public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
* Re: libstdc++/4533: mistaken assumption in find_first_not_of and find_last_not_of
@ 2001-10-20 19:03 rodrigc
  0 siblings, 0 replies; 2+ messages in thread
From: rodrigc @ 2001-10-20 19:03 UTC (permalink / raw)
  To: brendan, gcc-bugs, gcc-prs, nobody

Synopsis: mistaken assumption in find_first_not_of and find_last_not_of

State-Changed-From-To: open->closed
State-Changed-By: rodrigc
State-Changed-When: Sat Oct 20 19:03:20 2001
State-Changed-Why:
    Patch applied.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=4533&database=gcc


^ permalink raw reply	[flat|nested] 2+ messages in thread

* libstdc++/4533: mistaken assumption in find_first_not_of and find_last_not_of
@ 2001-10-11  5:26 brendan
  0 siblings, 0 replies; 2+ messages in thread
From: brendan @ 2001-10-11  5:26 UTC (permalink / raw)
  To: gcc-gnats

>Number:         4533
>Category:       libstdc++
>Synopsis:       mistaken assumption in find_first_not_of and find_last_not_of
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Oct 11 05:26:24 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Brendan Kehoe
>Release:        CVS tree
>Organization:
>Environment:

>Description:
In bits/basic_string.tcc, the function find_first_not_of includes a check in its for loop making sure the size specified in the third argument is non-zero.  The standard doesn't actually say it's invalid to have a zero-width size.  Thus we should be getting __pos back, not npos.

Similarly, the same goes with find_last_not_of.
>How-To-Repeat:

>Fix:
Take out `__n &&' from find_first_not_of.
Take out `&& __n' from find_last_not_of.
>Release-Note:
>Audit-Trail:
>Unformatted:
----gnatsweb-attachment----
Content-Type: text/plain; name="diffs-basic_string-1.txt"
Content-Disposition: inline; filename="diffs-basic_string-1.txt"

2001-10-10  Brendan Kehoe  <brendan@zen.org>

	* bits/basic_string.tcc (find_first_not_of): Take out check for
	__n being non-zero, since the standard does not mandate that.
	e.g., a search for "" in "" should yield position 0, not npos.
	(find_last_not_of): Likewise.

Index: include/bits/basic_string.tcc
===================================================================
RCS file: /cvs/gcc/egcs/libstdc++-v3/include/bits/basic_string.tcc,v
retrieving revision 1.6
diff -u -p -r1.6 basic_string.tcc
--- basic_string.tcc	2001/07/20 00:09:31	1.6
+++ basic_string.tcc	2001/10/10 11:44:12
@@ -698,7 +698,7 @@ namespace std
     find_first_not_of(const _CharT* __s, size_type __pos, size_type __n) const
     {
       size_t __xpos = __pos;
-      for (; __n && __xpos < this->size(); ++__xpos)
+      for (; __xpos < this->size(); ++__xpos)
 	if (!traits_type::find(__s, __n, _M_data()[__xpos]))
 	  return __xpos;
       return npos;
@@ -722,7 +722,7 @@ namespace std
     find_last_not_of(const _CharT* __s, size_type __pos, size_type __n) const
     {
       size_type __size = this->size();
-      if (__size && __n)
+      if (__size)
 	{ 
 	  if (--__size > __pos) 
 	    __size = __pos;


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2001-10-20 19:03 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-10-20 19:03 libstdc++/4533: mistaken assumption in find_first_not_of and find_last_not_of rodrigc
  -- strict thread matches above, loose matches on Subject: below --
2001-10-11  5:26 brendan

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).