From transfire at gmail.com  Mon Aug 14 10:24:22 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 14 Aug 2006 10:24:22 -0400
Subject: [libxml-devel] How to insert before, after and prepend,
	plus speed issue
Message-ID: <4b6f054f0608140724i4f24c234v78f39bf31f81b309@mail.gmail.com>

Hi--

I'm trying to figuere out how to add a node before and after another
node. I think for after I use sibling, but I have no ida about before.

Also I need to prepend a node to the inner xml of a node as opposed to
add it to the end. how is that done?

FYI, if interested I've been working on a lib called Cherry XML that
creates a nice interface much like Hpricot's, but instead uses
adapters  for either REXML or ruby-libxml.

I'm looking foward to using libxml primarily but there's still some
issues with the lib, so I have to fall back to REXML. The most
signifficant thing I've noticed it that I'm getting execution times 5
times slower than REXML. That mkes no sense. Has anyone else seen
this?

Thanks,
T.

From rosco at roscopeco.co.uk  Tue Aug 15 07:33:38 2006
From: rosco at roscopeco.co.uk (Ross Bamford)
Date: Tue, 15 Aug 2006 12:33:38 +0100
Subject: [libxml-devel] How to insert before, after and prepend,
	plus speed issue
In-Reply-To: <4b6f054f0608140724i4f24c234v78f39bf31f81b309@mail.gmail.com>
References: <4b6f054f0608140724i4f24c234v78f39bf31f81b309@mail.gmail.com>
Message-ID: <op.tebo2cukkc9f7k@jukebox.roscopeco>

Hi Trans,

On Mon, 14 Aug 2006 15:24:22 +0100, TRANS <transfire at gmail.com> wrote:

> I'm trying to figuere out how to add a node before and after another
> node. I think for after I use sibling, but I have no ida about before.
>
> Also I need to prepend a node to the inner xml of a node as opposed to
> add it to the end. how is that done?
>

I don't know of a way with the latest release, but if you don't mind using  
CVS head you get handy prev= and next= methods on XML::Node, e.g:

XML::Parser::VERSION
# => "0.3.9"

xd = XML::Parser.string('<foo><bar/><baz/></foo>').parse
# => <?xml version="1.0"?>
<foo>
   <bar/>
   <baz/>
</foo>

bar = xd.root.child
# => <bar/>

bar.prev = XML::Node.new('boo')
# => <boo/>

bar.next = XML::Node.new('too')
# => <too/>

xd
# => <?xml version="1.0"?>
<foo>
   <boo/>
   <bar/>
   <too/>
   <baz/>
</foo>

> FYI, if interested I've been working on a lib called Cherry XML that
> creates a nice interface much like Hpricot's, but instead uses
> adapters  for either REXML or ruby-libxml.
>

Sounds cool, is it online yet?

> I'm looking foward to using libxml primarily but there's still some
> issues with the lib, so I have to fall back to REXML. The most
> signifficant thing I've noticed it that I'm getting execution times 5
> times slower than REXML. That mkes no sense. Has anyone else seen
> this?
>

Hmm, are you using the SAX parser? My first implementation is painfully  
slow, I underestimated the overhead of proc callbacks. Since 0.4.0 is up  
next, I'm thinking of breaking the API (it's new anyway) and replacing the  
Sax parser with a method-based one... My prototype is a bit faster than  
the current one, and I'm still working on it.

If it's not that, I'd appreciate if you'd send over your benchmarks and  
I'll see if I can figure it out.

Cheers,
-- 
Ross Bamford - rosco at roscopeco.co.uk

From rosco at roscopeco.co.uk  Tue Aug 15 08:22:49 2006
From: rosco at roscopeco.co.uk (Ross Bamford)
Date: Tue, 15 Aug 2006 13:22:49 +0100
Subject: [libxml-devel] Request to add HTML parsing
In-Reply-To: <44BC3A9A.6070809@thomaszone.com>
References: <44BC3A9A.6070809@thomaszone.com>
Message-ID: <op.tebrcbu0kc9f7k@jukebox.roscopeco>

Hi Mark,

On Tue, 18 Jul 2006 02:34:18 +0100, Mark Thomas <ruby at thomaszone.com>  
wrote:

> I'm switching to Ruby from Perl, and currently I do all my HTML parsing  
> in
> perl's XML::LibXML. Applying XPath to parse HTML is extremely powerful  
> and
> fast, fast, fast in libxml.
>
> Can you add that feature to the Ruby one? I think it would be easy to do;
> it's just a flag on the parser, which tells libxml to create a DOM from
> HTML instead of XML, and all the XML methods then magically work on the
> HTML!
>

Sure, I'll look into this for 0.4.0.

Cheers,
-- 
Ross Bamford - rosco at roscopeco.co.uk

From rosco at roscopeco.co.uk  Tue Aug 15 08:33:56 2006
From: rosco at roscopeco.co.uk (Ross Bamford)
Date: Tue, 15 Aug 2006 13:33:56 +0100
Subject: [libxml-devel] minor typo in README file example
In-Reply-To: <11a866cb0606151154w29eeb17bj9a0049af6288c160@mail.gmail.com>
References: <11a866cb0606151154w29eeb17bj9a0049af6288c160@mail.gmail.com>
Message-ID: <op.tebruucykc9f7k@jukebox.roscopeco>

Hi,

On Thu, 15 Jun 2006 19:54:28 +0100, Aitor Garay-Romero <aitorgr at gmail.com>  
wrote:

>    In the README file the "reading xml" example is not correct.   
> Somewhere
> it says:
>
>    puts "Node path: #{node.path} \t Contents: #{node}"
>
>    But it should say:
>
>    puts "Node path: #{node.path} \t Contents: #{node.contents}"
>
>    If not the output will not be the one shown on the page.
>

Thanks for that, I've fixed it and will commit as soon as a small CVS  
hiccup is resolved.

-- 
Ross Bamford - rosco at roscopeco.co.uk

From rosco at roscopeco.co.uk  Tue Aug 15 08:37:19 2006
From: rosco at roscopeco.co.uk (Ross Bamford)
Date: Tue, 15 Aug 2006 13:37:19 +0100
Subject: [libxml-devel] Segfault when doing simple parsing
In-Reply-To: <62e8012c0606270614n4d6e7b1ao3c3cdac35a4b9a90@mail.gmail.com>
References: <62e8012c0606270614n4d6e7b1ao3c3cdac35a4b9a90@mail.gmail.com>
Message-ID: <op.tebr0hm2kc9f7k@jukebox.roscopeco>

On Tue, 27 Jun 2006 14:14:05 +0100, Pedro C?rte-Real <pedrocr at gmail.com>  
wrote:

> I'm trying to use libxml to run some XPath on a document I have as a
> string in memory. This is my code for parsing the document:
>
> parser = XML::Parser.new
> parser.string = self.content
> @xmldoc = parser.parse
>
> Doing the same thing in a test script seems to work without
> segfaulting. This is run inside a rails model and segfaults when
> testing it. Here's the backtrace:
>
> 0xb72b37c4 in ruby_xml_parser_parse (self=144875496) at  
> ruby_xml_parser.c:1124
> 1124          xmlFreeDoc(rxpc->ctxt->myDoc);
>
> #0  0xb72b37c4 in ruby_xml_parser_parse (self=144875496)
>     at ruby_xml_parser.c:1124
> #1  0xb7ef050b in rb_iterator_p () from /usr/lib/libruby1.8.so.1.8
> #2  0xb7efaf8c in rb_thread_trap_eval () from /usr/lib/libruby1.8.so.1.8
> (...)
>

We have a few memory handling bugs that are still being ironed out, so  
it's possible you're seeing one of those. Hopefully soon we'll be able to  
release a fixed up version, but in the meantime you may (or may not) get  
more joy from the current CVS version. I'll try to reproduce this and take  
a look at it.

Thanks,
-- 
Ross Bamford - rosco at roscopeco.co.uk

From steve at asklater.com  Fri Aug 18 13:50:34 2006
From: steve at asklater.com (SteveC)
Date: Fri, 18 Aug 2006 18:50:34 +0100
Subject: [libxml-devel] reading from a String?
Message-ID: <20060818175034.GA29102@asklater.com>

The Document api lists new(version) and file(filename):

http://libxml.rubyforge.org/doc/classes/XML/Document.html

I'd like to create a document from an xml file already in a string a-la
REXML::Document.new

Is there some other magic way of doing that?

have fun,

SteveC steve at asklater.com http://www.asklater.com/steve/

From grant at janrain.com  Fri Aug 18 13:53:46 2006
From: grant at janrain.com (Grant Monroe)
Date: Fri, 18 Aug 2006 10:53:46 -0700
Subject: [libxml-devel] reading from a String?
In-Reply-To: <20060818175034.GA29102@asklater.com>
References: <20060818175034.GA29102@asklater.com>
Message-ID: <826ff8bf0608181053u5881531ev3b2fe417625d2ad0@mail.gmail.com>

p = XML::Parser.new
p.string = yourstring
doc = p.parse


On 8/18/06, SteveC <steve at asklater.com> wrote:
> The Document api lists new(version) and file(filename):
>
> http://libxml.rubyforge.org/doc/classes/XML/Document.html
>
> I'd like to create a document from an xml file already in a string a-la
> REXML::Document.new
>
> Is there some other magic way of doing that?
>
> have fun,
>
> SteveC steve at asklater.com http://www.asklater.com/steve/
> _______________________________________________
> libxml-devel mailing list
> libxml-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/libxml-devel
>
>


-- 
 Grant Monroe
 JanRain, Inc.

From mvette13 at gmail.com  Fri Aug 18 13:56:27 2006
From: mvette13 at gmail.com (Mark Van Holstyn)
Date: Fri, 18 Aug 2006 13:56:27 -0400
Subject: [libxml-devel] reading from a String?
In-Reply-To: <826ff8bf0608181053u5881531ev3b2fe417625d2ad0@mail.gmail.com>
References: <20060818175034.GA29102@asklater.com>
	<826ff8bf0608181053u5881531ev3b2fe417625d2ad0@mail.gmail.com>
Message-ID: <e45a289c0608181056w1e4282at3402c1e60eab7a87@mail.gmail.com>

doc = XML::Parser.string( "...data..." ).parse

On 8/18/06, Grant Monroe <grant at janrain.com> wrote:
>
> p = XML::Parser.new
> p.string = yourstring
> doc = p.parse
>
>
> On 8/18/06, SteveC <steve at asklater.com> wrote:
> > The Document api lists new(version) and file(filename):
> >
> > http://libxml.rubyforge.org/doc/classes/XML/Document.html
> >
> > I'd like to create a document from an xml file already in a string a-la
> > REXML::Document.new
> >
> > Is there some other magic way of doing that?
> >
> > have fun,
> >
> > SteveC steve at asklater.com http://www.asklater.com/steve/
> > _______________________________________________
> > libxml-devel mailing list
> > libxml-devel at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/libxml-devel
> >
> >
>
>
> --
> Grant Monroe
> JanRain, Inc.
> _______________________________________________
> libxml-devel mailing list
> libxml-devel at rubyforge.org
> http://rubyforge.org/mailman/listinfo/libxml-devel
>



-- 
Mark Van Holstyn
mvette13 at gmail.com
http://lotswholetime.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/libxml-devel/attachments/20060818/c49ff910/attachment-0001.html 

From steve at asklater.com  Fri Aug 18 14:03:16 2006
From: steve at asklater.com (SteveC)
Date: Fri, 18 Aug 2006 19:03:16 +0100
Subject: [libxml-devel] reading from a String?
In-Reply-To: <826ff8bf0608181053u5881531ev3b2fe417625d2ad0@mail.gmail.com>
References: <20060818175034.GA29102@asklater.com>
	<826ff8bf0608181053u5881531ev3b2fe417625d2ad0@mail.gmail.com>
Message-ID: <20060818180316.GB29102@asklater.com>

Brilliant, thanks for the super quick response.

Can I suggest that this snippet be put in the Reading example?

Here's a little patch to the README

140,148d139
< If you want to parse a xml document already in a string, you can
< extract it with
< 
<   p = XML::Parser.new
<   p.string = yourstring
<   doc = p.parse
< 
< and then use doc as in the above example.
< 


* @ 18/08/06 06:53:46 PM grant at janrain.com wrote:
> p = XML::Parser.new
> p.string = yourstring
> doc = p.parse
> 
> 
> On 8/18/06, SteveC <steve at asklater.com> wrote:
> >The Document api lists new(version) and file(filename):
> >
> >http://libxml.rubyforge.org/doc/classes/XML/Document.html
> >
> >I'd like to create a document from an xml file already in a string a-la
> >REXML::Document.new
> >
> >Is there some other magic way of doing that?
> >
> >have fun,
> >
> >SteveC steve at asklater.com http://www.asklater.com/steve/
> >_______________________________________________
> >libxml-devel mailing list
> >libxml-devel at rubyforge.org
> >http://rubyforge.org/mailman/listinfo/libxml-devel
> >
> >
> 
> 
> -- 
> Grant Monroe
> JanRain, Inc.

have fun,

SteveC steve at asklater.com http://www.asklater.com/steve/

From shimbo at is.naist.jp  Sat Aug 19 11:14:28 2006
From: shimbo at is.naist.jp (Masashi Shimbo)
Date: Sun, 20 Aug 2006 00:14:28 +0900
Subject: [libxml-devel] [PATCH] libxml-ruby gem fails to build
Message-ID: <87lkpk3gh7.wl%shimbo@is.naist.jp>

Hi,

I tried to install libxml-ruby 0.3.8 gem on my Ubuntu Linux box
without success.  The installation apparently succeeded with no errors, 
but 'gem check -t libxml-ruby' fails with the message

  /usr/local/lib/site_ruby/1.8/rubygems/custom_require.rb:27:in
  `gem_original_require': no such file to load -- xml/libxml_so
  (LoadError)

It seems that RubyGem expects the source directory for extensions to be
different from where they are installed (typically 'lib/'), but the
current libxml-ruby gem uses 'ext/' for both.

Consequently, the following happens with 'gem install';
Note that 'gem install' runs 'extconf.rb' in 'ext/xml/' and then 
'make; make install; make clean':

1. The file 'ext/xml/libxml_so.so' compiles fine.

2. Because RUBYARCHDIR is 'ext/xml/', 'make install' does nothing.
   # N.B. gem overwrites RUBYLIBDIR and RUBYARCHDIR to the value of
   # 'require_path' field in the gem spec just before 'make' is run.

3. gem finally runs 'make clean' and 'ext/xml/libxml_so.so' is removed.

Attached is a patch that fixes the problem.  It essentially

- creates 'lib/xml' directory;
- sets the gem spec 'require_path' field to 'lib/xml' (in place of
  'ext/xml');
- moves 'ext/xml/libxml.rb' to 'lib/xml' directory; and
- installs libxml_so.so in 'lib/xml' upon 'gem install'.

My environment is: Ubuntu Linux 6.06 / RubyGem 0.9.0 / Ruby 1.8.4

Masashi Shimbo

-------------- next part --------------
A non-text attachment was scrubbed...
Name: libxml-ruby-0.3.8.diff
Type: application/octet-stream
Size: 5798 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/libxml-devel/attachments/20060820/581a5aae/attachment.obj 

From luc_van_deuren at yahoo.com  Tue Aug 29 21:27:02 2006
From: luc_van_deuren at yahoo.com (Luc Van Deuren)
Date: Tue, 29 Aug 2006 18:27:02 -0700 (PDT)
Subject: [libxml-devel] cdata comment patch
Message-ID: <20060830012702.92362.qmail@web32410.mail.mud.yahoo.com>

Hello,

I just needed #CDATA and comment into my xml output.
making a cdata or comment node did not seem to work,
so i made a quick patch,  (cdata_comment_patch.zip)

ruby_xml_node.c 
    
    node = XML::Node.new_cdata("mycdata")
        will create a #CDATA node
         "<![CDATA[mycdata]]>"
    node << "_add_to_it"
        will just add to the data.
        " <![CDATA[mycdata_add_to_it]]>
     
    node = XML::Node.new_comment("mycomment")
         will create a comment node
          "<!--mycomment-->"
     node << "_add_to_it"
         will just add to the comment.
         " <![CDATA[mycomment_add_to_it]]>

+ two testfiles
    tests/tc_xml_node7
    tests/tc_xml_node8

    for testing both
 

Question: does the XML::Node.new_cdata and XML::Node.new_comment
seems a good idea?
or is there a more sofisticated solution?

btw, i'm happy this libxml-ruby exists, thanks for the effort!

Luc,

 		
---------------------------------
Do you Yahoo!?
 Everyone is raving about the  all-new Yahoo! Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/libxml-devel/attachments/20060829/7505291a/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cdata_comment_patch.zip
Type: application/zip
Size: 2069 bytes
Desc: 1810481609-cdata_comment_patch.zip
Url : http://rubyforge.org/pipermail/libxml-devel/attachments/20060829/7505291a/attachment.zip