Trying to compile nedit-latest-sources-HEAD

Eddy De Greef degreef at imec.be
Mon Oct 16 09:38:26 CEST 2006


Scott Tringali wrote:
> Eddy De Greef wrote:
> 
>> Could you apply the attached debugging patch, compile it, run it, and 
>> send us the output that NEdit writes to stdout? You may need to 
>> redirect stdout to a file (nedit > debug.txt) if your server crashes 
>> immediately.
> 
> I didn't see a patch on this mail.  

Sorry, I forgot to attach it; 2nd try.

> (If nothing else, we can just jam in 
> XLIB_SKIP_ARGB_VISUALS into the environment.)

That's an option, but I still would like to know why the workaround 
doesn't work.

Eddy

-------------- next part --------------
Index: util/misc.c
===================================================================
RCS file: /cvsroot/nedit/nedit/util/misc.c,v
retrieving revision 1.82
diff -c -r1.82 misc.c
*** util/misc.c	17 Mar 2006 10:05:39 -0000	1.82
--- util/misc.c	13 Oct 2006 13:25:59 -0000
***************
*** 484,489 ****
--- 484,491 ----
      bestClass = 0;
      bestVisual = 0;
      for (i=0; i < nVis; i++) {
+         printf("Id: 0x%02x depth: %d vendor: %s\n", visList[i].visualid,
+                visList[i].depth, ServerVendor(display));
          if (visList[i].depth >= 32 && 
              strstr(ServerVendor(display), "X.Org") != 0) {
               /* Xorg 6.8.* 32-bit visuals (with alpha-channel) cause a lot
***************
*** 491,496 ****
--- 493,499 ----
                  Users can achieve the same effect with older versions of
                  NEdit by setting the environment variable 
                  XLIB_SKIP_ARGB_VISUALS. */
+             printf("Skipping\n");
              continue;
          }
  	if (visList[i].depth > maxDepth) {
***************
*** 511,516 ****
--- 514,522 ----
  	    }
  	}
      }
+     
+     printf("Using visual #%d: 0x%02x\n", bestVisual, 
+       visList[bestVisual].visualid);
      *visual = cachedVisual = visList[bestVisual].visual;
      *depth = cachedDepth = visList[bestVisual].depth;
      


More information about the Discuss mailing list