Re: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-26 Thread Reinhard Pötz
Jos Snellings wrote:
 Indeed, Robby, this is clearly the behaviour we are observing. But,
 even if you avoid the minimized tags, I have the impression that the
 XSLT step is converting empty tag pairs to minimized ones. Shouldn't
 we avoid this, given the recommendation?

Some weeks ago I added alternative serializers for HTML/XML/XHTML to
Cocoon 3 trunk. Those take care of many of those tricky serialization
problems. Look for Encoding Serializers in the samples.

-- 
Reinhard Pötz   Managing Director, {Indoqa} GmbH
 http://www.indoqa.com/en/people/reinhard.poetz/

Member of the Apache Software Foundation
Apache Cocoon Committer, PMC member  reinh...@apache.org


-
To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org
For additional commands, e-mail: users-h...@cocoon.apache.org



Re: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-26 Thread Jos Snellings
I have seen these, they are in 'cocoon-optional', but how to configure
in the pipeline? 
map:serialize type=exhtml/
or ehtml?
(are there new samples? I did not know)


On Thu, 2009-11-26 at 18:22 +0100, Reinhard Pötz wrote:
 Jos Snellings wrote:
  Indeed, Robby, this is clearly the behaviour we are observing. But,
  even if you avoid the minimized tags, I have the impression that the
  XSLT step is converting empty tag pairs to minimized ones. Shouldn't
  we avoid this, given the recommendation?
 
 Some weeks ago I added alternative serializers for HTML/XML/XHTML to
 Cocoon 3 trunk. Those take care of many of those tricky serialization
 problems. Look for Encoding Serializers in the samples.
 



-
To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org
For additional commands, e-mail: users-h...@cocoon.apache.org



Re: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-26 Thread Reinhard Pötz
Jos Snellings wrote:
 I have seen these, they are in 'cocoon-optional', but how to configure
 in the pipeline? 
 map:serialize type=exhtml/
 or ehtml?

yes, ehtml, exhtml and exml.

-- 
Reinhard Pötz   Managing Director, {Indoqa} GmbH
 http://www.indoqa.com/en/people/reinhard.poetz/

Member of the Apache Software Foundation
Apache Cocoon Committer, PMC member  reinh...@apache.org


-
To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org
For additional commands, e-mail: users-h...@cocoon.apache.org



Re: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-25 Thread Jos Snellings
Hi !

Here is what Firebug has to tell: 
when an xhtml serializer is selected, this is the output:

script src=/thesaurus/js/jquery.js type=text/javascript/
script src=/thesaurus/js/jquery_cookie.js type=text/javascript/
script src=/thesaurus/js/jquery_treeview.js type=text/javascript/
script type=text/javascript
$(function() {
$(#tree).treeview({
collapsed: true,
animated: medium,
control:#sidetreecontrol,
persist: location
});
})

/script
---
Firefox sees only the first javascript.
By the way: in het xslt rule the script tags are: 
script src=/thesaurus/js/jquery.js type=text/javascript/script
but these are converted to single elements script/, and it looks as if 
Firefox has a problem.
-
A DOM inspection confirms this: 
It looks as if the subsequent script tags are regarded as sub-elements of the 
first script tag.


I haven't tested with other browsers.
Is this to be considered as an issue? Fact is that the produced xhtml is valid 
and *should* work.

Kind regards,
Jos


On Wed, 2009-11-25 at 13:16 +0100, Jos Snellings wrote:
 I was about to do that, it really points in the direction of firefox.
 I will let you know
 
 
 On Wed, 2009-11-25 at 12:44 +0100, Reinhard Pötz wrote:
  Jos Snellings wrote:
   Hello !
   
   I have a strange observation to share:
   when multiple javascripts are included in an XSLT transformation, to
   land in the head of an html page, it is observed that some are never
   fetched, by FireFox 3.5.5 (Ubuntu). 
   
   I clarify: exerpt 1: from sitemap, pipeline for static resources:
map:pipeline type=async-caching 
 map:match pattern=images/{name}.jpg
  map:read src=presentation/images/{map:name}.jpg
   mime-type=image/jpeg /
 /map:match
  map:match pattern=images/{name}.gif
   map:read src=presentation/images/{map:name}.gif
   mime-type=image/gif /
 /map:match
 map:match pattern=css/{name}.css
  map:read src=presentation/style/{map:name}.css
   mime-type=text/css /
 /map:match
 map:match pattern=css/images/{name}
  map:read src=presentation/style/images/{map:name}
   mime-type=image/png /
 /map:match
 map:match pattern=js/{name}.js
   map:read src=presentation/javascript/{map:name}.js /
 /map:match
   
   --
   xsl:template match=/
   html
   head
   title Herein Thesaurus /title   
 
  link rel=stylesheet type=text/css
   href=/thesaurus/css/herein.css/
  link rel=stylesheet type=text/css
   href=/thesaurus/css/jquery_treeview.css/
   
 script type=text/javascript
   src=/thesaurus/js/jquery.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_cookie.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_treeview.js/script
   
  script type=text/javascript
   src=/thesaurus/js/thesaurus.js/script
  
  script type=text/javascript
  
 $(function() {
 $(#tree).treeview({
 collapsed: true,
 animated: medium,
 control:#sidetreecontrol,
 persist: location
 });
 })
 
 /script 
   /head
   -
   I detailed the log to debug level for the cocoon output, and, to my
   amazement, I find that the pipeline does not react at all: javascripts 
   jquery_treeview and jquery_cookie are never fetched.
   
   --
   Observation: when I place the same html under the form of a static html
   page, all the scripts are nicely fetched.
   
   
   What could be the cause of this - admittedly strange - phenomenon?
   Anybody observed a similar test case?
   
   Reinhard, as you make heavy use of rich client javascripting, do you
   know about this?
   Am I overlooking something essential?
   I see no reason why the result of a xslt transformation would behave
   different, so obviously I am.
  
  This sounds like some browser caching issue.
  Have you used Firebug to track down the problem?
  
 
 
 
 -
 To unsubscribe, e-mail: users-unsubscr...@cocoon.apache.org
 For additional commands, e-mail: users-h...@cocoon.apache.org
 
 



-
To unsubscribe, e-mail: 

RE: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-25 Thread Robby Pelssers
I've seen this behavior for other tags as well...

For instance with empty div tags

In that case subsequent tags are sometimes treated as children of the empty 
preceding div tag.


Kind regards,
Robby

-Original Message-
From: Jos Snellings [mailto:jos.snelli...@pandora.be] 
Sent: Wednesday, November 25, 2009 4:51 PM
To: users@cocoon.apache.org
Subject: Re: cocoon 3: xslt and pickup java scripts: Firebug

Hi !

Here is what Firebug has to tell: 
when an xhtml serializer is selected, this is the output:

script src=/thesaurus/js/jquery.js type=text/javascript/
script src=/thesaurus/js/jquery_cookie.js type=text/javascript/
script src=/thesaurus/js/jquery_treeview.js type=text/javascript/
script type=text/javascript
$(function() {
$(#tree).treeview({
collapsed: true,
animated: medium,
control:#sidetreecontrol,
persist: location
});
})

/script
---
Firefox sees only the first javascript.
By the way: in het xslt rule the script tags are: 
script src=/thesaurus/js/jquery.js type=text/javascript/script
but these are converted to single elements script/, and it looks as if 
Firefox has a problem.
-
A DOM inspection confirms this: 
It looks as if the subsequent script tags are regarded as sub-elements of the 
first script tag.


I haven't tested with other browsers.
Is this to be considered as an issue? Fact is that the produced xhtml is valid 
and *should* work.

Kind regards,
Jos


On Wed, 2009-11-25 at 13:16 +0100, Jos Snellings wrote:
 I was about to do that, it really points in the direction of firefox.
 I will let you know
 
 
 On Wed, 2009-11-25 at 12:44 +0100, Reinhard Pötz wrote:
  Jos Snellings wrote:
   Hello !
   
   I have a strange observation to share:
   when multiple javascripts are included in an XSLT transformation, to
   land in the head of an html page, it is observed that some are never
   fetched, by FireFox 3.5.5 (Ubuntu). 
   
   I clarify: exerpt 1: from sitemap, pipeline for static resources:
map:pipeline type=async-caching 
 map:match pattern=images/{name}.jpg
  map:read src=presentation/images/{map:name}.jpg
   mime-type=image/jpeg /
 /map:match
  map:match pattern=images/{name}.gif
   map:read src=presentation/images/{map:name}.gif
   mime-type=image/gif /
 /map:match
 map:match pattern=css/{name}.css
  map:read src=presentation/style/{map:name}.css
   mime-type=text/css /
 /map:match
 map:match pattern=css/images/{name}
  map:read src=presentation/style/images/{map:name}
   mime-type=image/png /
 /map:match
 map:match pattern=js/{name}.js
   map:read src=presentation/javascript/{map:name}.js /
 /map:match
   
   --
   xsl:template match=/
   html
   head
   title Herein Thesaurus /title   
 
  link rel=stylesheet type=text/css
   href=/thesaurus/css/herein.css/
  link rel=stylesheet type=text/css
   href=/thesaurus/css/jquery_treeview.css/
   
 script type=text/javascript
   src=/thesaurus/js/jquery.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_cookie.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_treeview.js/script
   
  script type=text/javascript
   src=/thesaurus/js/thesaurus.js/script
  
  script type=text/javascript
  
 $(function() {
 $(#tree).treeview({
 collapsed: true,
 animated: medium,
 control:#sidetreecontrol,
 persist: location
 });
 })
 
 /script 
   /head
   -
   I detailed the log to debug level for the cocoon output, and, to my
   amazement, I find that the pipeline does not react at all: javascripts 
   jquery_treeview and jquery_cookie are never fetched.
   
   --
   Observation: when I place the same html under the form of a static html
   page, all the scripts are nicely fetched.
   
   
   What could be the cause of this - admittedly strange - phenomenon?
   Anybody observed a similar test case?
   
   Reinhard, as you make heavy use of rich client javascripting, do you
   know about this?
   Am I overlooking something essential?
   I see no reason why the result of a xslt transformation would behave
   different, so obviously I am

RE: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-25 Thread Robby Pelssers
Some article explains a bit more:


http://forums.mozillazine.org/viewtopic.php?f=25t=297285

The first empty DIV closes itself with the trailing slash. This is allowed in 
XML (XHTML is XML), but it is not allowed in HTML. If your document (1) has an 
XHTML doctype declaration at the beginning of the document, (2) has the proper 
namespace declaration, and (3) is being served with the proper mime type by the 
server (application/xhtml+xml), there is no difference between those two empty 
DIVs.

However, if you miss any of those three criteria, your document is not going to 
be rendered as XML, and you will not get the results you expect. With the wrong 
mime type sent by the server, the document will be rendered as HTML. In HTML, 
div / syntax will be read as an opening div tag, and the browser will expect 
a closing tag.

The XHTML compatibility guidelines advise you not to use the minimized tag 
syntax for compatibility reasons. Even if Firefox renders your XHTML properly, 
older browsers that only understand HTML will choke and die on the DIV that is 
not properly closed in HTML. Read 4.3 and 4.6 here.



-Original Message-
From: Robby Pelssers [mailto:robby.pelss...@ciber.com] 
Sent: Wednesday, November 25, 2009 4:58 PM
To: users@cocoon.apache.org
Subject: RE: cocoon 3: xslt and pickup java scripts: Firebug

I've seen this behavior for other tags as well...

For instance with empty div tags

In that case subsequent tags are sometimes treated as children of the empty 
preceding div tag.


Kind regards,
Robby

-Original Message-
From: Jos Snellings [mailto:jos.snelli...@pandora.be] 
Sent: Wednesday, November 25, 2009 4:51 PM
To: users@cocoon.apache.org
Subject: Re: cocoon 3: xslt and pickup java scripts: Firebug

Hi !

Here is what Firebug has to tell: 
when an xhtml serializer is selected, this is the output:

script src=/thesaurus/js/jquery.js type=text/javascript/
script src=/thesaurus/js/jquery_cookie.js type=text/javascript/
script src=/thesaurus/js/jquery_treeview.js type=text/javascript/
script type=text/javascript
$(function() {
$(#tree).treeview({
collapsed: true,
animated: medium,
control:#sidetreecontrol,
persist: location
});
})

/script
---
Firefox sees only the first javascript.
By the way: in het xslt rule the script tags are: 
script src=/thesaurus/js/jquery.js type=text/javascript/script
but these are converted to single elements script/, and it looks as if 
Firefox has a problem.
-
A DOM inspection confirms this: 
It looks as if the subsequent script tags are regarded as sub-elements of the 
first script tag.


I haven't tested with other browsers.
Is this to be considered as an issue? Fact is that the produced xhtml is valid 
and *should* work.

Kind regards,
Jos


On Wed, 2009-11-25 at 13:16 +0100, Jos Snellings wrote:
 I was about to do that, it really points in the direction of firefox.
 I will let you know
 
 
 On Wed, 2009-11-25 at 12:44 +0100, Reinhard Pötz wrote:
  Jos Snellings wrote:
   Hello !
   
   I have a strange observation to share:
   when multiple javascripts are included in an XSLT transformation, to
   land in the head of an html page, it is observed that some are never
   fetched, by FireFox 3.5.5 (Ubuntu). 
   
   I clarify: exerpt 1: from sitemap, pipeline for static resources:
map:pipeline type=async-caching 
 map:match pattern=images/{name}.jpg
  map:read src=presentation/images/{map:name}.jpg
   mime-type=image/jpeg /
 /map:match
  map:match pattern=images/{name}.gif
   map:read src=presentation/images/{map:name}.gif
   mime-type=image/gif /
 /map:match
 map:match pattern=css/{name}.css
  map:read src=presentation/style/{map:name}.css
   mime-type=text/css /
 /map:match
 map:match pattern=css/images/{name}
  map:read src=presentation/style/images/{map:name}
   mime-type=image/png /
 /map:match
 map:match pattern=js/{name}.js
   map:read src=presentation/javascript/{map:name}.js /
 /map:match
   
   --
   xsl:template match=/
   html
   head
   title Herein Thesaurus /title   
 
  link rel=stylesheet type=text/css
   href=/thesaurus/css/herein.css/
  link rel=stylesheet type=text/css
   href=/thesaurus/css/jquery_treeview.css/
   
 script type=text/javascript
   src=/thesaurus/js/jquery.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_cookie.js/script
   script type=text/javascript
   src=/thesaurus/js/jquery_treeview.js/script
   
  script type=text/javascript
   src

RE: cocoon 3: xslt and pickup java scripts: Firebug

2009-11-25 Thread Jos Snellings
Indeed, Robby, this is clearly the behaviour we are observing.
But, even if you avoid the minimized tags, I have the impression 
that the XSLT step is converting empty tag pairs to minimized ones.
Shouldn't we avoid this, given the recommendation?

Jos


On Wed, 2009-11-25 at 17:02 +0100, Robby Pelssers wrote:
 Some article explains a bit more:
 
 
 http://forums.mozillazine.org/viewtopic.php?f=25t=297285
 
 The first empty DIV closes itself with the trailing slash. This is allowed in 
 XML (XHTML is XML), but it is not allowed in HTML. If your document (1) has 
 an XHTML doctype declaration at the beginning of the document, (2) has the 
 proper namespace declaration, and (3) is being served with the proper mime 
 type by the server (application/xhtml+xml), there is no difference between 
 those two empty DIVs.
 
 However, if you miss any of those three criteria, your document is not going 
 to be rendered as XML, and you will not get the results you expect. With the 
 wrong mime type sent by the server, the document will be rendered as HTML. In 
 HTML, div / syntax will be read as an opening div tag, and the browser will 
 expect a closing tag.
 
 The XHTML compatibility guidelines advise you not to use the minimized tag 
 syntax for compatibility reasons. Even if Firefox renders your XHTML 
 properly, older browsers that only understand HTML will choke and die on the 
 DIV that is not properly closed in HTML. Read 4.3 and 4.6 here.
 
 
 
 -Original Message-
 From: Robby Pelssers [mailto:robby.pelss...@ciber.com] 
 Sent: Wednesday, November 25, 2009 4:58 PM
 To: users@cocoon.apache.org
 Subject: RE: cocoon 3: xslt and pickup java scripts: Firebug
 
 I've seen this behavior for other tags as well...
 
 For instance with empty div tags
 
 In that case subsequent tags are sometimes treated as children of the empty 
 preceding div tag.
 
 
 Kind regards,
 Robby
 
 -Original Message-
 From: Jos Snellings [mailto:jos.snelli...@pandora.be] 
 Sent: Wednesday, November 25, 2009 4:51 PM
 To: users@cocoon.apache.org
 Subject: Re: cocoon 3: xslt and pickup java scripts: Firebug
 
 Hi !
 
 Here is what Firebug has to tell: 
 when an xhtml serializer is selected, this is the output:
 
 script src=/thesaurus/js/jquery.js type=text/javascript/
 script src=/thesaurus/js/jquery_cookie.js type=text/javascript/
 script src=/thesaurus/js/jquery_treeview.js type=text/javascript/
 script type=text/javascript
   $(function() {
   $(#tree).treeview({
   collapsed: true,
   animated: medium,
   control:#sidetreecontrol,
   persist: location
   });
   })
   
 /script
 ---
 Firefox sees only the first javascript.
 By the way: in het xslt rule the script tags are: 
 script src=/thesaurus/js/jquery.js type=text/javascript/script
 but these are converted to single elements script/, and it looks as if 
 Firefox has a problem.
 -
 A DOM inspection confirms this: 
 It looks as if the subsequent script tags are regarded as sub-elements of the 
 first script tag.
 
 
 I haven't tested with other browsers.
 Is this to be considered as an issue? Fact is that the produced xhtml is 
 valid and *should* work.
 
 Kind regards,
 Jos
 
 
 On Wed, 2009-11-25 at 13:16 +0100, Jos Snellings wrote:
  I was about to do that, it really points in the direction of firefox.
  I will let you know
  
  
  On Wed, 2009-11-25 at 12:44 +0100, Reinhard Pötz wrote:
   Jos Snellings wrote:
Hello !

I have a strange observation to share:
when multiple javascripts are included in an XSLT transformation, to
land in the head of an html page, it is observed that some are never
fetched, by FireFox 3.5.5 (Ubuntu). 

I clarify: exerpt 1: from sitemap, pipeline for static resources:
 map:pipeline type=async-caching 
  map:match pattern=images/{name}.jpg
   map:read src=presentation/images/{map:name}.jpg
mime-type=image/jpeg /
  /map:match
   map:match pattern=images/{name}.gif
map:read src=presentation/images/{map:name}.gif
mime-type=image/gif /
  /map:match
  map:match pattern=css/{name}.css
   map:read src=presentation/style/{map:name}.css
mime-type=text/css /
  /map:match
  map:match pattern=css/images/{name}
   map:read src=presentation/style/images/{map:name}
mime-type=image/png /
  /map:match
  map:match pattern=js/{name}.js
map:read src=presentation/javascript/{map:name}.js /
  /map:match

--
xsl:template match=/
html
head
title Herein Thesaurus /title   
  
   link rel=stylesheet type=text/css