On Wed, Sep 02, 2009 at 02:38:39PM +0200, Jim Meyering wrote:
> On request,

  Yup, thanks Jim !

> I ran clang against libxml2.  The result is a nice html summary,
> but the total weighs in at 220MB, uncompressed.  Here's the browsable
> result:
[...]
> Note that some of these are definitely false positives.
> 
> But here's a legitimate bug in xmlslave.c:
> 
>   
> http://meyering.net/code/tmp/clang/libxml2-vs-clang-syntax-checker/report-5Qxdd7.html#EndPath
> 
>     if (doc != NULL)
>         oldenc = doc->encoding;                   // one-line "then" clause
>       if (ctxt->encoding != NULL) {             // not part of "if block"
>           doc->encoding = BAD_CAST ctxt->encoding;
>       } else if (doc->encoding != NULL) {
>           encoding = doc->encoding;
>       }
>     }

  I went though all of them (173) there is an awful lot of redondant
reports, a reasonable amount of false positive, or rather systematic
checks that may not turn out realistic in real execution (like
ctxt->sax2 changing during a function call), but I think this raised
half a dozen real bugs, and allowed me to add safety check in places
where they shouldn't be really needed, but still good as a safety
feature.
  All changes now commited to git !

  The reports are pretty good except for the amount of duplication and
the size of the HTML generated :-)

 so thanks again for running the test, I think you can now safely clean
that tree up !

Daniel

-- 
Daniel Veillard      | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
dan...@veillard.com  | Rpmfind RPM search engine http://rpmfind.net/
http://veillard.com/ | virtualization library  http://libvirt.org/
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
xml@gnome.org
http://mail.gnome.org/mailman/listinfo/xml

Reply via email to