Re: [xml] Error on including and importing schemas already parsed

[Luke Robison <robison arlut utexas edu>]

Thank you Jason!  I can also confirm validate_me.xml validates with this
patch applied, although patching 2.7.3 doesn't work, I had to grab the
svn version (currently r3819).  Thanks for the help.

-Luke

-----------
Luke Robison
robison arlut utexas edu
Applied Research Laboratories: University of Texas
Engineering Scientist Associate

On Wed, 2009-03-04 at 18:19 -0500, Oblivian wrote:
> Here's a patch, seems to work ok but needs review.  validate_me.xml
> now validates...
> 
> On Wed, Mar 4, 2009 at 5:54 PM, Oblivian
> <oblivian users sourceforge net> wrote:
>         Nope, looks like your problem is different.  This appears to
>         be an incomplete or just an unsure implementation.  The code
>         that is causing it is in xmlschemas.c, just a different
>         location:
>         
>                    /*
>                     * Chameleons: the original target namespace will
>                     * differ from the resulting namespace.
>                     */
>                     isChameleon = 1;
>                     if (bucket->parsed &&
>                         (bucket->targetNamespace !=
>         pctxt->targetNamespace)) {
>                         /*
>                         * This is a sanity check, I dunno yet if this
>         can happen.
>                         */
>                         PERROR_INT("xmlSchemaParseIncludeOrRedefine",
>                             "trying to use an already parsed schema
>         for a "
>                             "different targetNamespace");
>                         return(-1);
>                     }
>         
>         Obviously the comment says it all.  Your case shows this can
>         happen and from the XML Schema Part 1: Structures document
>         
>         "The ·XML Schema· corresponding to <schema> contains not only
>         the components corresponding to its definition and declaration
>         [children], but also all the components of all the ·XML
>         Schemas· corresponding to any <include>d schema documents.
>         Such included schema documents must either (a) have the same
>         targetNamespace as the <include>ing schema document, or (b) no
>         targetNamespace at all, in which case the <include>d schema
>         document is converted to the <include>ing schema document's
>         targetNamespace."
>         
>         Your case is the last where no namespace is in the included
>         schema.  So it should allow a generic schema with no
>         targetNamespace to be pulled into the current document's
>         targetNamespace.
>         
>         The problem will be to implement pulling the same file into
>         multiple namespaces which is what it really seems to be
>         complaining about.  I'm not familiar enough with the code yet
>         to know if it is as simple as pulling this check out to fix,
>         and if that may create any wierd redefinition problems as the
>         xml file is parsed against common.xsd in different namespaces
>         at the same time.
>         
>         Hopefully this helps :)
>         
>         Jason
>         
>         
>         
>         On Wed, Mar 4, 2009 at 9:39 AM, Luke Robison
>         <robison arlut utexas edu> wrote:
>                 
>                 
>                 On Tue, 2009-03-03 at 20:53 -0500, Oblivian wrote:
>                 > And I was so busy getting the test cases and patch
>                 done, I didn't
>                 > check today's list threads:
>                 >
>                 >
>                 http://mail.gnome.org/archives/xml/2009-March/msg00007.html
>                 >
>                 > Seems related too...
>                 
>                 
>                 Indeed it does seem related.  I tried applying your
>                 patch to my 2.7.3
>                 code, and while it fixed your test case, it fails to
>                 fix mine.  Glad I
>                 could provide another corner case :-).
>                 
>                 -Luke
>                 
>                 -----------
>                 Luke Robison
>                 robison arlut utexas edu
>                 Applied Research Laboratories: University of Texas
>                 Engineering Scientist Associate
>                 
>         
>         
>