the char declarations to one initial declaration, but left the macros alone.

> >  
> > +   ed->dirty=1;
> > +   _edje_recalc(ed);
> > +
> >     return 0;
> >  }
> 
> OY! Always recalcing the Edje here seems too excessive. It used to work
> well without it for the ops we had before your patch AFAIK, so you
> should prolly only recalc for the new ops, and where it's really
> necessary.

I knew I was going to get a comment on this one.  For a reason I haven't tracked down yet.  The ops stopped doing a recalc on a mouse_event triggered .edc program.  They don't even work on a CVS build (i.e. not mine).  They did work (most of them) when I started working on it.  Halfway through due to some commit somewhere, this stopped working (I've looked through the edje cvs logs for the last month... I don't think it's in edje). I don't know if this is just me or not.  It'd be a great help if you could try this on your machine and let me know if it works.

However, this isn't why I put the recalc after every op.  I figured that if someone runs an edje script, they want whatever they've changed to appear right away.  And, since the actual recalc of the part doesn't happen until after the script in the .edc is finished I thought it'd be ok.  Do a gdb w/breakpoints on _edje_embryo_fn_set_state_val and _edje_part_recalc to see what I'm talking about.  I did notice that if I took out the _edje_recalc the effect was the same, there's plenty of calls to _edje_recalc after the script has run waiting for the object to thaw.  But the ed->dirty=1 needs to stay.

> 
> Regards,
> Tilman
> 
> -- 
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing on usenet and in e-mail?
> 

- Zigs