Sandy,
Actually the section you cite (3.3.3) clearly and repeatedly shows that
this is not the case and that
the change is in fact a regression. You are not supposed to touch the
non-0x20 whitespace chars
if they were literally specified with character references.
Regards,
Glenn
Sandy
Gao/Toronto/IB To: [EMAIL PROTECTED]
M@IBMCA cc: [EMAIL PROTECTED]
Subject: Re: [Bug 2761] New: - Incorrect
NMTOKENS attribute value
07/24/2001 normalization in Xerces v1.4.2
09:10 AM
Please respond
to
xerces-j-dev
Hi. By looking at "http://www.w3.org/TR/2000/REC-xml-20001006#AVNormalize";,
I believe the behavior of 1.4.2 is what's intended. For NMTOKEN, the parser
is required to replace #xD, #xA, and #X9 with #x20, then remove leading and
trailing #x20s, and replace sequences of #x20s by single #x20s.
Sandy Gao
Software Developer, IBM Canada
(1-416) 448-3255
[EMAIL PROTECTED]
bugzilla@apach
e.org To:
[EMAIL PROTECTED], [EMAIL PROTECTED]
cc:
07/24/2001 Subject: [Bug 2761] New: -
Incorrect NMTOKENS attribute value
04:22 AM normalization in Xerces v1.4.2
Please respond
to
xerces-j-dev
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2761
*** shadow/2761 Tue Jul 24 01:22:23 2001
--- shadow/2761.tmp.15785 Tue Jul 24 01:22:23 2001
***************
*** 0 ****
--- 1,33 ----
+
+============================================================================+
+ | Incorrect NMTOKENS attribute value normalization in Xerces v1.4.2
|
+
+----------------------------------------------------------------------------+
+ | Bug #: 2761 Product: Xerces-J
|
+ | Status: NEW Version: CVS extract
|
+ | Resolution: Platform: All
|
+ | Severity: Major OS/Version: All
|
+ | Priority: Other Component: Core
|
+
+----------------------------------------------------------------------------+
+ | Assigned To: [EMAIL PROTECTED]
|
+ | Reported By: [EMAIL PROTECTED]
|
+ | CC list: Cc:
|
+
+----------------------------------------------------------------------------+
+ | URL:
|
+
+============================================================================+
+ | DESCRIPTION
|
+ The attribute value (NMTOKENS) is incorrectly handed to the application
because
+ it is incorrectly normalized... (Only   can be stripped away)...
+
+ <!DOCTYPE doc [
+ <!ATTLIST normNames attr NMTOKENS #IMPLIED>
+ ]>
+ <doc>
+ <normNames attr=' A   
	 B '/>
+ </doc>
+
+ Output of the sample program is:
+
+ <!-- in Version Xerces 1.4.1 -->
+ attr="A 
	 B"
+
+ <!-- in Version Xerces 1.4.2 -->
+ attr="A B"
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
