Hi,
I've had a very frustrating time with this. I inadvertently upgraded a
bunch of packages which over-wrote the version of nedit I compiled
myself. The Debian Linux package version dies on cut/paste with SIGSEGV.
Subsequent attempts to rebuild my version were unsuccessful. Some
wouldn't even start, just crashed before CreateWindow completed.
A few things:
1) I removed all trace of Motif and Lesstif from my machine.
2) Fetched lesstif-0.95.2.tar.bz2 built and installed it.
3) Rebuild nedit (as from CVS)
nedit still faults with SIGSEGV.
I solved it with this patch to lesstif-0.95.2 for anyone interested.
=============================
--- lesstif-0.95.2.orig/lib/Xm-2.1/CutPaste.c 2007-09-12 22:05:58.000000000
+0200
+++ lesstif-0.95.2/lib/Xm-2.1/CutPaste.c 2009-09-05 16:13:09.000000000
+0200
@@ -1027,10 +1027,12 @@
nunits = transferlen;
/* XChangeProperty expects a buffer of longs when receiving 32 bits
data, MEUHH */
- if (sizeof(long) != 4)
+ if (sizeof(long) != 4) {
convert_buf = XtMalloc(len * sizeof(long));
- for (i = 0; i < len; i++)
- convert_buf[i] = data[i];
+ if (convert_buf != NULL)
+ for (i = 0; i < len; i++)
+ convert_buf[i] = data[i];
+ }
break;
case 16:
=============================
I've sent a copy to Paul, a Lesstif developer. This may or may not be
related to the Cut/Paste problem discussed on nedit-discuss recently.
It may also be worth noting that Debian Linux lists Lesstif 0.95.0 under
its unstable packages. To further the cause of NEdit, it may be worth
chasing the maintainer to move to a more recent version. I'm not sure
which version that is though. 0.95.2 ?
BTW I'm writing this with SciTE. It works. My productivity was screwed
with the bad nedit version and I am still fumbling around with SciTE.
Its not ideal but 'needs must', until I get NEdit sorted out again.
However, I've used nedit for probably 10 years now and want to continue
using it. I've helped debug issues with it in the past. It is bit-rotting
though. A stable version of Lesstif would help Linux users.
What I must also add is that whenever someone pops up on the list with
suggestions and sound keen on picking up development they eventually go
away. It seems that they are more often than not discouraged than encouraged.
I don't think its conscious but there should be reasons to want them to
stick around, rather than reasons why the things they want to do are
problematic.
Dale.
--
NEdit Develop mailing list - [email protected]
http://www.nedit.org/mailman/listinfo/develop
