[Mono-bugs] [Bug 36704][Nor] Changed - Assorted System.Xml.XmlSchema fixes

[bugzilla-daemon@rocky.ximian.com]

Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.

Changed by ginga@kit.hi-ho.ne.jp.

http://bugzilla.ximian.com/show_bug.cgi?id=36704

--- shadow/36704	Mon Feb 17 18:39:54 2003
+++ shadow/36704.tmp.2231	Tue Feb 18 13:15:06 2003
@@ -1,12 +1,12 @@
 Bug#: 36704
 Product: Mono/Class Libraries
 Version: unspecified
 OS: All
 OS Details: 
-Status: NEW   
+Status: NEEDINFO   
 Resolution: 
 Severity: Unknown
 Priority: Normal
 Component: System.XML
 AssignedTo: mono-bugs@ximian.com                            
 ReportedBy: roam@ringlet.net               
@@ -73,6 +73,63 @@
 Created an attachment (id=3132)
 Read multiple XmlNode's in an <appinfo> element; expect the correct closing tag on <appinfo> and <documentation> elements.
 
 
 ------- Additional Comments From gonzalo@ximian.com  2003-02-17 18:39 -------
 I'll ask Atsushi to take a look at this.
+
+------- Additional Comments From ginga@kit.hi-ho.ne.jp  2003-02-18 13:15 -------
+Hi Peter, thanks and sorry to be so late.
+
+In fact I'm not schema guy and maybe incompletely understanding.
+Anyway...
+
+> xmlschema-enumerator.patch:
+
+Ok.
+
+> xmlschema-iscompiled.patch:
+> - add actual handling of the XmlSchema.isCompiled member variable: 
+reset it
+>   in all routines that change the schema's textual description and 
+set it
+>   to true upon a successful run of the XmlSchema.Compile() method.
+
+Hmm... isn't it hard work to check *all* the members (including all  
+subcomponents)? It will make schema classes so complicated.
+
+One easier solution I thought is (1)to keep schema string using Write
+() method, and (2)when going to Compile(), create current schema 
+string as same way, compare with kept schema string, and if they are 
+different, then continue Compile() process.
+
+> xmlschema-ann.patch
+> - XmlSchemaAppInfo.cs: add the ability to read more than one 
+XmlNode;
+
+Doesn't your code handle all descendants as if they were children?
+
+> - XmlSchemaAnnotation.cs: note that XmlSchemaAppInfo.Read() and
+>   XmlSchemaDocumentation.Read() return with the reader positioned at
+>   the respective element's end, *not* at the next element or at
+>   the annotation end.  Thus, expect and swallow an EndElement for
+>   'appinfo' or 'documentation', instead of choking while expecting
+>   an EndElement for 'annotation'.
+
+Sure, you are right.
+
+> xmldocument-skipread.patch
+> - XmlDocument.ReadNode: check if the reader is already positioned at
+>   an Element, and if so, do not do a Read() that will skip the start
+>   tag of the element.
+
+If your patch is applied, ReadNode() will arise infinite loop. BTW,
+what XmlReader did you used at that time? I assume that XmlNodeReader
+had such bug that didn't handle starting node correctly, and maybe
+fixed in current cvs.
+
+I'll be grateful if you have some tests for those schema problems.
+
+Anyway, I've fixed XmlSchemaObjectEnumerator and XmlSchemaAnnotation.
+Thanks again for those checks and patches!
+
+ Atsushi Eno