X-Spam-Check-By: sourceware.org
Message-Id: <1165851287.32453.279885173@webmail.messagingengine.com>
From: "Charles Wilson" <cygwin AT cwilson DOT fastmail DOT fm>
To: cygwin AT cygwin DOT com
Content-Disposition: inline
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="ISO-8859-1"
MIME-Version: 1.0
X-Mailer: MessagingEngine.com Webmail Interface
Subject: pcre, C++
Date: Mon, 11 Dec 2006 10:34:47 -0500
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Id: <cygwin.cygwin.com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

In version 6.0, the (upstream) pcre maintainer added a C++ wrapper
contributed by Google.  However, in pcre-6.7 (and going back all the way
to 6.0), I can find no case where the C++ unittests
pcre_scanner_unittest or pcrecpp_unittest actually pass; they always
segfault on cygwin (gcc/g++ 3.4.4-1, binutils 20060817-1) [and also
mingw].

A little digging shows that the problem occurs in pcrepp.cc (RE::Init):

void RE::Init(const char* pat, const RE_Options* options) {
  pattern_ = pat;  <<< HERE

where std::string pattern_, a member of class RE, is assigned the const
char* pat -- and pat is NOT null (verified in gdb) (it is a valid,
null-terminated regex string).  This is certainly a legal thing to do,
as std::string has a operator=(const char*).  But for some reason, it
fails here.  I think that somehow this code is tickling a g++/stdlibc++
bug, but my gdb-foo is not sufficient to figure it out.  I tried
recompiling with -O0, but (a) the test progs still failed, and (b) there
was no additional enlightenment under gdb.

[Note: pcrecpp works fine on linux[gcc]/solaris[Sun Workshop
8.0]/irix[MIPS-Pro], at least as tested by pcre_scanner_unittest and
pcrecpp_unittest]

Has anyone else observed this failure on cygwin with pcrecpp?  Anybody
got a clue what's really going on here?

--
Chuck




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/