Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com Message-ID: <023f01c21069$d439bf40$42a18c09@wdg.uk.ibm.com> From: "Max Bowsher" To: "Pavel Tsekov" Cc: References: <000401c2105b$dacd6f40$42a18c09 AT wdg DOT uk DOT ibm DOT com> <14910202790 DOT 20020610120910 AT syntrex DOT com> Subject: Re: Diagnosis of setup crash (was: Re: Re[3]: cygwin download incomplete - try again - loop?) Date: Mon, 10 Jun 2002 11:30:23 +0100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Pavel Tsekov wrote: > MB> The crash occurs because rfc1738_do_escape is passed url=0x0. This zero pointer > MB> is gereated in rfc1738_escape_part by url.cstr_oneuse(), where url = String(""). > > MB> rfc_escape_part("") is called from check_for_cached(). The "" value comes from > pkgsource.sites[n]->>key. The site objects with a key of "" are created somewhere > MB> in the yacc/lex stuff (of which I know nothing) based upon the mirror value > MB> produced in IniParseFindVisitor::visitFile(). Changing < mirror = String (); > > MB> to < mirror = String ("."); > on line 69 of IniParseFindVisitor.cc eliminates > MB> the crash. > > Well, it seems like this is a result of the fix applied for the > previous bug. Btw IMO this can be fixed by skipping the ini_init() > part and all its dependencies when the mirror is determined to be > empty. I.e. the previous fix is OK , but some more code has to be > moved to the block "prefix.size() > 0". Hang on! Won't that mean that setup.ini files in the root dir of the package cache are ignored? -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/