problem building Xerces-J

[Guillaume Rousse <Gu...@univ-reunion.fr>]

Hello.
I was attempting to package Xerces for Linux Mandrake, but i had the
following error when trying to build documentation of Xerces-J from
sources :
	[guillaume@agathe xerces-1_2_3]$ make docs
	Building Stylebook docs in docs directory
	 -p  docs/html
	java -classpath
"./tools/xalan.jar:./tools/xerces.jar:./src:./samples:./tools/stylebook-1.0-b2.jar:"
org.apache.stylebook.StyleBook 				"targetDirectory=docs/html"
docs/docs-book.xml tools/style-apachexml.jar
	[StyleBook] Overriding targetDirectory="docs/html" (Old=".")
	[StyleBook] Project URL: "sbk:/sources/docs-book.xml"
	[BasicEngine] Initializing
	[Loader] Parsing Project file
	[XalanProcessor] Applying XSL sheet
"sbk:/style/stylesheets/book2project.xsl"
	[Loader] Caught java.util.zip.ZipException
	[StyleBook] Caught org.apache.stylebook.LoadingException
	org.apache.stylebook.LoadingException: Aucun fichier ou répertoire de
ce type
		at org.apache.stylebook.Loader.load(Loader.java:61)
		at org.apache.stylebook.StyleBook.getProject(StyleBook.java:121)
		at org.apache.stylebook.StyleBook.main(StyleBook.java:85)
	[StyleBook] Error creating project
	make: *** [docs] Erreur 3
Apparently, tools/style-apachexml.jar, that contains the needed xsl
file, is not in the classpath, altough it is present as a command line
option. As i don't know anything in XSL, and i don't have the source for
SourceBook, i didn't looked further.

Morevoer, once this target has been selected, even if it failed, it
can't be run again :
	[guillaume@agathe xerces-1_2_3]$ make docs
	make: `docs' est à jour.
And just removing src/classfiles_updated doesn't work either :
	[guillaume@agathe xerces-1_2_3]$ make docs
	make: *** Pas de règle pour fabriquer la cible
`src/classfiles_updated', nécessaire pour `docs'. Arrêt.

BTW, there is also a problem with make clean:
	[guillaume@agathe xerces-1_2_3]$ make clean
	/bin/sh: -c: line 1: syntax error near unexpected token `;d'
	/bin/sh: -c: line 1: `if [ -n "" ];  then for i in ;do  echo "make -C
$i clean"; make -C $i clean;  done;  fi'
	make: *** [cleandirs] Erreur 2

Just changing line 104 of src/Makefile.incl to
	then for i in "$(DIRS)"; do \
solved at least the syntax error, but resulted in another failure i
couldn't solve :
	[guillaume@agathe xerces-1_2_3]$ make clean
	make -C src clean
	make[1]: Entre dans le répertoire `/home/guillaume/xerces-1_2_3/src'
	make -C org javax clean
	make[2]: Entre dans le répertoire
`/home/guillaume/xerces-1_2_3/src/org'
	make[2]: *** Pas de règle pour fabriquer la cible `javax'. Arrêt.
	make[2]: Quitte le répertoire `/home/guillaume/xerces-1_2_3/src/org'
	make[1]: *** [cleandirs] Erreur 2
	make[1]: Quitte le répertoire `/home/guillaume/xerces-1_2_3/src'
	make: *** [clean] Erreur 2

Hope this helps.

-- 
Guillaume Rousse

Murphy's law : If anything can go wrong, it will.
O'Tool's commentary : Murphy was an optimist.

RE: [PATCH] Preserve element identifiers when cloning Document

[Jay Cain <ja...@cett.msstate.edu>]

I've submitted two patches concerning this issue and have gotten no feedback
whatsoever. Is the patch going to be applied or what? Isn't this supposed to
be an open-source project? As far as I know, I've followed the proceedures
for submitting a patch and there have been no replies. What gives?

- - - - -
Jay Cain
Center for Educational and Training Technology
Mississippi State University

[PATCH] Preserve element identifiers when cloning Document

[Jay Cain <ja...@cett.msstate.edu>]

Here is an updated patch for DocumentImpl. The element identifiers are
preserved when a deep clone is performed. Is there a need to be able to turn
this feature off?

Also, a minor change was made for DOM conformity. importNode now throws
NOT_SUPPORTED_ERR instead of HIERARCHY_REQUEST_ERR when an import of a
document node is attempted, which is according the DOM specifications. This
shouldn't cause any problems, especially if the DOM Java bindings were used
as the reference. If it will cause problems, it's only one line of code to
change.

The tests passed and the documentation has been updated.

- - - - -
Jay Cain
Center for Educational and Training Technology
Mississippi State University

[PATCH] Preserve element identifiers when cloning Document

[Jay Cain <ja...@cett.msstate.edu>]

This is a patch for DocumentImpl.java allowing it to preserve element
identifiers when deep cloning. Basically, it checks a hashtable for the
element being cloned and assigns the associated identifier to the new
element.

The descriptions of the modified methods need to be rehashed.

- - - - -
Jay Cain
Center for Educational and Training Technology
Mississippi State University

> -----Original Message-----
> From: Peter Kharchenko [mailto:peterk@interactivebusiness.com]
> Sent: Thursday, January 04, 2001 3:38 PM
> To: xerces-j-dev@xml.apache.org
> Subject: cloning DocumentImpl object
>
>
> Hi there,
>  I can't seem to find a way to make a complete clone of a Document object
> (that would fully replicate the behavior of the original, for example
> identifier table is copied, so one can do getElementById() etc). Note
> that cloning must be deep, that is all of the nodes belonging to that
> document must be copied.
>  I've seen similar questions brought up before, but never saw a good
> answer to that.
>
>  Here's the problem in details:
> There are a few ways one could attempt to clone that DocumentImpl object.
> 	- using overloaded Object.clone() method. That doesn't work, since
> clone() doesn't copy children.
> 	- using cloneNode(true) doesn't work either since
> identifier tables are
> not copied.
>
>  Is there a standard way of creating a fully-functional deep clone of a
> DocumentImpl object ?
>
> thanks in advance,
>
> -peter.
>
> P.S. since what I really need is a deep clone that preserves identifier
> table, I might just rewrite importNode() function to do just that, but
> that feels like a cheapo hack.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
> For additional commands, e-mail: xerces-j-dev-help@xml.apache.org

cloning DocumentImpl object

[Peter Kharchenko <pe...@interactivebusiness.com>]

Hi there,
 I can't seem to find a way to make a complete clone of a Document object 
(that would fully replicate the behavior of the original, for example 
identifier table is copied, so one can do getElementById() etc). Note 
that cloning must be deep, that is all of the nodes belonging to that 
document must be copied. 
 I've seen similar questions brought up before, but never saw a good 
answer to that. 

 Here's the problem in details:
There are a few ways one could attempt to clone that DocumentImpl object.
	- using overloaded Object.clone() method. That doesn't work, since 
clone() doesn't copy children.
	- using cloneNode(true) doesn't work either since identifier tables are 
not copied.

 Is there a standard way of creating a fully-functional deep clone of a 
DocumentImpl object ?
 
thanks in advance,

-peter.

P.S. since what I really need is a deep clone that preserves identifier 
table, I might just rewrite importNode() function to do just that, but 
that feels like a cheapo hack.