xmlxslt-devel

Re: [eugene@snailgem.org: [XML::XSLT] XML::XSLT homepage]

[Mike808 <mi...@ne...>]

Pavel Nejedly wrote:
>     I mean I've removed the font, not the changes :))

I don't think it is the stylesheet. Netscape is rendering the <H#>
tags in the Comic Sans font.

Is what you're hosting what you checked out from the repository?

I'd very much like to see Egon's example content restored, since I 
think that was a major part of what was missing (as well as prior
releases). The most glaring problem was that there is no download
available from the site. It probably ought to point them to the
Sourceforge download area rather than retain the older content.

I don't have the website repository up, but I can validate and 
"DeCSS" the pages so as to be viewable by any browser. But, since
I've yet to get the maintenance fork up to snuff, don't count on
me.

Also, there's some webalyzer statistics that can be viewed, but I
can't seem to get cron working on the new host. Haven't had time
to look into it yet. And I think either Marc or Bron worked
out a trick with the web pages being in the CVS repository and
being extracted automatically.

Mike808/
perl -e "print pack\"H*\",\"6d696b653830384075736572732e736f75726365666f
7267652e6e65740a687474703a2f2f77616c64656e2e6d6f2e6e65742f7e6d696b653830
382f726573756d652e68746d0a452d427573696e657373205370656369616c697374\""

Re: [eugene@snailgem.org: [XML::XSLT] XML::XSLT homepage]

[Pavel N. <bi...@at...>]

On Sun, Sep 24, 2000 at 07:11:10PM +0200, Pavel Nejedly wrote:
#   Wholesale replacement? I've made only minor changes (except, of 
#   course, for the font you dislike :) But I've removed it.

    I mean I've removed the font, not the changes :))

# 
#   Pavel
# _______________________________________________
# Xmlxslt-devel mailing list
# Xml...@li...
# http://lists.sourceforge.net/mailman/listinfo/xmlxslt-devel

Re: [eugene@snailgem.org: [XML::XSLT] XML::XSLT homepage]

[Pavel N. <bi...@at...>]

On Sun, Sep 24, 2000 at 09:43:29AM -0500, Mike808 wrote:
# What was wrong with the site, specifically? I found most of the
# broken links were due to things being located in other parts of the
# original site before we moved it to SourceForge. All we should need
# to do would be to either add the content to make the local link work
# (and maybe make it relative too, instead of absolute), or use the
# absolute URL to Egon's original host.
# 
# None of these things require wholesale replacement of the site, but
# more fixing 'bugs' in the website.

  Wholesale replacement? I've made only minor changes (except, of 
  course, for the font you dislike :) But I've removed it.

  Pavel

Re: [eugene@snailgem.org: [XML::XSLT] XML::XSLT homepage]

[Mike808 <mi...@ne...>]

What was wrong with the site, specifically? I found most of the
broken links were due to things being located in other parts of the
original site before we moved it to SourceForge. All we should need
to do would be to either add the content to make the local link work
(and maybe make it relative too, instead of absolute), or use the
absolute URL to Egon's original host.

None of these things require wholesale replacement of the site, but
more fixing 'bugs' in the website.

Pavel Nejedly wrote:
> ----- Forwarded message from Eugene Sotirescu <eu...@sn...> -----
> # Question to those in the know: what is the homepage of XML::XSLT?
> # I only found http://xmlxslt.sourceforge.net, on which all the local
> # links are broken.
> 
> I've created new website with almost the same content (added new releases to
> the changelog, some text removed and, of course, fixed the links). Its URI is:

My personal taste is not for the comic sans font that Netscape renders
for this page with your stylesheet.

I'd take the approach of correcting specific items in the existing
pages of the website. Don't fix what ain't broken.

Mike808/
perl -e "print pack\"H*\",\"6d696b653830384075736572732e736f75726365666f
7267652e6e65740a687474703a2f2f77616c64656e2e6d6f2e6e65742f7e6d696b653830
382f726573756d652e68746d0a452d427573696e657373205370656369616c697374\""

[eugene@snailgem.org: [XML::XSLT] XML::XSLT homepage]

[Pavel N. <bi...@at...>]

----- Forwarded message from Eugene Sotirescu <eu...@sn...> -----
# Question to those in the know: what is the homepage of XML::XSLT?
# I only found http://xmlxslt.sourceforge.net, on which all the local
# links are broken.


I've created new website with almost the same content (added new releases to
the changelog, some text removed and, of course, fixed the links). Its URI is:

  http://atrey.karlin.mff.cuni.cz/~bim/temp/

Could you look at it and send me your comments? I'd like to commit it to CVS on Tuesday


  Pavel

Re: Xmlxslt-devel digest, Vol 1 #19 - 1 msg

[Geert J. <Gee...@Da...>]

Thanks for reporting this bug...

Greets,
Geert

xml...@li... wrote:
> 
> Send Xmlxslt-devel mailing list submissions to
>         xml...@li...
> 
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://lists.sourceforge.net/mailman/listinfo/xmlxslt-devel
> or, via email, send a message with subject or body 'help' to
>         xml...@li...
> 
> You can reach the person managing the list at
>         xml...@li...
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Xmlxslt-devel digest..."
> 
> Today's Topics:
> 
>   1. XML Tag Names (Marcel Schwarz)
> 
> --__--__--
> 
> Message: 1
> Date: Thu, 14 Sep 2000 15:04:32 +0200 (CEST)
> Reply-To: m.s...@mi...
> Organization: mikan media design
> From: Marcel Schwarz <ms...@mi...>
> To: xml...@li...
> Subject: XML Tag Names
> 
> Hi,
> 
> I have a question concerning XML Tag Names.
> 
> I use the XML::XSLT package to Transform a NITF document (News Industry Text
> Format, www.nitf.org) to HTML.
> The NITF DTD describes the elements <body.head> and <body.content>
> 
> The XSLT-Parser doesn't find these elements, unless I rename them to
> <bodyhead> or <bodycontent>.
> But the XML-Namespace provides "." in element names.
> 
> I operated the same documents with the XML::Sablotron package:
> The Sablotron Parser accepted <body.head> etc.
> 
> Greetings, Marcel.
> 
> ---
> --------------------------------------------------
> die mikan homepage: http://www.mikan.de
> 
> m.s...@mi...
> 
> PGP-Fingerprint
> 1CDB E6DC 02CC 0347 C77C  D8E9 709B 468A 3FAB D3CA
> --------------------------------------------------
> 
> --__--__--
> 
> _______________________________________________
> Xmlxslt-devel mailing list
> Xml...@li...
> http://lists.sourceforge.net/mailman/listinfo/xmlxslt-devel
> 
> End of Xmlxslt-devel Digest_______________________________________________
> Xmlxslt-devel mailing list
> Xml...@li...
> http://lists.sourceforge.net/mailman/listinfo/xmlxslt-devel

XML Tag Names

[Marcel S. <ms...@mi...>]

Hi,

I have a question concerning XML Tag Names.

I use the XML::XSLT package to Transform a NITF document (News Industry Text
Format, www.nitf.org) to HTML.
The NITF DTD describes the elements <body.head> and <body.content>

The XSLT-Parser doesn't find these elements, unless I rename them to
<bodyhead> or <bodycontent>. 
But the XML-Namespace provides "." in element names.

I operated the same documents with the XML::Sablotron package: 
The Sablotron Parser accepted <body.head> etc.

Greetings, Marcel.



---
--------------------------------------------------
die mikan homepage: http://www.mikan.de
 
m.s...@mi...
 
PGP-Fingerprint
1CDB E6DC 02CC 0347 C77C  D8E9 709B 468A 3FAB D3CA
--------------------------------------------------

Perl XML::XSLT Bug in apply-template?

[Erik M. <eri...@st...>]

It seems like apply-templates only make a search for the first template
match that contain a part of the name of the tag. (I.e. it searches for
=~ /$tag/ when it should search for eq $tag)

Or have I read the specification of XSL wrong. In my opinion
<xsl:template match="LINK"> should only match those that are exactly
LINK not H_LINK.


I have the following in my XSL-file (and a lot more...):
------XSL-file------
<xsl:template match="LINK">
  <p><a href="{@URL}"><xsl:value-of select="."/></a></p>
</xsl:template>

<xsl:template match="H_LINK">
  <a href="{@URL}"><xsl:value-of select="."/></a> - 
</xsl:template>

<xsl:template match="HISTORY">
  <p class="navig">
  <xsl:apply-templates select="H_LINK"/>  
  </p>
</xsl:template>

-----XML-file-----
  <HISTORY>
    <H_LINK URL="http://www.liu.se">LiU</H_LINK>
  </HISTORY>

-------------------
It gives 
<p><a href=....</a></p>
when it should have given:
<a href=...</a> -

XML::XSLT says $VERSION = '0.24';

Thankful for an answer and/or bugfix or pointer to where I should search
for it in the code.

/Erik Melkerson

-----------------------------------
Erik Melkersson
eri...@st...
http://www.edu.isy.liu.se/~erime844
ICQ: 3821584

Re: failure notice

[<ma...@ev...>]

Mike,  I'm having mail trouble so I didn't see this till today:


    M> Would it be OK if they were to publish the development
    M> lastest-before-deadline version? Or the not-yet-ready
    M> maintenance version? or should we hint that they will want to
    M> use version ZZZ?

I think the main branch would work.

    M> Just my thoughts. If the consensus is to send a non-legally
    M> binding e-mail that can't be authenticated, then I'll go
    M> along. It doesn't bother me enough to worry too much about it.

I agree that the agreement is silly.

-- 
Hackers view advertising as damage and route around it.
			 -- gowen on slashdot.org

Re: zero bug, again

[<ma...@ev...>]

>>>>> "PN" == Pavel Nejedly <bi...@at...> writes:

    PN> I've fixed that in the MAIN branch and added a test script for
    PN> it (t/zeroes.t)

Excellent!  Glad to have you aboard!

-- 
Hackers view advertising as damage and route around it.
			 -- gowen on slashdot.org

zero bug, again

[Pavel N. <bi...@at...>]

Hi all,

the ``zero bug'' (I mean disappearing of text nodes containing single '0')
was not fixed for xsl:text (_text):

  $Self->[INDENT] -= $Self->[INDENT_INCR];

  if($text) {
   $Self->_move_node ($Self->[XML_DOCUMENT]->createTextNode ($text), $current_result_node); 
  }else{

I suggest to change if($text) to if($text ne '')
Also note that the mechanism used for xsl:value-of is also not good (it works,
but it may create empty text nodes if __string__ returns ''):


    my $text = undef;
    $text = $Self->__string__ ($$xml_node[0]) if @$xml_node;
    $Self->[INDENT] -= $Self->[INDENT_INCR];

    if (defined $text) {
      $Self->_add_node ($Self->[XML_DOCUMENT]->createTextNode($text), $current_result_node);
    } else {


It's suggest to write

   
    my $text = '';
    $text = $Self->__string__ ($$xml_node[0]) if @$xml_node;
    $Self->[INDENT] -= $Self->[INDENT_INCR];

    if ($text ne '') {
      $Self->_add_node ($Self->[XML_DOCUMENT]->createTextNode($text), $current_result_node);
    } else {


I've fixed that in the MAIN branch and added a test script for it (t/zeroes.t)

Mark : I've also fixed __add_default_templates there.

  Pavel

Translating ; ?

[Bruce L. (c) <bl...@sa...>]

Hi,

How do you use translate() to change the ; character to any other character?

Bruce

Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

translate() headache

[Bruce L. (c) <bl...@sa...>]

Hi All,

Can anyone explain why, when I use 

<xsl:variable name="filteredparagraph" select="translate($paragraphstring,
'&#8216;' , ' ' )"/>

on a text node that looks like:

<body_text_paragraph>can obtain  a &#8216;safety net&#8217; letter from the
</body_text_paragraph>

Xalan translates the output thus:

	can obtain  a 

It truncates the text at the first occurance of the unicode string? Why?

Thanks,

Bruce


Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Hairy translate() Problem

[Bruce L. (c) <bl...@sa...>]

Hi All,

I am trying to use the following code involving the translate() function to
replace any 'macintosh quote marks' with Interwoven friendly quote marks. I
use the unicodes to identify the Mac characters.

.
.
.
<xsl:variable name="paragraphstring" select="string(text())"/>
<xsl:variable name="filteredparagraph" select="translate($paragraphstring,
'&#8216;&#8217;&#8220;&#8221;', '``')"/>
.
.
.

PROBLEM: Excelon stylus runs this XSLT with no problems. However, Xalan
truncates the text in the PCDATA element to which the translate() function
is being applied when it comes across one of the control characters.


Source
<body_text_paragraph>Farmers have been handed an 11th hour GST safety net.
Those who have not registered for the GST, or have not received their
Australian Business Number (ABN), have been thrown a temporary lifeline to
prevent their buyers retaining 48.5pc of the gross sale proceeds in
withholding tax.
Farmers who do not possess an ABN can obtain  a &#8216;safety net&#8217;
letter from the Australian Tax Office so when they supply other businesses,
tax will not be required to be withheld under the new Pay As You Go
arrangements. 
For  businesses with the letter, the rate of withholding tax has been
reduced from 48.5 per cent to nil. The variation covers the period from July
1 to July 31 inclusive, but will lapse as soon as the business is notified
of their ABN.</body_text_paragraph>
</body_text>

Excelon Stylus Results/Output (GOOD)
<value>Farmers have been handed an 11th hour GST safety net. Those who have
not registered for the GST, or have not received their Australian Business
Number (ABN), have been thrown a temporary lifeline to prevent their buyers
retaining 48.5pc of the gross sale proceeds in withholding tax.
Farmers who do not possess an ABN can obtain  a `safety net` letter from the
Australian Tax Office so when they supply other businesses, tax will not be
required to be withheld under the new Pay As You Go arrangements. 
For  businesses with the letter, the rate of withholding tax has been
reduced from 48.5 per cent to nil. The variation covers the period from July
1 to July 31 inclusive, but will lapse as soon as the business is notified
of their ABN.
</value>

Xalan Results/Output (TRUNCATED)
<value>Farmers have been handed an 11th hour GST safety net. Those who have
not registered for the GST, or have not received their Australian Business
Number (ABN), have been thrown a temporary lifeline to prevent their buyers
retaining 48.5pc of the gross sale proceeds in withholding tax. Farmers who
do not possess an ABN can obtain  a </value>

Why does Excelon translate this while Xalan has an error? Is it something to
do with Unicode code pages or sets?

Thanks and Regards,

Bruce.


Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

RE: Excluding nodes

[Kay M. <Mic...@ic...>]

> How do I match everything except a given node?
> 
> I know //*/* matches any node with an element as a parent.

No, it matches every element with an element as a parent.
  
> I want to match every node with an element as a parent except for one node

> with a specific name (news_detail.)
> 

//*/node()[not(self::news_detail)]

Mike Kay 

RE: Preceding-siblings

[Kay M. <Mic...@ic...>]

> Can someone give me a simple example of how to use the 
> preceding-siblings expression properly?

To test if an <item> element is the first <item> child of its parent:

<xsl:if test="not(preceding-sibling::item)">This is the first</xsl:if>

Mike Kay

One last Question

[Bruce L. (c) <bl...@sa...>]

Thanks all of you for your help today.

I have one last question:

The following gets rid of apple mac originated characters in the input XML
file:

translate($paragraphstring, '‘’“”', '``')

However, some of them I want to translate to the " character. This presents
a problem since I don't know how to escape it. So - what would the translate
statement to translate any character to a " character ?

Thanks a Bundle

Bruce



Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Re: Newby Question

[Bron G. <br...@ne...>]

On Mon, Aug 07, 2000 at 10:40:40PM -0500, Bruce Long (c) wrote:
> Hi All,
> 
> I am trying to match all nodes with elements as parents but not the ones
> with a parent element of news_detail ...


Ouch, this is mega expensive..

> but am having trouble with avoiding the news_detail node's children. If I
> can't put the ancestor:: axes in the match pattern as Xalan says, how do I
> exclude the child nodes of the news_detail node?

What are you trying to do with the output?

One possible method would be something like

<xsl:template match="*|@*|node()">
  <xsl:copy>
     <xsl:apply-templates match="*|@*"/>
  </xsl:copy>
</xsl:template>
<xsl:template match="news_detail"/>  <!-- ignore news detail -->

This should copy every node except news_detail, and obviously its children.

-- 
Bron ( hope this is what you needed )

Newby Question

[Bruce L. (c) <bl...@sa...>]

Hi All,

I am trying to match all nodes with elements as parents but not the ones
with a parent element of news_detail ...

match="//*/ancestor::*[name()!='news_detail']"

Using Excelon Stylus I get 
'Expected node test in pattern'

Xalan gives me

Only child:: and attribute:: axes are allowed in match patterns! offending
axes = ancestor.

I get joy from 

match="//*/*[name()!='news_detail']"

thanks to Bron Gondwana

but am having trouble with avoiding the news_detail node's children. If I
can't put the ancestor:: axes in the match pattern as Xalan says, how do I
exclude the child nodes of the news_detail node?

Thanks in Advance for your help.

Best Regards,

Bruce.

Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Re: Should I optimize this?

[Bron G. <br...@ne...>]

On Sun, Aug 06, 2000 at 11:43:03PM -0500, Mike808 wrote:
> In the maintenance branch, I found several instances of the following
> kind of code:
> 
>   %{$obj->{hashref}} = (%{$obj->{hashref}}, $name => $value);
> 
> On the face of it, this appears to perform a copy of the entire hash
> on every add. It would seem to me that the more direct coding of
> 
>   $obj->{hashref}->{$name} = $value;
> 
> would result in faster execution times.

I changed this in my code, but yes - that should definitely be changed.

Good work!

-- 
Bron ( catching up on a few days email here )

Pavel Nejedly

[Bron G. <br...@ne...>]

I've added Pavel to the developers list on sourceforge.

Apologies for my general crapness recently, I've been very busy with
other things.  I probably won't be working on XML::XSLT for at least
3 weeks at this stage, but I'm happy to do the build management still.

I will be checking my email frequently...

-- 
Bron ( project management is fun.. )

Excluding nodes

[Bruce L. (c) <bl...@sa...>]

Hi,

How do I match everything except a given node?

I know //*/* matches any node with an element as a parent. I want to match
every node with an element as a parent except for one node with a specific
name (news_detail.)

Something like -

//*/*/@!=news_detail ???

I can't find an example in Michael Kay's book - is there one?

Bruce.

Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Preceding-siblings

[Bruce L. (c) <bl...@sa...>]

Hi,

Can someone give me a simple example of how to use the preceding-siblings
expression properly?

Thanks much

Bruce

Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Simple Newby XPath Question

[Bruce L. (c) <bl...@sa...>]

Hi All,

How do I match the second teir elements but not their children:

<?xml...?>
	<news_detail>
		<author>
			<author_name>
				<first_name></first_name>
				<middle_initial></middle_initial>
				<last_name></last_name>
			</author_name>
		</author>
	</news_detail>

I have been using 

	<xsl:template match="news_detail/*" name="news_detail">
		.
		.
		.
	</xsl:template>

and calling it thus:
			.
			.
			.
			<xsl:apply-templates select="//*/*"/> <!-- for
another template -->
			<xsl:apply-templates select="news_detail"/>
			.
			.
			.

The problem is that I get the author_name node and the first_name, last_name
and middle_initial child nodes. I only want the node list to include the
<author> tag.

 <item delimiter="," name="author_name"/>
  <item delimiter="," name="first_name"/>
  <item delimiter="," name="middle_initial"/> <!-- I DONT WANT THESE IN THE
OUTPUT -->

What XPath expression do I need to use ?
Also - even though I am calling the two templates explicitly using select,
does the pattern from the first template still carry over ? Why ? How do I
stop this?

Thanks Heaps,

Bruce.

Bruce Long (c)
Sapient | Architects for the New Economy
Level 13, 60 Castlereagh Street	Phone:	+612 9210.2067
Sydney, NSW 2000			Fax:	+612 9210.2500
Australia				
Email:	<bl...@sa...>

Re: Should I optimize this?

[<ma...@ev...>]

>>>>> "M" == Mike808  <mi...@ne...> writes:

    M> Is there some reason I should *not* implement this throughout
    M> the module?

I can't think of a reason not to.

Mark.