tclxml-users — Discussion on using the TclXML package

Re: [Tclxml-users] How to generate an XML document using TCL

[Peter F. <pet...@zv...>]

Steve Cassidy wrote:
> 
> 
> Human readability isn't really one of XML's properties -- more like
> 'potential human decodeability' in that it's possible to use XML in such a
> way that the meaning of the structure is transparent.

very true.

> Of course it's also
> possible to use it to make data formats just as opaque as MSWord's binary
> format.

sorry i have to disagree. I never seen xml from any source that was
anywhere near obscure as MSWord's proprietary binary format .. 
Having said that, its true that anyone can make XML content more obscure
than it needs to be by ignorance or lack of effort ...

This leads me to some other benefits of XML - in an emergency I can
_edit_ any arbitrary pieces of xml with just about any simple, ascii
capable editor (i.e just about everything) on any platform without
having to spend a cent.  I doubt that you can compare MSW binary fmts
and XML on any level. <PhilosophicalMode> For lots of reasons MSW
document formats will never be ubiquitous, but XML has the chance to be.
I think thats the key to improving our lot - ubiquitous technology
</PhilosophicalMode>. 

> 
> One of my main motivations for using XML as a serialisation format is that it
> removes the need to invent a new format and write a parser for it.  I'm bad
> at inventing new formats and when I've done it I've always found a little
> later that I need to extend it in some way that I didn't forsee. XML makes it
> much more likely that I'll be able to do that.  

a key feature - the promise of a ubiquitous array of standards based,
extensible, portable tools for a wide range of document and data
manipulation tasks. 

And dont get me started on the raft of applications and/or application
instances that have expensive database backends (purchase cost / system
resource or admin labour wise) when all they need is a decent filesystem
and structured content ....

> It also means that I don't
> need to write the parser (of course this argument doesn't hold if you're
> Steve Ball :-)

All hail Steve "lets write our CMS in xslt" Ball - he is a guru  8-). 
I know he will gladly accept offerings at the TCLXML alter to ease his
pain ;-) 

And I'd me amiss not to take this opportunity to remind everyone that as
a Steve and I are very busy these days earning a quid with our company,
so any help in writing parsers and assorted peripheral stuff is most
welcome ...



Peter Farmer            |  Custom XML software   | Internet Engineering 
Zveno Pty Ltd           | Website XML Solutions  | Training & Seminars
http://www.zveno.com/   |   Open Source Tools    |   - XML XSL Tcl
Pet...@zv...  +------------------------+---------------------
Ph. +61 8 92036380      | Mobile +61 417 906 851 | Fax +61 8 92036380

Re: [Tclxml-users] How to generate an XML document using TCL

[Steve C. <ste...@mq...>]

On Fri, 26 Oct 2001 10:33, Philip Ehrens wrote:
> Without defending XML itself, it is trivially easy to convert XML
> to HTML ;^)
>
> That said, I would like to tell a little story.
>
> On our project, we intended to adopt XML as our lightweight
> data format, because it is supposed to be human readable.
>
> Unfortunately, our imagined idea of what XML looks like was
> naive; XML is not very readable!!

Human readability isn't really one of XML's properties -- more like 
'potential human decodeability' in that it's possible to use XML in such a 
way that the meaning of the structure is transparent. Of course it's also 
possible to use it to make data formats just as opaque as MSWord's binary 
format. 

One of my main motivations for using XML as a serialisation format is that it 
removes the need to invent a new format and write a parser for it.  I'm bad 
at inventing new formats and when I've done it I've always found a little 
later that I need to extend it in some way that I didn't forsee. XML makes it 
much more likely that I'll be able to do that.  It also means that I don't 
need to write the parser (of course this argument doesn't hold if you're 
Steve Ball :-)

Steve

Re: [Tclxml-users] How to generate an XML document using TCL

[Cameron L. <cl...@st...>]

	From tcl...@li...  Thu Oct 25 19:45:53 2001
	From: Philip Ehrens <pe...@li...>
			.
			.
			.
	The internal format has nice things like this:

	(line has been wrapped)

	<int_4s compression='9' format='base64' name='my_data'
	        comment='this is my data'> ... </int_4s>

	So, I guess XML is to prevent us from writing non-portable
	but more readable formats.  But I, for one, think maybe
	it's got a bit out of hand.
			.
			.
			.
I, for a second, sure can't figure out some
of the marketing perpetrated on XML's behalf.

Back to technical matters:  Phil, I don't
understand of what the above is an example.
Why can't it be XML?

Re: [Tclxml-users] How to generate an XML document using TCL

[<Ste...@zv...>]

ski wrote:
> 
> > > I need to generate an XML document using a tcl script, how do i
> > > go about generating the same ?
> 
> as i read Sunil's post [and Steve's answer] i wondered out loud:
> "why XML?" Sunil talks of inventory data, can i assume a RDBMS
> holds this information?
> 
> Sunil suggests [reading between the lines] the end result is to
> view this inventory data with a browser. so again, "why XML?"
> 
> i have yet to write one line of XML. i use PHP to extract the
> proper data from any RDBMS and dispaly it in a browser. there
> is no 'middle' step. there is only [almost] real time access
> to the REAL data in the database.
> 
> how does tcl and XML help in this scenario?

I don't think Sunil gives enough information about the application
to answer this question (at least in a meaningful way - we can always
debate the issues regardless ;-)

The typical answer in this scenario is that the application is
using XML as a data exchange language.  XML is not necessarily the
best way to *store* data (an RDBMS may be a better solution, 
my favourite DBMS would be MetaKit) but it is an excellent language
for *transmitting* data.  The benefits of using XML for data exchange
are many, but include platform- and language-independence,
structuring of data, etc, etc.

The fact that Sunil wants to display the data in a browser does
not preclude XML - you can do client-side transformations using XSL.

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

Re: [Tclxml-users] How to generate an XML document using TCL

[Philip E. <pe...@li...>]

Without defending XML itself, it is trivially easy to convert XML
to HTML ;^)

That said, I would like to tell a little story.

On our project, we intended to adopt XML as our lightweight
data format, because it is supposed to be human readable.

Unfortunately, our imagined idea of what XML looks like was
naive; XML is not very readable!!

We wound up adopting a special "internal" lightweight format
that looks like we wished XML looked, and everybody who sees
it can read it almost from the get-go.

We also wrote conversion libraries that convert our internal
format to XML, but it comes out so ugly that nobody ever does
it.

The internal format has nice things like this:

(line has been wrapped)

<int_4s compression='9' format='base64' name='my_data'
        comment='this is my data'> ... </int_4s>

So, I guess XML is to prevent us from writing non-portable
but more readable formats.  But I, for one, think maybe
it's got a bit out of hand.

ski wrote:
> 
> Sunil suggests [reading between the lines] the end result is to
> view this inventory data with a browser. so again, "why XML?"

Re: [Tclxml-users] How to generate an XML document using TCL

[ski <sk...@eD...>]

> > I need to generate an XML document using a tcl script, how do i
> > go about generating the same ?

as i read Sunil's post [and Steve's answer] i wondered out loud:
"why XML?" Sunil talks of inventory data, can i assume a RDBMS
holds this information?

Sunil suggests [reading between the lines] the end result is to
view this inventory data with a browser. so again, "why XML?"

i have yet to write one line of XML. i use PHP to extract the
proper data from any RDBMS and dispaly it in a browser. there
is no 'middle' step. there is only [almost] real time access
to the REAL data in the database. 

how does tcl and XML help in this scenario?

-ski

p.s. i am a huge fan of tcl... and do use it often in my PHP
	solutions. i just do not see the value of XML here.

[Tclxml-users] Re: [tclxml] How to generate an XML document using TCL

[<Ste...@zv...>]

Please use the SourceForge mailing list in future:
tcl...@li...

"Rajapurohit, Sunil" wrote:
> I need to generate an XML document using a tcl script, how do i
> go about generating the same ?

When creating XML, I generally use TclDOM.  Create a DOM tree in memory,
and then use 'dom::DOMImplementation serialize $doc' to generate the
XML.  The TclDOM package will make sure that the generated XML is 
well-formed.

Alternatively, XML is just text so there's no reason why you can't
just create the string directly.  Eg:

	puts "<document>$content</document>"

The problem with this is that (a) you have to worrying about the
XML syntax nitty-gritty and (b) the content variable may contain
special characters which you have to deal with.

There are also some generation packages available, like the 'html' 
package in tcllib (this will be added to TclXML RSN, when my 
workload permits).

Hope that helps,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

[Tclxml-users] Re: dom 1.6, a bug?

[<Ste...@zv...>]

Dear Yong Chen,

You wrote:
> I'm trying your tclxml 2.1theata and tcldom-1.6 (I can't find the
> place to download tcldom2.0, if it exists).

TclDOM 2.0 is in development.  You can get it from the CVS repository.

> Basically these 2 packages are great, but I'm having a problem with
> your dom package, can you please help?

Your document uses XML Namespaces.  Unfortunately, although TclXML v2.x
does support XML Namespaces, TclDOM v1.6 does not support XML Namespaces 
at all.

I've started adding support for XML Namespaces in v2.0, 
but it is by no means complete yet (serialisation is a major issue).
That work is part of the effort to add DOM Level 2 support.

Sorry I can't help further at this time,
Steve Ball

The rest of what you wrote:
> under tclsh8.3 shell:
> 
> %package require dom
> 1.6
> %% proc readAndModifyEnvelope { args } {
>     set f [open "SessionManagement_SessionManager_createSession.xml"]
>     set all_lines [ read $f]
> 
>     set mydoc [dom::DOMImplementation parse $all_lines]
>     set rr [dom::DOMImplementation serialize $mydoc]
>     puts $rr
>     return $rr
> }
> 
> The previous proc simply reads in an XML file
> "SessionManagement_SessionManager_createSession.xml", converts it into
> dom doc, then serializes it back (and saves the result in rr) and
> returns the result.
> 
> The problem is, the value in 'rr' is different now from the file. Some
> attributes of some elements such as Envelope, soapServiceHandler are
> missing after conversion. For example, Envelope had an attribute named
> xmlns:s, it doesn't show up after the conversion.
> 
> Also <s:Body> becomes <Body>.
> 
> Here is the original file:
> SessionManagement_SessionManager_createSession.xml
> 
> <s:Envelope xmlns:s="http://schemas.xmlsoap.org/soap/envelope/"
>             xmlns:xsd="http://www.w3.org/1999/XMLSchema"
>             xmlns:xsi="http://www.w3.org/1999/XMLSchema-instance"
> 
> s:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/">
>   <s:Body>
>     <soapServiceHandler xmlns="urn:MyService">
>       <createEnableSession module="SessionManagement"
> interface="SessionManager">
>         <userName>
>           admin
>         </userName>
>         <password>
>           admin
>         </password>
>         <enablePassword>
>           enable
>         </enablePassword>
>       </createEnableSession>
>     </soapServiceHandler>
>   </s:Body>
> </s:Envelope>
> 
> the file is changed to following (result from puts $rr):
> 
> <?xml version='1.0'?>
> <!DOCTYPE Envelope>
> <Envelope s:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/">
> 
>   <Body>
>     <soapServiceHandler>
>       <createEnableSession interface="SessionManager"
> module="SessionManagement">
>         <userName>
>           admin
>         </userName>
>         <password>
>           admin
>         </password>
>         <enablePassword>
>           enable
>         </enablePassword>
>       </createEnableSession>
>     </soapServiceHandler>
>   </Body>
> </Envelope>
> 
> thanks for your help
> Yong Chen

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

[Tclxml-users] Use of the Xerces C++ compiler

[Kevin S. <Kev...@te...>]

All,

I would like to implement a simple interface to the Xerces C++ parser in TCL
(specifically dealing with the generation of an XML file from a DOM structure).

I am working on HP-UX 11, and used the aCC compiler to build the Xerces C++
parser shared library.

I have built a stub TCL package shared library (with only a package init and a
single simple command). This stub makes no reference to the Xerces code.

I am using tclsh 8.3 compiled using an ansi C compiler on HP-UX 11.

If I link my stub TCL package shared library using aCC (the C++ compiler) and
include the Init and the TCL command implementation function in an extern "C"
block, and DONT link the Xerces shared library in during the link phase, then my
stub shared library loads successfully.

However, if I link the Xerces shared library in during the link phase without
changing the object, then I get the following errors on TCL load:

/usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_Start (data)  from
/usr/lib/libCsup.2
/usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_End (data)  from
/usr/lib/libCsup.2
/usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_Start (data)  from
/usr/lib/libCsup.2
/usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_End (data)  from
/usr/lib/libCsup.2
/usr/lib/dld.sl: Unresolved module for symbol: __shlinit (code)  from
/usr/lib/libstream.2
couldn't load file "../lib/libxmlgenxerces.sl": unresolved symbol name

(where "../lib/libxmlgenxerces.sl" is my stubb TCL package library)

Does anyone have any ideas ?

Kev Shaw



Tertio Limited  -  One Angel Square,   Torrens Street,   London  EC1V 1PL
Tel: +44 (0)207 843 4000 Fax: +44 (0)207 843 4001 Web http://www.tertio.com
Any views expressed in this message are those of the individual sender,
except where the sender specifically states them to be the views of Tertio Ltd.

[Tclxml-users] Re: -errorcommand

[Mats B. <ma...@pr...>]

Ste...@zv... wrote:
> 
> Mats Bengtsson wrote:
> > Steve Ball wrote:
> > > Mats Bengtsson wrote:
> > > >
> > > > The docs for TclXML specify the -errorcommand proc to take two parameters,
> > > > 'tclProc errorcode errormessage', but browsing through the 'sgmlparser.tcl'
> > > > file, the error proc is usually called as:
> > > >
> > > > uplevel #0 $options(-errorcommand) "$attr around line $state(line)"
> > ...
> > > >
> > > > Am I right here?
> > >
> > > Hmmm... yes I'd say so.  The Error procedure in sgml-8.[01].tcl simply
> > > uses 'args'.  Sorry about that!  Would you care to submit a patch? ;-)
> > >
> >
> > Since I don't know the code that well, the risk is that I create more
> > errors than I fix.
> 
> That's what the regression test suite is for.  Make your changes, and
> then
> run 'make test'.
> 

Ok, I'll give it a try. But I need to know your intentions with the
'errorcode errormessage' parameters. The 'errormessage' is obviously
the present text string, but what about 'errorcode'? Is that a numerical
three digit string, or just a brief 'errormessage'?


/Mats

Re: [Tclxml-users] Re: Tclxml-users digest, Vol 1 #57 - 1 msg

[<Ste...@zv...>]

Mats Bengtsson wrote:
> Steve Ball wrote:
> > Mats Bengtsson wrote:
> > >
> > > The docs for TclXML specify the -errorcommand proc to take two parameters,
> > > 'tclProc errorcode errormessage', but browsing through the 'sgmlparser.tcl'
> > > file, the error proc is usually called as:
> > >
> > > uplevel #0 $options(-errorcommand) "$attr around line $state(line)"
> ...
> > >
> > > Am I right here?
> >
> > Hmmm... yes I'd say so.  The Error procedure in sgml-8.[01].tcl simply
> > uses 'args'.  Sorry about that!  Would you care to submit a patch? ;-)
> >
> 
> Since I don't know the code that well, the risk is that I create more
> errors than I fix.

That's what the regression test suite is for.  Make your changes, and
then
run 'make test'.

> One related issue. If I make a crude workaround by having 'args' in
> the errorcommand proc, see below, the elementendcommand fails,
> see stack below. What is the correct procedure to make the best
> error recovery?
> 
> code-----------------------------------
> set data2 {<stream:error>Invalid XML</stream:error>}
> .
> proc HandleErrorFix {args} {
>     global  myParser
> 
>     puts "HandleErrorFix ==> args=$args"
>     $myParser reset
>     $myParser free $myParser
> }
> 
> set myParser [xml::parser]
> $myParser configure -elementstartcommand HandleStart  \
>   -elementendcommand HandleEnd -characterdatacommand HandleText \
>   -final 0 -errorcommand HandleErrorFix
> 
> $myParser parse $data2

You're destroying the parser state while it is in the middle of parsing.
You must return an error code of TCL_ERROR or TCL_BREAK in order to
terminate
the parser.  For example:

proc HandleErrorFix {args} {
    global  myParser

    puts "HandleErrorFix ==> args=$args"
    #$myParser reset
    #$myParser free $myParser
    return -code break
}

The reset and free calls are commented out because I don't believe
that's
a reasonable thing to do at that point.

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

[Tclxml-users] Re: Tclxml-users digest, Vol 1 #57 - 1 msg

[Mats B. <ma...@pr...>]

>
> Mats Bengtsson wrote:
> >
> > The docs for TclXML specify the -errorcommand proc to take two parameters,
> > 'tclProc errorcode errormessage', but browsing through the 'sgmlparser.tcl'
> > file, the error proc is usually called as:
> >
> > uplevel #0 $options(-errorcommand) "$attr around line $state(line)"
...
> >
> > Am I right here?
> 
> Hmmm... yes I'd say so.  The Error procedure in sgml-8.[01].tcl simply
> uses 'args'.  Sorry about that!  Would you care to submit a patch? ;-)
> 

Since I don't know the code that well, the risk is that I create more
errors than I fix. 

One related issue. If I make a crude workaround by having 'args' in
the errorcommand proc, see below, the elementendcommand fails,
see stack below. What is the correct procedure to make the best
error recovery?

/Mats

code-----------------------------------
set data2 {<stream:error>Invalid XML</stream:error>}
.
proc HandleErrorFix {args} {
    global  myParser
    
    puts "HandleErrorFix ==> args=$args"
    $myParser reset
    $myParser free $myParser
}

set myParser [xml::parser]
$myParser configure -elementstartcommand HandleStart  \
  -elementendcommand HandleEnd -characterdatacommand HandleText \
  -final 0 -errorcommand HandleErrorFix

$myParser parse $data2


backtrace***************************************

can't read "namespaces()": no such variable
    while executing
"set nsuri $namespaces([lindex [lsort -dictionary -decreasing 
           [array names namespaces $prefix,*]] 0])"
    (procedure "ParseEvent:ElementClose" line 17)
    invoked from within
"ParseEvent:ElementClose $tag [array get options]"
    invoked from within
"::sgml::parseEvent {stream:error {} {} {Invalid XML} stream:error / {} {}} 
 -emptyelement {namespace inscope ::xml::tclparser ParseEmpty} -parseattribu..."
    ("eval" body line 1)
    invoked from within
"eval ::sgml::parseEvent [list $tokenised] $parseOptions"
    (procedure "parse" line 38)
    invoked from within
"parse xmlparser1 {<stream:error>Invalid XML</stream:error>}"
    (in namespace inscope "::xml::tclparser" script line 1)
    invoked from within
"namespace inscope ::xml::tclparser parse xmlparser1 
            {<stream:error>Invalid XML</stream:error>}"
    ("eval" body line 1)
    invoked from within
"eval $classinfo(-parsecommand) [list $name] $args"
    ("parse" arm line 5)
    invoked from within
"switch -- $method {

	configure {
	    # BUG: We're not checking for legal options
	    array set data $args
	    eval $classinfo(-configurecommand) [..."
    (procedure "::xml::ParserCmd" line 7)
    invoked from within
"::xml::ParserCmd xmlparser1 parse {<stream:error>Invalid XML</stream:error>}"
    ("eval" body line 1)
    invoked from within
"eval ::xml::ParserCmd xmlparser1 [list $method] $args"
    (procedure "xmlparser1" line 1)
    invoked from within
"$myParser parse $data2"
    (file "" line 46)

Re: [Tclxml-users] bugs -errorcommand ?

[<Ste...@zv...>]

Mats Bengtsson wrote:
> 
> The docs for TclXML specify the -errorcommand proc to take two parameters,
> 'tclProc errorcode errormessage', but browsing through the 'sgmlparser.tcl'
> file, the error proc is usually called as:
> 
> uplevel #0 $options(-errorcommand) "$attr around line $state(line)"
> 
> and in some places as:
> 
> uplevel #0 $options(-errorcommand) [list $attr around line $state(line)]
> 
> The uplevel command concatenates all its arguments which breaks the
> list structure above. I guess that a number of [list [list ....] ...]
> commands are needed here to build the proper list structure.
> 
> There are many 'uplevel #0 $options(-errorcommand)' statements to fix.
> 
> Am I right here?

Hmmm... yes I'd say so.  The Error procedure in sgml-8.[01].tcl simply
uses 'args'.  Sorry about that!  Would you care to submit a patch? ;-)

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

[Tclxml-users] bugs -errorcommand ?

[Mats B. <ma...@pr...>]

The docs for TclXML specify the -errorcommand proc to take two parameters,
'tclProc errorcode errormessage', but browsing through the 'sgmlparser.tcl'
file, the error proc is usually called as:

uplevel #0 $options(-errorcommand) "$attr around line $state(line)"

and in some places as:

uplevel #0 $options(-errorcommand) [list $attr around line $state(line)]

The uplevel command concatenates all its arguments which breaks the 
list structure above. I guess that a number of [list [list ....] ...]
commands are needed here to build the proper list structure.

There are many 'uplevel #0 $options(-errorcommand)' statements to fix.

Am I right here?

/Mats

code------

set data2 {<stream:error>Invalid XML</stream:error>}
proc HandleError {errcode errmsg} {
    puts "Error ==> errcode=$errcode, errmsg=$errmsg"
}
...
set myParser [xml::parser]
$myParser configure -elementstartcommand HandleStart  \
  -elementendcommand HandleEnd -characterdatacommand HandleText \
  -final 0 -errorcommand HandleError
$myParser parse $data2
> called "HandleError" with too many arguments

[Tclxml-users] Internet worm warning!!!

[J. L. <jl...@cs...>]

Hi!

There's a new internet worm out there! It's called "nimda".

Our virus scanner detect this worm earlier this morning (MEZ)
and the related mail comes definetly from this mailing-list.

So be carefull opening attachments like readme, readme.wav,
readme.exe if you using Windows boxes.

Cheers

___________________________________________________
Jörg Linnenkohl
e-mail: jl...@cs...
phone : +49 069 50952 225
fax   : +49 069 50952 299
CSK Software AG * Holzhausenstraße 44 * D-60322 Frankfurt
http://www.csksoftware.de

[Tclxml-users] tclxml: xerces wrapper compilation problem

[Sreng T. <sre...@ax...>]

Hi,

I checked-out the "xerces" branch in tclxml module. But I can't compile.
I tried to fix some missing includes as
#include <framework/XMLPScanToken.hpp>
#include <validators/DTD/DTDValidator.hpp>
#include <validators/DTD/ContentSpecNode.hpp>

in tclXercesHandlers.hpp, misspellings as TclXercesParserData against
XercesParserData, but there is still too much.

Is there another version on another tag ?

Thanks,
Sreng.

Re: [Tclxml-users] How do I build tcldom?

[Joe E. <jen...@fl...>]

Larry W. Virden  wrote:

> When I get a copy of the tcldom in the tclxml cvs, I find I don't
> have a complete build environment:
> 
> CVS	     aclocal.m4  configure.in  library	   tclxml.m4
> Makefile.in  changes	 docs	       src	   test
> README	     config	 examples      tcldom.cpp  tests
> 
> Notice there isn't a configure file; I'm not certain what else
> is missing.


For the C implementation, cd to the 'tcldom/src' directory first 
and run 'sh ./configure ; make ; make install' there.  (Let me 
know if this doesn't work.  You may need to install expat first.)

For the Tcl implementation, just copy tcldom/library/*.tcl 
to somewhere that Tcl can find it.  (You may need to tweak
the pkgIndex.tcl file too.)

You *could* try running 'autoconf' followed by
configure; make; make install in the tcldom top-level
directory; it looks like the top-level configure.in
is TEA-based (and you're not Jeff, so it probably won't 
work for you :-)


--Joe English

  jen...@fl...

Re: [Tclxml-users] tclDOM: how to free memory after removing a child ?

[Steve B. <Ste...@zv...>]

Sreng Truong wrote:
> here is a code I tested. After removing a child from the DOM tree, I try
> 
> to free the child from memory.
> 
> ---------- destroy.tcl
> package require dom
> 
> set doc [dom::DOMImplementation create]
> set root [dom::document createElement $doc top]
> set node1 [dom::document createElement $root node]
> set node2 [dom::document createElement $root node]
> 
> puts $node1
> dom::node removeChild $root $node1
> dom::DOMImplementation destroy $node1
> puts [dom::DOMImplementation serialize $doc]
> 
> -------
> That doesn't work at all! 
[...snip...]
> 
> Do I use the destroy function correctly ?

It looks OK to me.  Please submit this as a bug report on SourceForge.

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

Re: [Tclxml-users] How do I build tcldom?

[Mark H. <ma...@us...>]

Larry W. Virden <lv...@ca...> wrote:
> Notice there isn't a configure file; I'm not certain what else
> is missing.
> 
> What steps do I have to take to get that resolved?

Run "autoconf" and the configure file will be generated.
It is not necessary to do this on the same system as you
will be configuring, so if you don't want to d/l and install
autoconf, it might be easier to run it on a handy linux box.

Cheers,
Mark

--
Mark Harrison             ma...@us...  ma...@ma...
Chief Software Architect         www.asiainfo.com    www.markharrison.net
AsiaInfo Holdings, Inc.                               +1-408-970-9356 (o)
Beijing/Santa Clara                                   +1-408-590-8606 (m)

[Tclxml-users] How do I build tcldom?

[Larry W. V. <lv...@ca...>]

When I get a copy of the tcldom in the tclxml cvs, I find I don't
have a complete build environment:

CVS	     aclocal.m4  configure.in  library	   tclxml.m4
Makefile.in  changes	 docs	       src	   test
README	     config	 examples      tcldom.cpp  tests


Notice there isn't a configure file; I'm not certain what else
is missing.

What steps do I have to take to get that resolved?
-- 
Never apply a Star Trek solution to a Babylon 5 problem.
Larry W. Virden <mailto:lv...@ca...> <URL: http://www.purl.org/NET/lvirden/>
Even if explicitly stated to the contrary, nothing in this posting should 
be construed as representing my employer's opinions.
-><-

[Tclxml-users] tclDOM: how to free memory after removing a child ?

[Sreng T. <sre...@ax...>]

Hi,

here is a code I tested. After removing a child from the DOM tree, I try

to free the child from memory.

---------- destroy.tcl
package require dom

set doc [dom::DOMImplementation create]
set root [dom::document createElement $doc top]
set node1 [dom::document createElement $root node]
set node2 [dom::document createElement $root node]

puts $node1
dom::node removeChild $root $node1
dom::DOMImplementation destroy $node1
puts [dom::DOMImplementation serialize $doc]

-------
That doesn't work at all!  The following error message is displayed on
my shell:


::dom::document1(node4)
invalid token "::dom::document1(node1)"
    while executing
"GetHandle documentFragment [lindex $args 0] node"
    ("serialize" arm line 7)
    invoked from within
"switch -- $method {

        hasFeature {

            if {[llength $args] != 2} {
                return -code error "wrong number of arguments"
            }

            # Later on, could u..."
    (procedure "dom::DOMImplementation" line 5)
    invoked from within
"dom::DOMImplementation serialize $doc"
    invoked from within
"puts [dom::DOMImplementation serialize $doc]
"
    (file "destroy.tcl" line 13)




Do I use the destroy function correctly ?


Thanks for your help.
Sreng.

Re: [Tclxml-users] when will the full distribution be available ?

[Steve B. <Ste...@zv...>]

Sreng Truong wrote:
> When will the full distribution be available ? Approximately ?
> Is 1.6 the latest released version of TclDOM ?

Version 2.0 is currently under development.  You can checkout
the latest code from the SourceForge CVS repository.

At the moment I'm inclined to release the Tcl code, similarly
to the TclXML 'theta' release.

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

[Tclxml-users] when will the full distribution be available ?

[Sreng T. <sre...@ax...>]

Hello,

When will the full distribution be available ? Approximately ?
Is 1.6 the latest released version of TclDOM ?

Thanks.

Re: [Tclxml-users] Is anyone aware of a location to obtain a tar of the tcldom package?

[Steve B. <Ste...@zv...>]

Steve Cassidy wrote:
> 
> On Thursday 16 August 2001 08:38, Steve Ball wrote:
> > Do folks want an official v2.0 release sooner, rather than later?
> 
> Well, I can say that the current version is very useful, I've just started
> working with it to translate some legacy data into xml and it's proved it's
> worth already.

Cool!

> I'm using the tcl only version, I notice that there's c code in the CVS
> package which doesn't get built when I say 'make'. Does this provide an
> alternate version?

Scriptics/Ajuba contributed their 'TclDOMPro' package, which was a
C implementation of the TclDOM API.  Joe English has been working
on it.  The aim is to have a version of TclDOM implemented in C
which is 'plug-n-play' compatible with the Tcl version.

However, I've started work on a wrapper for libxml, which includes a
DOM implementation.  The aim of this subproject is to have a TclDOM
package which integrates with the TclXSLT package (announced recently).

Alternatively, there's tDOM.

At this stage we may end up with four DOM packages; 3 C and 1 Tcl,
3 API-compatible and 1 incompatible.  This situation may need to be
rationalised, time will tell.

Cheers,
Steve Ball

-- 
Steve Ball            |   XSLT Standard Library   | Training & Seminars
Zveno Pty Ltd         |     Web Tcl Complete      |   XML XSL Schemas
http://www.zveno.com/ |      TclXML TclDOM        | Tcl, Web Development
Ste...@zv...  +---------------------------+---------------------
Ph. +61 2 6242 4099   |   Mobile (0413) 594 462   | Fax +61 2 6242 4099

Re: [Tclxml-users] Is anyone aware of a location to obtain a tar of the tcldom package?

[Steve C. <ste...@mq...>]

On Thursday 16 August 2001 08:38, Steve Ball wrote:
> Larry W. Virden wrote:
> > I am investigating tclxml, tcldom, and tclsoap and have gotten the first
> > and last of these packages - but I can't seem to locate a place to
> > download a tar file of tcldom.  Anyone know where I can turn?  Thanks!
>
> TclDOM v2.0 has not yet been officially released, so there's no
> tarball.  While I'd like to do more work on the package (more Level 2/3
> support, XPath support, etc), perhaps there's no point in delaying
> release.  At this stage there's some significant release engineering
> to be done.
>
> Do folks want an official v2.0 release sooner, rather than later?


Well, I can say that the current version is very useful, I've just started 
working with it to translate some legacy data into xml and it's proved it's 
worth already.  I'm not pushing it by any means and I'm only using it to 
generate xml, not parse.  

I'm using the tcl only version, I notice that there's c code in the CVS 
package which doesn't get built when I say 'make'. Does this provide an 
alternate version? 

Steve