tclxml-users

[Tclxml-users] Question on -final usage in tclxml-2.4

[Ted N. S. A. GA <te...@ag...>]

Hello folks,

I am transitioning a program I had running with tclexpat to use
tclxml.  I had some problems with -final in tclexpat, (had to patch the
code someway or another) and seem to be finding something similar here
also.

I am using the pure tcl version of tclxml version 2.4.  It is installed
in a director "lib" beneath my current directory, as is tcllib1.2

Here is my first test script

====Start====
#!/usr/local/bin/tclsh8.4

lappend auto_path lib

package require xml

proc cdata {data args} {
	puts $data
}

set parser [::xml::parser parseit \
	-characterdatacommand cdata ]

$parser parse "<the>world</the>"
====End====

As expected, it prints

	world

Here is my second test script

====Start====
#!/usr/local/bin/tclsh8.4

lappend auto_path lib

package require xml

proc cdata {data args} {
	puts $data
}

set parser [::xml::parser parseit \
	-characterdatacommand cdata -final 0]

$parser parse "<the>"
$parser parse "world"
$parser parse "</the>"
====End====

It does not print "world" and errors out with a long backtrace:


====Start====
  can't read "opts(-statevariable)": no such variable
      while executing
  "upvar #0 $opts(-statevariable) state"
      (procedure "::sgml::tokenise" line 29)
      invoked from within
  "::sgml::tokenise $xml $tokExpr $substExpr -internaldtdvariable ::xml::tclparser::parseit(internaldtd) -statevariable ::xml::tclparser::parseit -final ..."
      ("eval" body line 1)
      invoked from within
  "eval {::sgml::tokenise $xml $tokExpr $substExpr} $tokenOptions"
      (procedure "parse" line 57)
      invoked from within
  "parse parseit <the>"
      (in namespace inscope "::xml::tclparser" script line 1)
      invoked from within
  "::namespace inscope ::xml::tclparser parse parseit <the>"
      ("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 parseit parse <the>"
      ("eval" body line 1)
      invoked from within
  "eval ::xml::ParserCmd parseit [list $method] $args"
      (procedure "parseit" line 1)
      invoked from within
  "$parser parse "<the>""
      (file "./doit2" line 14)
====End====

This seems like a pretty basic thing.  Any suggestions?

Please reply directly to me as well as to the list because sourceforge
seems to be unhappy today, and I have not been able to subscribe so
far.

			Thanks,

			Ted Nolan

Re: [Tclxml-users] Question on -final usage in tclxml-2.4

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

Dear Ted,

You wrote:
> I am transitioning a program I had running with tclexpat to use
> tclxml.  I had some problems with -final in tclexpat, (had to patch the
> code someway or another) and seem to be finding something similar here
> also.
> 
> Here is my first test script
[...snip...]

> As expected, it prints
> 
> 	world

Good - that's a start ;-)

> Here is my second test script
> 
> ====Start====
> #!/usr/local/bin/tclsh8.4
> 
> lappend auto_path lib
> 
> package require xml
> 
> proc cdata {data args} {
> 	puts $data
> }
> 
> set parser [::xml::parser parseit \
> 	-characterdatacommand cdata -final 0]
> 
> $parser parse "<the>"
> $parser parse "world"
> $parser parse "</the>"
> ====End====
> 
> It does not print "world" and errors out with a long backtrace:

Your script should include:

$parser configure -final 1

at the end, just to wind-up the parser.  As it is, I would expect
it to at least not produce an error, if not to emit "world".

> This seems like a pretty basic thing.  Any suggestions?

There is already an open bug on "-final" in the SF bug
tracker (#413341).  Please feel free to append your comments
to that bug.  The bug is currently assigned to Mats, and may be
worthwhile reminding him (gently) to fix it ;-)

Of course, if you can provide a patch I would be only too
happy to incorporate it myself and close the bug report if
it is fixed.

> Please reply directly to me as well as to the list because sourceforge
> seems to be unhappy today, and I have not been able to subscribe so
> far.

Fixed.  You are now subscribed.  (New subscribers must be
approved, to avoid spammers having access to the list).

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