rn bug #26
Larry Wall
lwall at sdcrdcf.UUCP
Sat Dec 1 11:20:07 AEST 1984
System: rn version 4.1
Bug #: 26
Priority: MEDIUM (HIGH for pyramids)
Subject: bitmap to rcline conversion may fail
Description:
If you have more than a buffer's worth of unread article numbers upon
exiting from a newsgroup, restore_ng() will misfire. If you are
on a Pyramid, for some reason, it blows up all the time.
Repeat-By:
Better not to repeat it.
Fix: From rn, say "| patch -d DIR", where DIR is your rn source directory.
Outside of rn, say "cd DIR; patch <thisarticle". If you don't have
the patch program, apply the following by hand, or get patch.
Index: bits.c
Prereq: 4.1.1.2
***************
*** 1,4
! /* $Header: bits.c,v 4.1.1.2 84/10/31 11:58:50 lwall Exp $
*
* $Log: bits.c,v $
* Revision 4.1.1.2 84/10/31 11:58:50 lwall
--- 1,4 -----
! /* $Header: bits.c,v 4.1.1.3 84/11/30 17:08:15 lwall Exp $
*
* $Log: bits.c,v $
* Revision 4.1.1.3 84/11/30 17:08:15 lwall
***************
*** 1,6
/* $Header: bits.c,v 4.1.1.2 84/10/31 11:58:50 lwall Exp $
*
* $Log: bits.c,v $
* Revision 4.1.1.2 84/10/31 11:58:50 lwall
* initctl() wrongly assumed that expired articles were unread when trying
* to adjust firstart, and misadjusted toread[ng]. toread[ng] is now
--- 1,9 -----
/* $Header: bits.c,v 4.1.1.3 84/11/30 17:08:15 lwall Exp $
*
* $Log: bits.c,v $
+ * Revision 4.1.1.3 84/11/30 17:08:15 lwall
+ * Had an = where it should have been ==.
+ *
* Revision 4.1.1.2 84/10/31 11:58:50 lwall
* initctl() wrongly assumed that expired articles were unread when trying
* to adjust firstart, and misadjusted toread[ng]. toread[ng] is now
***************
*** 154,160
}
#endif
free(rcline[ng]); /* return old rc line */
! if (mybuf = buf) {
rcline[ng] = safemalloc((MEM_SIZE)(s-buf)+1);
/* grab a new rc line */
strcpy(rcline[ng], buf); /* and load it */
--- 157,163 -----
}
#endif
free(rcline[ng]); /* return old rc line */
! if (mybuf == buf) {
rcline[ng] = safemalloc((MEM_SIZE)(s-buf)+1);
/* grab a new rc line */
strcpy(rcline[ng], buf); /* and load it */
More information about the Comp.sources.bugs
mailing list