Birgi Tamersoy, 09.12.2008:
>
> Hello Markus,
>
> On Tue, Dec 09, 2008 at 01:41:37AM +0100, Markus Heidelberg wrote:
> >
> > Birgi Tamersoy, 08.12.2008:
> > > Hello Everybody,
> > >
> > > I fixed this problem with a simple patch. I also had to add something to
> > > the Makefile. I decided to separate the Vim source tree patches related
> > > to CodeCheck from the direct code_check.c patches. That is why I am
> > > attaching two files. Please let me know if this is not a good idea, or
> > > there is another common way of doing it. I will fix it accordingly.
> >
> > What's the benefit from splitting up the patch?
> >
> I initially thought having a separate list of patches to code_check.c
> will be cleaner, at least until the project gets mature enough. But, I
> just realized that it will make things much more complex for the users.
I wouldn't split up the patch only to separate the changes in
code_check.c from the rest. But in this case there are two independent
changes, so two patches are appropriate.
> > Yet some comments. After discussion I will include the fixes.
> >
> > > +++ code_check.c 2008-12-08 10:59:10.000000000 -0600
> > > @@ -361,7 +363,9 @@
> > > */
> > > void
> > > cc_update_screen(void) {
> > > +#ifdef FEAT_GUI
> > > if (!gui.in_use) {
> > > +#endif
> > > /* in console mode updating the screen from the worker thread
> > > * does not cause any problems. */
> > > update_topline();
> > > @@ -370,6 +374,7 @@
> > > setcursor();
> > > cursor_on();
> > > out_flush();
> > > +#ifdef FEAT_GUI
> > > } else {
> > > /* updating the screen in gui mode is troublesome. */
> > > char_u bytes[3];
> > > @@ -380,6 +385,7 @@
> > >
> > > add_to_input_buf(bytes, 3);
> > > }
> > > +#endif
> > > }
> > >
> > > /*
> >
> > Why not interchanging the "then" and "else" path?
> >
> I did not see any clear advantage in doing that. Do you think it will be
> more readable?
Yes, I think so. You'd only have one "#ifdef FEAT_GUI", there are
several other places in the code with the same two paths in the other
order. But I can apply it nevertheless.
> > > +++ Makefile 2008-12-08 11:45:56.000000000 -0600
> > > @@ -1363,6 +1363,9 @@
> > >
> > > TAGS_INCL = *.h
> > >
> > > +# CodeCheck requires the pthreads library.
> > > +EXTRA_LIBS = -lpthread
> > > +
> > > BASIC_SRC = \
> > > buffer.c \
> > > charset.c \
> >
> > Is this needed? It already linked and worked without it. And doesn't
> > this break linking on systems without pthread?
> >
> This is actually needed in the ``./configure --disable-gui'' option.
> Previously it was linking because, one of the gui libraries was
> including the pthread library.
>
> However you are right, it will break in systems without pthread.
>
> I will try to come up with a clean solution to this library problem
> and create a new combined patch.
That'd be fine.
Currently the code_check.txt help file also has the *intro* and
*progress* keywords which "overwrite" the default goals to intro.txt and
the progress filetype. Could this be adjusted?
Thanks, Markus
--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---
