Issues - tenderlove/nokogiri

tenderlove / nokogiri

668

Description:

Nokogiri (鋸) is an HTML, XML, SAX, and Reader parser with XPath and CSS selector support.

Sort by: Priority | Votes | Last Updated

Loading…

159.

0 votes

JRuby version test on Windows is broken
5 comments Created 5 days ago by casebook

Title

Body
raise(RuntimeError, "Nokogiri requires JRuby 1.4.0RC1 or later on Windows") if JRUBY_VERSION < "1.4.0RC1" fails if my version is 1.4.0 reported by bhauff
or cancel
```
 raise(RuntimeError, "Nokogiri requires JRuby 1.4.0RC1 or later on Windows") if JRUBY_VERSION < "1.4.0RC1"
```
fails if my version is 1.4.0

reported by bhauff
Comments

bhauff Mon Nov 09 14:08:06 -0800 2009 | link

Thanks for the help on this.

Comment
Thanks for the help on this.
or cancel

casebook Mon Nov 09 14:30:18 -0800 2009 | link

fixed in 6510fe7

Comment
fixed in 6510fe7142feb1c6e26143008cc984a74769415e
or cancel

casebook Mon Nov 09 14:31:11 -0800 2009 | link

ah, for clarity, this is flavorjones. wrong login.

Comment
ah, for clarity, this is flavorjones. wrong login.
or cancel

flavorjones Mon Nov 09 14:35:02 -0800 2009 | link

I concur. that was me.

Comment
I concur. that was me.
or cancel

flavorjones Mon Nov 09 14:37:57 -0800 2009 | link

closing. this will be in 1.4.1 sometime in the next week or so.

Comment
closing. this will be in 1.4.1 sometime in the next week or so.
or cancel

Parsed with GitHub Flavored Markdown
158.

0 votes

1.4.1
x

tenderlove
x

Ancestor search doesn't work with a css query.
1 comment Created 9 days ago by chriseppstein

Title

Body
See this script for an example: http://gist.github.com/227429
or cancel

See this script for an example: http://gist.github.com/227429

Comments

tenderlove Thu Nov 05 19:46:04 -0800 2009 | link

Node#matches? works in nodes contained by a DocumentFragment. closed by d41db1a

Comment
Node#matches? works in nodes contained by a DocumentFragment. closed by d41db1a1d949ab7827b54e2789b5ee1addf2c45b
or cancel

Parsed with GitHub Flavored Markdown
157.

0 votes

1.4.1
x

tenderlove
x

Add the ":self" psuedo selector
1 comment Created 9 days ago by tenderlove

Title

Body
Add the ":self" pseudo selector so that people can have CSS expressions like this: ":self > foo" which would be equivalent to this: "./foo"
or cancel
Add the ":self" pseudo selector so that people can have CSS expressions like this:
```
":self > foo"
```
which would be equivalent to this:
```
"./foo"
```
Comments

tenderlove Mon Nov 09 15:37:11 -0800 2009 | link

Fixed in 55fbf25

Comment
Fixed in 55fbf2508892a479ac556081409d299115bcc83d
or cancel

Parsed with GitHub Flavored Markdown
155.

0 votes

Nokogiri::HTML::DocumentFragment#parse regression in how UTF8 is handled
2 comments Created 15 days ago by naofumi

Title

Body
I'm comparing 1.3.3 on rubyforge, with the Oct. 29 09:05:43 2009 -0700 commit 7fbf2622ddb7364e21bfed70759203f7f86da596 . I'm on Ruby 1.8.7 ruby 1.8.7 (2009-06-12 patchlevel 174) [i686-darwin10] LibXML2 is 2.7.3. With 1.3.3, DocumentFragment works OK with UTF8 strings. <code> puts Nokogiri::HTML::DocumentFragment.parse(%Q(<body>こんにちは</body>)).to_html(:encoding => 'UTF-8') => <body>こんにちは</body> </code> However, with the latest GitHub version, the result is <code>=><body>ããã«ã¡ã¯</body></code> Interestingly, if I add a <meta charset> tag, the latest GitHub version recognizes it and processes UTF8 strings correctly. The following is an example. <code> puts Nokogiri::HTML::DocumentFragment.parse(%Q(<meta content="text/html; charset=UTF8" http-equiv="content-type">\n<body>こんにちは</body>)).to_html(:encoding => 'UTF-8') => <meta content="text/html; charset=UTF8" http-equiv="content-type"><body>こんにちは</body> </code> I looked into the source, but I couldn't get any closer to the cause.
or cancel

I'm comparing 1.3.3 on rubyforge, with the Oct. 29 09:05:43 2009 -0700 commit 7fbf262 .
I'm on Ruby 1.8.7 ruby 1.8.7 (2009-06-12 patchlevel 174) [i686-darwin10]
LibXML2 is 2.7.3.

With 1.3.3, DocumentFragment works OK with UTF8 strings.
puts Nokogiri::HTML::DocumentFragment.parse(%Q(<body>こんにちは</body>)).to_html(:encoding => 'UTF-8') => <body>こんにちは</body>
However, with the latest GitHub version, the result is
=><body>ããã«ã¡ã¯</body>

Interestingly, if I add a <meta charset> tag, the latest GitHub version recognizes it and processes UTF8 strings correctly. The following is an example.
puts Nokogiri::HTML::DocumentFragment.parse(%Q(<meta content="text/html; charset=UTF8" http-equiv="content-type">\n<body>こんにちは</body>)).to_html(:encoding => 'UTF-8') => <meta content="text/html; charset=UTF8" http-equiv="content-type"><body>こんにちは</body>
I looked into the source, but I couldn't get any closer to the cause.

Comments

tenderlove Fri Oct 30 17:45:51 -0700 2009 | link

バグレポートを送ってくれてありがとう。

修正しました！

Thanks for using Nokogiri.

Comment
バグレポートを送ってくれてありがとう。修正しました！ Thanks for using Nokogiri.
or cancel

tenderlove Fri Oct 30 17:46:44 -0700 2009 | link

Fixed in 7a77846

Comment
Fixed in 7a7784609f9ea588c7fc5f8023e4c0920dac4b1e
or cancel

Parsed with GitHub Flavored Markdown
154.

0 votes

Segmentation fault when creating new node with 'Self' reference
1 comment Created 16 days ago by jhingo

Title

Body
I get a [BUG] Segmentation Fault with the following code: class XMLDocument < Nokogiri::XML::Document def initialize() super() body_node = Nokogiri::XML::Node.new("body",self) body_node.content = "stuff" self.root = body_node end end If I execute the following 'xmldoc=XMLDocument.new()' the seg fault occurs on the line: body_node = Nokogiri::XML::Node.new("body",self) If I create a separate 'tempdoc=Nokogiri::XML::Document.new' in the initialize() and change the references from 'self' to 'tempdoc' in the rest of the code then there's no fault. So the fault is being triggered by the 'self' reference. Running Windows XP ruby 1.8.6 nokogiri 1.3.3 libxml 2.7.3 I'm a newbie so apologies if I'm doing anything moronic. Cheers!
or cancel
I get a [BUG] Segmentation Fault with the following code:
```
class XMLDocument < Nokogiri::XML::Document
 def initialize()
 super()
 body_node = Nokogiri::XML::Node.new("body",self)
 body_node.content = "stuff"
 self.root = body_node
 end
end
```
If I execute the following 'xmldoc=XMLDocument.new()' the seg fault occurs on the line:
body_node = Nokogiri::XML::Node.new("body",self)

If I create a separate 'tempdoc=Nokogiri::XML::Document.new' in the initialize() and change the references from 'self' to 'tempdoc' in the rest of the code then there's no fault. So the fault is being triggered by the 'self' reference.

Running
Windows XP
ruby 1.8.6
nokogiri 1.3.3
libxml 2.7.3

I'm a newbie so apologies if I'm doing anything moronic.

Cheers!
Comments
tenderlove Thu Oct 29 09:05:48 -0700 2009 | link
ext/nokogiri/xml_document.c (Nokogiri_wrap_xml_document) init called after the tuple is set up. closed by 7fbf262
Comment
* ext/nokogiri/xml_document.c (Nokogiri_wrap_xml_document) init called after the tuple is set up. closed by 7fbf2622ddb7364e21bfed70759203f7f86da596
or cancel
Parsed with GitHub Flavored Markdown
153.

0 votes

Fragment nodes with namespaces should work properly
1 comment Created 18 days ago by flavorjones

Title

Body
Reported by Iñaki Baz Castillo on the mailing list. Creating a fragment with a namespace makes the prefix part of the tag name, and (arbitrarily?) uses the namespace of the document root's first child.
or cancel

Reported by Iñaki Baz Castillo on the mailing list.

Creating a fragment with a namespace makes the prefix part of the tag name, and (arbitrarily?) uses the namespace of the document root's first child.

Comments

flavorjones Tue Oct 27 07:58:44 -0700 2009 | link

Closed by 597195f

Comment
Closed by 597195ff8fe471e5350581c2d5cce704fcf87439
or cancel

Parsed with GitHub Flavored Markdown
152.

0 votes

NodeSet.wrap does not preserve document structure
2 comments Created 25 days ago by flavorjones

Title

Body
Failing spec: def test_wrap_preserves_document_structure assert_equal "employeeId", @xml.at_xpath("//employee").children.detect{|j| ! j.text? }.name @xml.xpath("//employeeId[text()='EMP0001']").wrap("<wrapper/>") assert_equal "wrapper", @xml.at_xpath("//employee").children.detect{|j| ! j.text? }.name end
or cancel
Failing spec:
```
 def test_wrap_preserves_document_structure
 assert_equal "employeeId",
 @xml.at_xpath("//employee").children.detect{|j| ! j.text? }.name
 @xml.xpath("//employeeId[text()='EMP0001']").wrap("<wrapper/>")
 assert_equal "wrapper",
 @xml.at_xpath("//employee").children.detect{|j| ! j.text? }.name
 end
```
Comments

flavorjones Mon Oct 19 20:10:58 -0700 2009 | link

NodeSet.wrap now preserves document structure. closed by f7388be.

Comment
NodeSet.wrap now preserves document structure. closed by f7388bebc9dbb8dfb0ba4f48e18c80452498b943.
or cancel

flavorjones Mon Oct 19 20:12:12 -0700 2009 | link

and 2d3db36

Comment
and 2d3db36732af54b6ced5ec23a8db5a8a6016b993
or cancel

Parsed with GitHub Flavored Markdown
151.

-2 votes

Allow builder.to_xml to be passed options like :new_lines => false, or :spacing => false
4 comments Created 27 days ago by KieranP

Title

Body
Would also be nice to see a :new_lines => false option at some point, so you get back to back elements, no extra spacing. Making the builder object searchable would be awesome too. e.g. builder.at('//root').to_xml(:indent => 0, :new_lines => false) or better yet: builder.root.to_xml(:spacing => false)
or cancel
Would also be nice to see a :new_lines => false option at some point, so you get back to back elements, no extra spacing. Making the builder object searchable would be awesome too. e.g.

builder.at('//root').to_xml(:indent => 0, :new_lines => false)

or better yet:

builder.root.to_xml(:spacing => false)

Comments
tenderlove Thu Oct 29 09:06:37 -0700 2009 | link

This was merged in with fd6fc58

Comment
This was merged in with fd6fc58de82e834248f7cd5c0b1c22cab3442f99
or cancel

KieranP Thu Oct 29 12:53:06 -0700 2009 | link

Right, but the :new_lines and :spacing options don't exist yet. I was meaning someone add code for these options so we can make the resulting xml completely compact (no indents, no new lines etc).

Comment
Right, but the :new_lines and :spacing options don't exist yet. I was meaning someone add code for these options so we can make the resulting xml completely compact (no indents, no new lines etc).
or cancel
tenderlove Thu Oct 29 13:47:11 -0700 2009 | link
ah, oops. I thought this was the same thing.

The document on the builder is searchable. You can do:

builder.doc.at('//root')

In fact, Builder#doc just returns a Nokogiri::XML::Document that you can manipulate as your normally would.
Comment
ah, oops. I thought this was the same thing. The document on the builder is searchable. You can do: builder.doc.at('//root') In fact, Builder#doc just returns a Nokogiri::XML::Document that you can manipulate as your normally would.
or cancel
KieranP Thu Oct 29 16:05:00 -0700 2009 | link
Awesome. So all that's left is a way to remove new lines. At the moment we do

.gsub(/\n/, '')

But any content with new lines gets compacted too.

>> builder = Nokogiri::XML::Builder.new >> builder.root { |xml| xml.test('hey') } >> builder.doc.root.to_xml(:indent => 0) => "<root>\n<test>hey</test>\n</root>"

Is there a way to do this?

>> builder = Nokogiri::XML::Builder.new >> builder.root { |xml| xml.test('hey') } >> builder.doc.root.to_xml(:indent => 0, :new_lines => false) => "<root><test>hey</test></root>"
Comment
Awesome. So all that's left is a way to remove new lines. At the moment we do .gsub(/\n/, '') But any content with new lines gets compacted too. >> builder = Nokogiri::XML::Builder.new >> builder.root { |xml| xml.test('hey') } >> builder.doc.root.to_xml(:indent => 0) => "<root>\n<test>hey</test>\n</root>" Is there a way to do this? >> builder = Nokogiri::XML::Builder.new >> builder.root { |xml| xml.test('hey') } >> builder.doc.root.to_xml(:indent => 0, :new_lines => false) => "<root><test>hey</test></root>"
or cancel
Parsed with GitHub Flavored Markdown
150.

0 votes

1.4.1
x

tenderlove
x

Nokogiri::XML ignore my set encoding
2 comments Created about 1 month ago by paranormal

Title

Body
Reproduce Nokogiri::XML(open('http://www.cite-sciences.fr/rss/ressources/fr/faq_fr_20.xml').read, nil, 'UTF-8', 18543) Nokogiri::XML::SyntaxError: Unsupported encoding UTF-85 This xml, not valid becouse <?xml version="1.0" encoding="UTF-85"> but encoding set global. Thanks for nokogiri ^-^.
or cancel
Reproduce

Nokogiri::XML(open('http://www.cite-sciences.fr/rss/ressources/fr/faq_fr_20.xml').read, nil, 'UTF-8', 18543)

Nokogiri::XML::SyntaxError: Unsupported encoding UTF-85

This xml, not valid becouse <?xml version="1.0" encoding="UTF-85"> but encoding set global.

Thanks for nokogiri ^-^.

Comments
paranormal Mon Nov 02 00:50:22 -0800 2009 | link

This is important for me. Because of my program is web robot that detects encoding and merge to utf before nokogiri work.

I think, nokogiri must ignore tag encoding if they set in initialize.

Comment
This is important for me. Because of my program is web robot that detects encoding and merge to utf before nokogiri work. I think, nokogiri must ignore tag encoding if they set in initialize.
or cancel
tenderlove Mon Nov 09 17:47:05 -0800 2009 | link
Unfortunately the encoding declared inside the document is king.

In 1.4.1 you'll be able to alias UTF-85 to UTF-8 like so:

Nokogiri::EncodingHandler.alias('UTF-8', 'UTF-85')

And that will let you parse the strange encoding declaration.

b631d67
38fdb08
a073613
Comment
Unfortunately the encoding declared inside the document is king. In 1.4.1 you'll be able to alias UTF-85 to UTF-8 like so: Nokogiri::EncodingHandler.alias('UTF-8', 'UTF-85') And that will let you parse the strange encoding declaration. b631d6796f597aa6a55cdf20c203e9af1a96e0cf 38fdb08532bf07a2b46c9b08b07082f25578d85d a07361323e21746ab7a25b12d89ff0c30b784aef
or cancel
Parsed with GitHub Flavored Markdown
149.

2 votes

Anchor tags tightly wrapping another element generate unwanted whitespace on #to_xhtml
3 comments Created about 1 month ago by dasil003

Title

Body
This is just weird. Observe the reduced test cases: s = '<a>see</a>' n = Nokogiri::HTML::DocumentFragment.parse(s) n.to_xhtml => "<a>\n see\n</a>" to_html works: s = '<a>see</a>' n = Nokogiri::HTML::DocumentFragment.parse(s) n.to_html => "<a>see</a>" as does adding a text node in the source: s = '<a> see</a>' n = Nokogiri::HTML::DocumentFragment.parse(s) n.to_xhtml => "<a> see</a>"
or cancel
This is just weird. Observe the reduced test cases:
```
s = '<a>see</a>'
n = Nokogiri::HTML::DocumentFragment.parse(s)
n.to_xhtml
=> "<a>\n see\n</a>"
```
to_html works:
```
s = '<a>see</a>'
n = Nokogiri::HTML::DocumentFragment.parse(s)
n.to_html
=> "<a>see</a>"
```
as does adding a text node in the source:
```
s = '<a> see</a>'
n = Nokogiri::HTML::DocumentFragment.parse(s)
n.to_xhtml
=> "<a> see</a>"
```
Comments
dasil003 Wed Oct 14 14:55:15 -0700 2009 | link

BTW, I just discovered it also affects the OBJECT tag.

Comment
BTW, I just discovered it also affects the OBJECT tag.
or cancel
tenderlove Wed Oct 14 15:06:55 -0700 2009 | link
I don't think this is a bug. The default save options for to_xhtml say to format or "pretty print" the document. If your document contains space nodes, it will preserve them. If there are no blank nodes in the document, it will add them to make the output formatted.

If you don't want formatting, you can change the to_xhtml save options:

s = '<a>see</a>' n = Nokogiri::HTML::DocumentFragment.parse(s) puts n.to_xhtml(:save_with => Nokogiri::XML::Node::SaveOptions::AS_XHTML)
Comment
I don't think this is a bug. The default save options for to_xhtml say to format or "pretty print" the document. If your document contains space nodes, it will preserve them. If there are no blank nodes in the document, it will add them to make the output formatted. If you don't want formatting, you can change the to_xhtml save options: s = '<a>see</a>' n = Nokogiri::HTML::DocumentFragment.parse(s) puts n.to_xhtml(:save_with => Nokogiri::XML::Node::SaveOptions::AS_XHTML)
or cancel
dasil003 Wed Oct 14 16:43:06 -0700 2009 | link

ah thanks for the tip. Might be nice to document these somewhere. I noticed libxml2 also doesn't document them.

Comment
ah thanks for the tip. Might be nice to document these somewhere. I noticed libxml2 also doesn't document them.
or cancel

Parsed with GitHub Flavored Markdown
148.

0 votes

1.4.0
x

Segmentation Fault on modified re-raise
1 comment Created about 1 month ago by Empact

Title

Body
<pre> xml = Nokogiri::XML('<xml />') begin xml.xpath('http://') rescue Nokogiri::XML::XPath::SyntaxError => e raise e, "howdy" end </pre> results in: <pre> [BUG] Segmentation fault ruby 1.8.7 (2009-06-12 patchlevel 174) [i686-darwin10] Abort trap </pre> for <pre> --- warnings: [] libxml: loaded: 2.7.5 binding: extension compiled: 2.7.5 nokogiri: 1.3.3 </pre>
or cancel
```
 xml = Nokogiri::XML('<xml />')
begin
 xml.xpath('http://')
rescue Nokogiri::XML::XPath::SyntaxError => e
 raise e, "howdy"
end
```
results in:
```
[BUG] Segmentation fault
ruby 1.8.7 (2009-06-12 patchlevel 174) [i686-darwin10]

Abort trap
```
for
```
--- 
warnings: []

libxml: 
 loaded: 2.7.5
 binding: extension
 compiled: 2.7.5
nokogiri: 1.3.3
```
Comments

tenderlove Tue Oct 13 20:49:06 -0700 2009 | link

duplicating erorrs works. yay! closed by 33922d7

Comment
duplicating erorrs works. yay! closed by 33922d779346d8c8f6caaadf4995fba7738b52d0
or cancel

Parsed with GitHub Flavored Markdown
147.

0 votes

0 comments Created about 1 month ago by klancaster

Title

Body

or cancel

Comments

Parsed with GitHub Flavored Markdown
146.

0 votes

OS X 10.5 + 1.8.6 + 1.3.3 segfault
4 comments Created about 1 month ago by sporkmonger

Title

Body
This segfaults, if you look at it funny out of the corner of your eye: require "nokogiri" class TextHandler < Nokogiri::XML::SAX::Document def initialize @chunks = [] end attr_reader :chunks def cdata_block(string) characters(string) end def characters(string) @chunks << string.strip if string.strip != "" end end th = TextHandler.new parser = Nokogiri::XML::SAX::Parser.new(th) parser.parse(<<-XML) <?xml version="1.0" encoding="utf-8"?> <root> <stuff> one </stuff> <stuff> two </stuff> </root> XML I was able to duplicate consistently for awhile, but I uninstalled and reinstalled nokogiri a few times, and now it works. It would reach the end of the document before segfaulting. The end_document event would fire, and then it would segfault shortly thereafter.
or cancel
This segfaults, if you look at it funny out of the corner of your eye:
```
require "nokogiri"

class TextHandler < Nokogiri::XML::SAX::Document
 def initialize
 @chunks = []
 end

 attr_reader :chunks

 def cdata_block(string)
 characters(string)
 end

 def characters(string)
 @chunks << string.strip if string.strip != ""
 end
end
th = TextHandler.new
parser = Nokogiri::XML::SAX::Parser.new(th)
parser.parse(<<-XML)
<?xml version="1.0" encoding="utf-8"?>
<root>
 <stuff>
 one
 </stuff>
 <stuff>
 two
 </stuff>
</root>
XML
```
I was able to duplicate consistently for awhile, but I uninstalled and reinstalled nokogiri a few times, and now it works. It would reach the end of the document before segfaulting. The end_document event would fire, and then it would segfault shortly thereafter.
Comments

jmhodges Sat Oct 10 17:14:55 -0700 2009 | link

What version of libxml2?

Comment
What version of libxml2?
or cancel

sporkmonger Sat Oct 10 21:10:51 -0700 2009 | link

Looks like 2.7.3.

Comment
Looks like 2.7.3.
or cancel

tenderlove Mon Oct 12 17:54:21 -0700 2009 | link

If you were on libxml2, 2.6.16, then I wouldn't be surprised. That version was very old an unstable.

I can't repro this (even with thousands of iterations), so I will assume it's a bug with 2.6.16. I am going to close this, but if you are able to repro with 2.7.3, please reopen this ticket. Thanks!

Comment
If you were on libxml2, 2.6.16, then I wouldn't be surprised. That version was very old an unstable. I can't repro this (even with thousands of iterations), so I will assume it's a bug with 2.6.16. I am going to close this, but if you are able to repro with 2.7.3, please reopen this ticket. Thanks!
or cancel

sporkmonger Mon Oct 12 17:59:47 -0700 2009 | link

Shouldn't require thousands of iterations, it's a happens-every-time kind of bug. However, I may have been wrong about the version of nokogiri I was using. It might have been edge.

Comment
Shouldn't require thousands of iterations, it's a happens-every-time kind of bug. However, I may have been wrong about the version of nokogiri I was using. It might have been edge.
or cancel

Parsed with GitHub Flavored Markdown
145.

0 votes

Adding a Document to a Node causes segfault on program exit
1 comment Created about 1 month ago by david

Title

Body
$ ruby -rubygems -rnokogiri -e 'Nokogiri::XML("<yo />").root << Nokogiri::XML::Document.new' <main>: [BUG] Segmentation fault ruby 1.9.1p243 (2009-07-16 revision 24175) [i486-linux] -- control frame ---------- c:0001 p:0000 s:0002 b:0002 l:0011a4 d:0011a4 TOP --------------------------- -- Ruby level backtrace information----------------------------------------- -- C level backtrace information ------------------------------------------- 0xb76cd6e9 /usr/lib/libruby-1.9.1.so.1.9(rb_vm_bugreport+0x69) [0xb76cd6e9] 0xb75e907f /usr/lib/libruby-1.9.1.so.1.9 [0xb75e907f] 0xb75e911a /usr/lib/libruby-1.9.1.so.1.9(rb_bug+0x3a) [0xb75e911a] 0xb7674fa4 /usr/lib/libruby-1.9.1.so.1.9 [0xb7674fa4] 0xb7768410 [0xb7768410] 0xb767b817 /usr/lib/libruby-1.9.1.so.1.9(st_foreach+0x17) [0xb767b817] 0xb72d0aa9 /var/lib/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.so [0xb72d0aa9] 0xb75f954d /usr/lib/libruby-1.9.1.so.1.9 [0xb75f954d] 0xb75f96e4 /usr/lib/libruby-1.9.1.so.1.9 [0xb75f96e4] 0xb75f98dc /usr/lib/libruby-1.9.1.so.1.9(rb_gc_call_finalizer_at_exit+0x17c) [0xb75f98dc] 0xb75eb0ee /usr/lib/libruby-1.9.1.so.1.9 [0xb75eb0ee] 0xb75ec436 /usr/lib/libruby-1.9.1.so.1.9(ruby_cleanup+0x116) [0xb75ec436] 0xb75ec5ee /usr/lib/libruby-1.9.1.so.1.9(ruby_run_node+0x5e) [0xb75ec5ee] 0x80487e8 ruby(main+0x68) [0x80487e8] 0xb73dfb56 /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe6) [0xb73dfb56] 0x80486e1 ruby [0x80486e1] [NOTE] You may encounter a bug of Ruby interpreter. Bug reports are welcome. For details: http://www.ruby-lang.org/bugreport.html ^CAborted (core dumped) -- SNIP -- This is ruby 1.9.1, but the same thing happened to me on 1.8.7. The reason why I think this is significant is because I mistakenly was adding a Document to a Node in my code, and that kept failing with me understanding why, so the same thing may happen to other people. Thanks.
or cancel

$ ruby -rubygems -rnokogiri -e 'Nokogiri::XML("").root << Nokogiri::XML::Document.new'

: [BUG] Segmentation fault
ruby 1.9.1p243 (2009-07-16 revision 24175) [i486-linux]

-- control frame ----------

c:0001 p:0000 s:0002 b:0002 l:0011a4 d:0011a4 TOP

-- Ruby level backtrace information-----------------------------------------

-- C level backtrace information ------------------------------------------- 0xb76cd6e9 /usr/lib/libruby-1.9.1.so.1.9(rb_vm_bugreport+0x69) [0xb76cd6e9]
0xb75e907f /usr/lib/libruby-1.9.1.so.1.9 [0xb75e907f]
0xb75e911a /usr/lib/libruby-1.9.1.so.1.9(rb_bug+0x3a) [0xb75e911a]
0xb7674fa4 /usr/lib/libruby-1.9.1.so.1.9 [0xb7674fa4]
0xb7768410 [0xb7768410]
0xb767b817 /usr/lib/libruby-1.9.1.so.1.9(st_foreach+0x17) [0xb767b817]
0xb72d0aa9 /var/lib/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.so [0xb72d0aa9]
0xb75f954d /usr/lib/libruby-1.9.1.so.1.9 [0xb75f954d]
0xb75f96e4 /usr/lib/libruby-1.9.1.so.1.9 [0xb75f96e4]
0xb75f98dc /usr/lib/libruby-1.9.1.so.1.9(rb_gc_call_finalizer_at_exit+0x17c) [0xb75f98dc]
0xb75eb0ee /usr/lib/libruby-1.9.1.so.1.9 [0xb75eb0ee]
0xb75ec436 /usr/lib/libruby-1.9.1.so.1.9(ruby_cleanup+0x116) [0xb75ec436]
0xb75ec5ee /usr/lib/libruby-1.9.1.so.1.9(ruby_run_node+0x5e) [0xb75ec5ee]
0x80487e8 ruby(main+0x68) [0x80487e8]
0xb73dfb56 /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe6) [0xb73dfb56]
0x80486e1 ruby [0x80486e1]

[NOTE] You may encounter a bug of Ruby interpreter. Bug reports are welcome.
For details: http://www.ruby-lang.org/bugreport.html

^CAborted (core dumped)

-- SNIP --

This is ruby 1.9.1, but the same thing happened to me on 1.8.7.

The reason why I think this is significant is because I mistakenly was adding a Document to a Node in my code, and that kept failing with me understanding why, so the same thing may happen to other people.

Thanks.

Comments

tenderlove Tue Oct 13 11:17:11 -0700 2009 | link

raising an exception if someone tries to reparent a *::Document. closed by c557764

Comment
raising an exception if someone tries to reparent a *::Document. closed by c557764f88d91d025c63b56b3e63a1c0f9302e69
or cancel

Parsed with GitHub Flavored Markdown
144.

0 votes

Segmentation fault
7 comments Created about 1 month ago by mawiza

Title

Body
Hi guys, I get the following error message while downloading an XML file and opening it using nokogiri: res = Net::HTTP.post_form(URI.parse(....), {...}) doc = Nokogiri::XML(Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("& lt;", "<").gsub("& gt;", ">")) I have installed the latest nightly on OS X 10.5.6. /Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3.20091004000018/lib/nokogiri/xml/document.rb:33: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0] Abort trap I have also tried to split the constructor calls: doc = Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("&lt ;", "<").gsub("&gt ;", ">") I have about 10 different XML files, and it crashes randomly on the different files, so I can't say that it's one specific file. The XML files vary in size from 3mb to 150mb. The files are very basic XML: <?xml version="1.0" encoding="utf-8"?> <string>....</string> where the string element contains escaped XML. Unfortunately the XML files are data we receive from an external vendor, so not really anything we can do about that. I can try to normalize the data using gsub and then use nokogiri on the xmlified data. Regards William
or cancel
Hi guys,

I get the following error message while downloading an XML file and opening it using nokogiri:
```
res = Net::HTTP.post_form(URI.parse(....), {...})
doc = Nokogiri::XML(Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("& lt;", "<").gsub("& gt;", ">"))
```
I have installed the latest nightly on OS X 10.5.6.
```
/Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3.20091004000018/lib/nokogiri/xml/document.rb:33: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0]

Abort trap
```
I have also tried to split the constructor calls:
```
doc = Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("&lt ;", "<").gsub("&gt ;", ">")
```
I have about 10 different XML files, and it crashes randomly on the different files, so I can't say that it's one specific file.

The XML files vary in size from 3mb to 150mb.

The files are very basic XML:
```
<?xml version="1.0" encoding="utf-8"?>
<string>....</string>
```
where the string element contains escaped XML. Unfortunately the XML files are data we receive from an external vendor, so not really anything we can do about that.

I can try to normalize the data using gsub and then use nokogiri on the xmlified data.

Regards
William
Comments

tenderlove Tue Oct 06 08:49:35 -0700 2009 | link

Can you run "nokogiri -v" and add the output to this ticket?

Comment
Can you run "nokogiri -v" and add the output to this ticket?
or cancel

mawiza Tue Oct 06 08:49:50 -0700 2009 | link

It seems that that did the trick. I have not had any segfaults since.

Comment
It seems that that did the trick. I have not had any segfaults since.
or cancel

mawiza Tue Oct 06 08:53:21 -0700 2009 | link

libxml:
loaded: 2.7.3 binding: extension compiled: 2.7.3 nokogiri: 1.3.3

I am still busy testing it, but so far so good. I will keep you posted and close this ticket when I am sure that it is not an issue anymore.

Comment
libxml: loaded: 2.7.3 binding: extension compiled: 2.7.3 nokogiri: 1.3.3 I am still busy testing it, but so far so good. I will keep you posted and close this ticket when I am sure that it is not an issue anymore.
or cancel

tenderlove Tue Oct 06 09:05:55 -0700 2009 | link

Okay, sounds good. You might also want to try upgrading libxml2. The latest libxml2 is 2.7.5 and I know they've packed in a bunch of bug fixes. If that doesn't do the trick, would you mind sending us a sample of the XML you're using to make it crash? It shouldn't SEGV under any circumstances. :-)

Comment
Okay, sounds good. You might also want to try upgrading libxml2. The latest libxml2 is 2.7.5 and I know they've packed in a bunch of bug fixes. If that doesn't do the trick, would you mind sending us a sample of the XML you're using to make it crash? It shouldn't SEGV under any circumstances. :-)
or cancel

mawiza Tue Oct 06 09:44:52 -0700 2009 | link

:) I will do that. Thanks for the advice.

Comment
:) I will do that. Thanks for the advice.
or cancel

tenderlove Tue Oct 13 13:49:22 -0700 2009 | link

Any updates on this?

Comment
Any updates on this?
or cancel

tenderlove Thu Oct 15 09:16:59 -0700 2009 | link

I'm closing this since there have been no updates. Please reopen if you're still having problems! Thanks!

Comment
I'm closing this since there have been no updates. Please reopen if you're still having problems! Thanks!
or cancel

Parsed with GitHub Flavored Markdown
143.

0 votes

test_empty_string_returns_empty_doc is empty
2 comments Created about 1 month ago by djsun

Title

Body
There is no test in the code below... # test_document.rb def test_empty_string_returns_empty_doc doc = Nokogiri::HTML('') end
or cancel
There is no test in the code below...
```
# test_document.rb
def test_empty_string_returns_empty_doc
  doc = Nokogiri::HTML('')
end
```
Comments

tenderlove Sun Oct 04 20:32:52 -0700 2009 | link

Thanks for the heads up! http://ihighfive.com/

Comment
Thanks for the heads up! http://ihighfive.com/
or cancel

djsun Mon Oct 05 05:54:59 -0700 2009 | link

You are very whalecome: http://ihighfive.com/whale-high-five.php

Comment
You are very whalecome: http://ihighfive.com/whale-high-five.php
or cancel

Parsed with GitHub Flavored Markdown
142.

0 votes

WARNING: Nokogiri was built against LibXML version 2.7.3, but has dynamically loaded 2.6.32
3 comments Created about 1 month ago by ponny

Title

Body
Any idea how do I make it load the correct version? I'm on Ubuntu. I've followed the directions here: http://wiki.github.com/tenderlove/nokogiri/use-libxml-from-source but it's just not flying. Thanks in advance.
or cancel

Any idea how do I make it load the correct version? I'm on Ubuntu. I've followed the directions here: http://wiki.github.com/tenderlove/nokogiri/use-libxml-from-source but it's just not flying.

Thanks in advance.

Comments

tenderlove Sun Oct 04 20:41:02 -0700 2009 | link

have you added '/usr/local/lib' to your ld.so.conf file?

Comment
have you added '/usr/local/lib' to your ld.so.conf file?
or cancel

tenderlove Tue Oct 13 13:45:05 -0700 2009 | link

hello?

Comment
hello?
or cancel

tenderlove Wed Oct 14 18:49:33 -0700 2009 | link

Closing. If you're still having troubles with this, please send an email to the mailing list:

http://groups.google.com/group/nokogiri-talk

Comment
Closing. If you're still having troubles with this, please send an email to the mailing list: http://groups.google.com/group/nokogiri-talk
or cancel

Parsed with GitHub Flavored Markdown
141.

0 votes

[PATCH] adding Builder#<< for appending raw strings
4 comments Created about 1 month ago by dudleyf

Title

Body
Here's a tiny patch implementing the functionality talked about here[0]. [0] http://rubyforge.org/pipermail/nokogiri-talk/2009-March/000224.html <pre> diff --git a/lib/nokogiri/xml/builder.rb b/lib/nokogiri/xml/builder.rb index 89cd63a..5cdcafd 100644 --- a/lib/nokogiri/xml/builder.rb +++ b/lib/nokogiri/xml/builder.rb @@ -277,6 +277,12 @@ module Nokogiri @doc.to_xml end + ### + # Append the given raw XML +string+ to the document + def << string + @doc.fragment(string).children.each { |x| insert(x) } + end + def method_missing method, *args, &block # :nodoc: if @context && @context.respond_to?(method) @context.send(method, *args, &block) diff --git a/test/xml/test_builder.rb b/test/xml/test_builder.rb index d4a6e26..12b1f86 100644 --- a/test/xml/test_builder.rb +++ b/test/xml/test_builder.rb @@ -117,6 +117,26 @@ module Nokogiri assert_equal 'hello', builder.doc.at('baz').content end + def test_raw_append + builder = Nokogiri::XML::Builder.new do |xml| + xml.root do + xml << '<foo>hello</foo>' + end + end + + assert_equal 'hello', builder.doc.at('//root/foo').content + end + + def test_raw_append_with_instance_eval + builder = Nokogiri::XML::Builder.new do + root do + self << '<foo>hello</foo>' + end + end + + assert_equal 'hello', builder.doc.at('//root/foo').content + end + def test_cdata builder = Nokogiri::XML::Builder.new do root { -- 1.6.4.3 </pre>
or cancel
Here's a tiny patch implementing the functionality talked about here[0].

[0] http://rubyforge.org/pipermail/nokogiri-talk/2009-March/000224.html
```
diff --git a/lib/nokogiri/xml/builder.rb b/lib/nokogiri/xml/builder.rb
index 89cd63a..5cdcafd 100644
--- a/lib/nokogiri/xml/builder.rb
+++ b/lib/nokogiri/xml/builder.rb
@@ -277,6 +277,12 @@ module Nokogiri
 @doc.to_xml
 end
 
+ ###
+ # Append the given raw XML +string+ to the document
+ def << string
+ @doc.fragment(string).children.each { |x| insert(x) }
+ end
+
 def method_missing method, *args, &block # :nodoc:
 if @context && @context.respond_to?(method)
 @context.send(method, *args, &block)
diff --git a/test/xml/test_builder.rb b/test/xml/test_builder.rb
index d4a6e26..12b1f86 100644
--- a/test/xml/test_builder.rb
+++ b/test/xml/test_builder.rb
@@ -117,6 +117,26 @@ module Nokogiri
 assert_equal 'hello', builder.doc.at('baz').content
 end
 
+ def test_raw_append
+ builder = Nokogiri::XML::Builder.new do |xml|
+ xml.root do
+ xml << 'hello'
+ end
+ end
+
+ assert_equal 'hello', builder.doc.at('//root/foo').content
+ end
+
+ def test_raw_append_with_instance_eval
+ builder = Nokogiri::XML::Builder.new do
+ root do
+ self << 'hello'
+ end
+ end
+
+ assert_equal 'hello', builder.doc.at('//root/foo').content
+ end
+
 def test_cdata
 builder = Nokogiri::XML::Builder.new do
 root {
-- 
1.6.4.3
```
Comments
tenderlove Sun Oct 04 20:40:05 -0700 2009 | link
I've applied the patch, but next time please make sure the tests pass. After applying the patch, I got these errors:

1) Error: test_raw_append(Nokogiri::XML::TestBuilder): NoMethodError: undefined method `content' for nil:NilClass test/xml/test_builder.rb:127:in `test_raw_append' 2) Error: test_raw_append_with_instance_eval(Nokogiri::XML::TestBuilder): NoMethodError: undefined method `content' for nil:NilClass test/xml/test_builder.rb:137:in `test_raw_append_with_instance_eval'
Comment
I've applied the patch, but next time please make sure the tests pass. After applying the patch, I got these errors: 1) Error: test_raw_append(Nokogiri::XML::TestBuilder): NoMethodError: undefined method `content' for nil:NilClass test/xml/test_builder.rb:127:in `test_raw_append' 2) Error: test_raw_append_with_instance_eval(Nokogiri::XML::TestBuilder): NoMethodError: undefined method `content' for nil:NilClass test/xml/test_builder.rb:137:in `test_raw_append_with_instance_eval'
or cancel
tenderlove Sun Oct 04 20:40:29 -0700 2009 | link

XML Builder can append raw strings. closed by 98b10d2

Comment
XML Builder can append raw strings. closed by 98b10d222c7ac246e2e8267c96ad1d38942e52ad
or cancel

dudleyf Mon Oct 05 06:34:54 -0700 2009 | link

Yeah, I pasted the wrong patch. Sorry about that.

Comment
Yeah, I pasted the wrong patch. Sorry about that.
or cancel

tenderlove Mon Oct 05 08:38:30 -0700 2009 | link

No problem! :-)

Comment
No problem! :-)
or cancel

Parsed with GitHub Flavored Markdown
140.

0 votes

REE
x

tcmalloc error
4 comments Created about 1 month ago by cjbottaro

Title

Body
Nokogiri::HTML(data) src/tcmalloc.cc:186] Attempt to free invalid pointer: 0x20e030 Nokogiri::VERSION => "1.3.2" Nokogiri::LIBXML_VERSION => "2.6.32" ruby -v => ruby 1.8.6 (2008-08-11 patchlevel 287) [i686-darwin9.7.0] Ruby Enterprise Edition 20090610 Where 'data' is... http://gist.github.com/199496 Thanks for the help!
or cancel
Nokogiri::HTML(data)
src/tcmalloc.cc:186] Attempt to free invalid pointer: 0x20e030

Nokogiri::VERSION => "1.3.2"
Nokogiri::LIBXML_VERSION => "2.6.32"
ruby -v => ruby 1.8.6 (2008-08-11 patchlevel 287) [i686-darwin9.7.0] Ruby Enterprise Edition 20090610

Where 'data' is... http://gist.github.com/199496

Thanks for the help!

Comments
mperham Fri Oct 02 08:51:33 -0700 2009 | link

Works for me with libxml 2.7.3 and nokogiri 1.3.3

Comment
Works for me with libxml 2.7.3 and nokogiri 1.3.3
or cancel

cjbottaro Fri Oct 02 16:40:22 -0700 2009 | link

Still crashes for me using libxml2 2.7.5 and nokogiri 1.3.3 and ree-1.8.6-20090610. It does however work with normal MRI: ruby 1.8.6 (2009-08-04 patchlevel 383) [i686-darwin9.8.0]

Comment
Still crashes for me using libxml2 2.7.5 and nokogiri 1.3.3 and ree-1.8.6-20090610. It does however work with normal MRI: ruby 1.8.6 (2009-08-04 patchlevel 383) [i686-darwin9.8.0]
or cancel
tenderlove Sun Oct 04 21:03:28 -0700 2009 | link
This isn't crashing for me. I'm using:

[apatterson@higgins nokogiri (master)]$ ruby -v ruby 1.8.6 (2008-08-11 patchlevel 287) [i686-darwin10.0.0] Ruby Enterprise Edition 20090610 [apatterson@higgins nokogiri (master)]$ ruby -I lib bin/nokogiri -v --- nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.7.5 loaded: 2.7.5 binding: extension [apatterson@higgins nokogiri (master)]$

Can you try with the nightly? To install the nightly build, do this:

$ sudo gem install nokogiri -s http://tenderlovemaking.com
Comment
This isn't crashing for me. I'm using: [apatterson@higgins nokogiri (master)]$ ruby -v ruby 1.8.6 (2008-08-11 patchlevel 287) [i686-darwin10.0.0] Ruby Enterprise Edition 20090610 [apatterson@higgins nokogiri (master)]$ ruby -I lib bin/nokogiri -v --- nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.7.5 loaded: 2.7.5 binding: extension [apatterson@higgins nokogiri (master)]$ Can you try with the nightly? To install the nightly build, do this: $ sudo gem install nokogiri -s http://tenderlovemaking.com
or cancel
tenderlove Tue Oct 13 13:50:41 -0700 2009 | link

I can't reproduce this. Please reopen if the problem persists. I need more details to fix this if there is a problem. Thanks!

Comment
I can't reproduce this. Please reopen if the problem persists. I need more details to fix this if there is a problem. Thanks!
or cancel

Parsed with GitHub Flavored Markdown
139.

0 votes

REE
x

extconf.rb have_func() always fails under Ruby Enterprise build system
7 comments Created about 1 month ago by flavorjones

Title

Body
ruby-enterprise-1.8.6-20090610: checking for xmlRelaxNGSetParserStructuredErrors()... no checking for xmlRelaxNGSetParserStructuredErrors()... no checking for xmlRelaxNGSetValidStructuredErrors()... no checking for xmlSchemaSetValidStructuredErrors()... no checking for xmlSchemaSetParserStructuredErrors()... no
or cancel
ruby-enterprise-1.8.6-20090610:
```
checking for xmlRelaxNGSetParserStructuredErrors()... no
checking for xmlRelaxNGSetParserStructuredErrors()... no
checking for xmlRelaxNGSetValidStructuredErrors()... no
checking for xmlSchemaSetValidStructuredErrors()... no
checking for xmlSchemaSetParserStructuredErrors()... no
```
Comments
flavorjones Wed Sep 30 22:58:10 -0700 2009 | link
root cause:

"gcc -o conftest -I/usr/include/libxml2 -I/usr/include -I. -I/home/mike/builds/ruby-enterprise-1.8.6-20090610-install/lib/ruby/1.8/i686-linux -I/home/mike/code/nokogiri/ext/nokogiri -I/usr/include/libxml2 -I/usr/include -I. -I/home/mike/builds/ruby-enterprise-1.8.6-20090610-install/lib/ruby/1.8/i686-linux -I/home/mike/code/nokogiri/ext/nokogiri -g -O2 -g -DXP_UNIX -O3 -Wall -Wcast-qual -Wwrite-strings -Wconversion -Wmissing-noreturn -Winline conftest.c -L/opt/local/lib -Wl,-R/opt/local/lib -L. -rdynamic -Wl,-export-dynamic -lexslt -lxslt -lxml2 -lruby-static -lexslt -lxslt -lxml2 -ldl -lcrypt -lm -lc" conftest.c: In function ‘t’: conftest.c:3: warning: implicit declaration of function ‘xmlRelaxNGSetParserStructuredErrors’ /usr/bin/ld: cannot find -lruby-static collect2: ld returned 1 exit status checked program was: /* begin */ 1: /*top*/ 2: int main() { return 0; } 3: int t() { xmlRelaxNGSetParserStructuredErrors(); return 0; } /* end */
Comment
root cause: "gcc -o conftest -I/usr/include/libxml2 -I/usr/include -I. -I/home/mike/builds/ruby-enterprise-1.8.6-20090610-install/lib/ruby/1.8/i686-linux -I/home/mike/code/nokogiri/ext/nokogiri -I/usr/include/libxml2 -I/usr/include -I. -I/home/mike/builds/ruby-enterprise-1.8.6-20090610-install/lib/ruby/1.8/i686-linux -I/home/mike/code/nokogiri/ext/nokogiri -g -O2 -g -DXP_UNIX -O3 -Wall -Wcast-qual -Wwrite-strings -Wconversion -Wmissing-noreturn -Winline conftest.c -L/opt/local/lib -Wl,-R/opt/local/lib -L. -rdynamic -Wl,-export-dynamic -lexslt -lxslt -lxml2 -lruby-static -lexslt -lxslt -lxml2 -ldl -lcrypt -lm -lc" conftest.c: In function ‘t’: conftest.c:3: warning: implicit declaration of function ‘xmlRelaxNGSetParserStructuredErrors’ /usr/bin/ld: cannot find -lruby-static collect2: ld returned 1 exit status checked program was: /* begin */ 1: /*top*/ 2: int main() { return 0; } 3: int t() { xmlRelaxNGSetParserStructuredErrors(); return 0; } /* end */
or cancel
mreinsch Wed Sep 30 23:01:45 -0700 2009 | link
suggested fix:

diff --git a/ext/nokogiri/extconf.rb b/ext/nokogiri/extconf.rb index 7c21e7d..b77552d 100644 --- a/ext/nokogiri/extconf.rb +++ b/ext/nokogiri/extconf.rb @@ -129,7 +129,7 @@ unless find_library('exslt', 'exsltFuncRegister', *LIB_DIRS) abort "libxslt is missing. try 'port install libxslt' or 'yum install libxslt-devel'" end -def nokogiri_link_command ldflags, opt='', libpath=$LIBPATH +def nokogiri_link_command ldflags, opt='', libpath=$DEFLIBPATH|$LIBPATH old_link_command ldflags, opt, libpath end
Comment
suggested fix: diff --git a/ext/nokogiri/extconf.rb b/ext/nokogiri/extconf.rb index 7c21e7d..b77552d 100644 --- a/ext/nokogiri/extconf.rb +++ b/ext/nokogiri/extconf.rb @@ -129,7 +129,7 @@ unless find_library('exslt', 'exsltFuncRegister', *LIB_DIRS) abort "libxslt is missing. try 'port install libxslt' or 'yum install libxslt-devel'" end -def nokogiri_link_command ldflags, opt='', libpath=$LIBPATH +def nokogiri_link_command ldflags, opt='', libpath=$DEFLIBPATH|$LIBPATH old_link_command ldflags, opt, libpath end
or cancel
flavorjones Wed Sep 30 23:03:20 -0700 2009 | link
proposed fix from Michael Reinsch:

- def nokogiri_link_command ldflags, opt='', libpath=$LIBPATH + def nokogiri_link_command ldflags, opt='', libpath=$DEFLIBPATH|$LIBPATH

which appears to work for me. Aaron, thoughts?
Comment
proposed fix from Michael Reinsch: - def nokogiri_link_command ldflags, opt='', libpath=$LIBPATH + def nokogiri_link_command ldflags, opt='', libpath=$DEFLIBPATH|$LIBPATH which appears to work for me. Aaron, thoughts?
or cancel
mreinsch Wed Sep 30 23:12:41 -0700 2009 | link

(which is what is used in mkmf.rb, link_command for ruby enterprise)

Comment
(which is what is used in mkmf.rb, link_command for ruby enterprise)
or cancel

tenderlove Sun Oct 04 20:53:33 -0700 2009 | link

Ugh. This is a PITA, but basically we can't make everyone happy. If I add this patch, I may as well remove that "nokogiri_link_command" stuff all together. Let me try to explain why:

Someone has ruby installed in /usr/lib, they also have libxml2 installed in /usr/lib. They've installed a newer version of libxml2 in /usr/local. We try to be nice and search /opt/local/lib in addition to /usr/local/lib before falling back to /usr/lib. Unfortunately, if the custom directory (/opt/local or /usr/lib) isn't supplied to dir_config(), it won't search that path. We can only supply one directory. If mkmf doesn't find it in that directory, then it falls back to /usr/lib.

That means we either get /opt/local/lib or /usr/lib, unless the user intervenes with a --with-xml-lib=/whatever --with-xml-include=/whataver, or we use my Super Hack® code. Unfortunately my Super Hack® screws over people with custom ruby installs because it will never find the ruby-static library.

I'm going to apply this fix (and by apply, I mean remove my custom code). I'd rather it "just work" for people with custom ruby installs. People with custom libxml2 installs can use the command line arguments.

Comment
Ugh. This is a PITA, but basically we can't make everyone happy. If I add this patch, I may as well remove that "nokogiri_link_command" stuff all together. Let me try to explain why: Someone has ruby installed in /usr/lib, they also have libxml2 installed in /usr/lib. They've installed a newer version of libxml2 in /usr/local. We try to be nice and search /opt/local/lib in addition to /usr/local/lib before falling back to /usr/lib. Unfortunately, if the custom directory (/opt/local or /usr/lib) isn't supplied to dir_config(), it won't search that path. We can only supply one directory. If mkmf doesn't find it in that directory, then it falls back to /usr/lib. That means we either get /opt/local/lib or /usr/lib, unless the user intervenes with a --with-xml-lib=/whatever --with-xml-include=/whataver, or we use my Super Hack® code. Unfortunately my Super Hack® screws over people with custom ruby installs because it will never find the ruby-static library. I'm going to apply this fix (and by apply, I mean remove my custom code). I'd rather it "just work" for people with custom ruby installs. People with custom libxml2 installs can use the command line arguments.
or cancel

tenderlove Sun Oct 04 20:54:09 -0700 2009 | link

removing my Super Hack® closed by fbe7217

Comment
removing my Super Hack® closed by fbe7217056217b27d41f0eb5d2bdc3e91f0d8add
or cancel

flavorjones Mon Oct 05 05:33:42 -0700 2009 | link

Aaron, the --with-xml-lib and --with-xml-include options do not appear to affect have_func(), since it consistently uses the wrong header files.

Comment
Aaron, the --with-xml-lib and --with-xml-include options do not appear to affect have_func(), since it consistently uses the wrong header files.
or cancel

Parsed with GitHub Flavored Markdown
138.

1 vote

1.4.0
x

Segfaults / memory corruption
16 comments Created about 1 month ago by retoo

Title

Body
Hi I've got some code here which replaces some parts of a html document with new content. This works pretty well in 99 out of 100 times, but in certain situation nokogiri segfaults. Yesterday we were able to capture a 'crashing' situation, this is: a) the original document b) a start and end dom-id and c) the new content. Based on this I was able to create a simpel example[0] which shows this bug. I suspected that some char / element causes the crash, but during my first inspection it turned out that it must be a combination of some weird circumstances. For example when I reduce the size of c) (new content), the crashes happen less often. The same happens when I reduce the range between b) start- and end-id. It also appears to crash more often when rails is loaded as well (although I was now able to get it 'reliably' crashing wihtout rails). I was only able to reproduce it on linux, it doesn't crash with the same ruby/libxml/nokogiri version on OS X. I'm pretty out of ideas, I hope somebody can checkout my example[0] and try to find the root cause. My current theory is that nokogiri/libxml somewhere corrupts memory, either on the stack or heap and that later somebody chokes on that corruption. I also think it is somehow related to the use of the french é, written as é, when I remove these chars the crash can't be reproduced. You can see the output plus stacktrace of a run on my machine on: https://gist.github.com/f37eda8131e39fac9dd4 Thanks for the help! Cheers Reto [0] http://github.com/retoo/nokogiri-bug
or cancel
Hi

I've got some code here which replaces some parts of a html document with new content. This works pretty well in 99 out of 100 times, but in certain situation nokogiri segfaults. Yesterday we were able to capture a 'crashing' situation, this is: a) the original document b) a start and end dom-id and c) the new content. Based on this I was able to create a simpel example[0] which shows this bug.

I suspected that some char / element causes the crash, but during my first inspection it turned out that it must be a combination of some weird circumstances. For example when I reduce the size of c) (new content), the crashes happen less often. The same happens when I reduce the range between b) start- and end-id. It also appears to crash more often when rails is loaded as well (although I was now able to get it 'reliably' crashing wihtout rails). I was only able to reproduce it on linux, it doesn't crash with the same ruby/libxml/nokogiri version on OS X.

I'm pretty out of ideas, I hope somebody can checkout my example[0] and try to find the root cause. My current theory is that nokogiri/libxml somewhere corrupts memory, either on the stack or heap and that later somebody chokes on that corruption. I also think it is somehow related to the use of the french é, written as é, when I remove these chars the crash can't be reproduced.

You can see the output plus stacktrace of a run on my machine on:
https://gist.github.com/f37eda8131e39fac9dd4

Thanks for the help!
Cheers
Reto

[0] http://github.com/retoo/nokogiri-bug

Comments
flavorjones Mon Sep 28 23:39:13 -0700 2009 | link

will look at this ASAP.

Comment
will look at this ASAP.
or cancel

flavorjones Wed Sep 30 23:04:24 -0700 2009 | link

i've got a repro case and a possible fix.

Comment
i've got a repro case and a possible fix.
or cancel
tenderlove Wed Sep 30 23:05:10 -0700 2009 | link
Can you try out the nightly build? I believe this may already be fixed.

To get the nightly, do this:

$ sudo gem install nokogiri -s http://tenderlovemaking.com
Comment
Can you try out the nightly build? I believe this may already be fixed. To get the nightly, do this: $ sudo gem install nokogiri -s http://tenderlovemaking.com
or cancel
retoo Thu Oct 01 01:11:07 -0700 2009 | link

I tested it against nokogiri/master and the nightly, the crash still happens.

But I increased the 'input' length in the test script, the crash should now happen more often (almost every time now on three of my machines).

git://github.com/retoo/nokogiri-bug.git

Thanks!

Comment
I tested it against nokogiri/master and the nightly, the crash still happens. But I increased the 'input' length in the test script, the crash should now happen more often (almost every time now on three of my machines). git://github.com/retoo/nokogiri-bug.git Thanks!
or cancel

flavorjones Thu Oct 01 07:48:29 -0700 2009 | link

retoo - the fix tenderlove is referring to does not apply to your/my situation, sorry about that. We have a repro case and a potential fix, and we are working on it. Thanks!

Comment
retoo - the fix tenderlove is referring to does not apply to your/my situation, sorry about that. We have a repro case and a potential fix, and we are working on it. Thanks!
or cancel

tenderlove Sat Oct 03 22:05:16 -0700 2009 | link

@retoo I think we've got a fix in. Can you pull the latest nightly and see if that fixes your crash?

Comment
@retoo I think we've got a fix in. Can you pull the latest nightly and see if that fixes your crash?
or cancel

retoo Mon Oct 05 00:34:47 -0700 2009 | link

It doesn't crash anymore here. Bisect says that it has been fixed in c753c8d.

Thank you guys!

Reto

Comment
It doesn't crash anymore here. Bisect says that it has been fixed in c753c8d. Thank you guys! Reto
or cancel

retoo Mon Oct 05 00:38:07 -0700 2009 | link

Closing...

Comment
Closing...
or cancel

mawiza Tue Oct 06 02:14:54 -0700 2009 | link

Hi guys,

I might have the same issue, I get the following error message while downloading a an xml file and opening it using nokogiri:

res = Net::HTTP.post_form(URI.parse(....), {...})
doc = Nokogiri::XML(Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("& lt;", "<").gsub("& gt;", ">"))

I have installed the latest nightly on OS X 10.5.6.

/Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3.20091004000018/lib/nokogiri/xml/document.rb:33: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0]

Abort trap

Regards
William

Comment
Hi guys, I might have the same issue, I get the following error message while downloading a an xml file and opening it using nokogiri: res = Net::HTTP.post_form(URI.parse(....), {...}) doc = Nokogiri::XML(Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("& lt;", "<").gsub("& gt;", ">")) I have installed the latest nightly on OS X 10.5.6. /Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3.20091004000018/lib/nokogiri/xml/document.rb:33: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0] Abort trap Regards William
or cancel
retoo Tue Oct 06 02:34:58 -0700 2009 | link
Hmm, probably not, 1. It didn't crash on os x in my case, 2. it crashed when I tried to replace() some parts of it, not during any xpath/constructor calls.

Nonetheless, I would report this (perhaps in a new, unused fresh ticket :D). Does this crash everytime? If yes, seperate the two constructor calls like:

step 1 = XML(input).xpath.to_s.gub puts "I survived so far step2 = XML(step1) puts "should never reach this point"

It would be very helpful if you could give the output of res.body. Or even better, the minimum of res.body which still triggers the crash.

Cheers.
Comment
Hmm, probably not, 1. It didn't crash on os x in my case, 2. it crashed when I tried to replace() some parts of it, not during any xpath/constructor calls. Nonetheless, I would report this (perhaps in a new, unused fresh ticket :D). Does this crash everytime? If yes, seperate the two constructor calls like: step 1 = XML(input).xpath.to_s.gub puts "I survived so far step2 = XML(step1) puts "should never reach this point" It would be very helpful if you could give the output of res.body. Or even better, the minimum of res.body which still triggers the crash. Cheers.
or cancel
mawiza Tue Oct 06 05:50:05 -0700 2009 | link
Hey Reto, it errors out on

doc = Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("&lt ;", "<").gsub("&gt ;", ">")

I have about 10 different XML files, and it crashes randomly on the different files, so I can't say that it's one specific file.

The XML files vary in size from 3mb to 150mb.

The files are very basic XML:

<?xml version="1.0" encoding="utf-8"?> <string>....</string>

where the string element contains escaped XML. Unfortunately the XML files are data we receive from an external vendor, so not really anything we can do about that.

I can try to normalize the data using gsub and then use nokogiri on the xmlified data.
Comment
Hey Reto, it errors out on doc = Nokogiri::XML(res.body).xpath("//text()").to_s.gsub("&lt ;", "<").gsub("&gt ;", ">") I have about 10 different XML files, and it crashes randomly on the different files, so I can't say that it's one specific file. The XML files vary in size from 3mb to 150mb. The files are very basic XML: <?xml version="1.0" encoding="utf-8"?> <string>....</string> where the string element contains escaped XML. Unfortunately the XML files are data we receive from an external vendor, so not really anything we can do about that. I can try to normalize the data using gsub and then use nokogiri on the xmlified data.
or cancel
flavorjones Tue Oct 06 05:58:36 -0700 2009 | link

Hi, can you please open a new issue for this? It appears to be unrelated to this actual issue, which has been solved in the nightlies.

Comment
Hi, can you please open a new issue for this? It appears to be unrelated to this actual issue, which has been solved in the nightlies.
or cancel

mawiza Tue Oct 06 06:02:26 -0700 2009 | link

Ok, I'll do that.

Comment
Ok, I'll do that.
or cancel

mawiza Tue Oct 06 06:10:51 -0700 2009 | link

Created a new ticket: http://github.com/tenderlove/nokogiri/issues/#issue/144

Comment
Created a new ticket: http://github.com/tenderlove/nokogiri/issues/#issue/144
or cancel

retoo Thu Oct 08 06:15:39 -0700 2009 | link

i close this one again. thx for the help guys!

Comment
i close this one again. thx for the help guys!
or cancel

tenderlove Tue Oct 13 13:49:01 -0700 2009 | link

Closing this because we fixed it. :-)

Comment
Closing this because we fixed it. :-)
or cancel

Parsed with GitHub Flavored Markdown
137.

1 vote

ffi
x

jruby
x

Support for JRuby on Windows
7 comments Created about 1 month ago by dlee

Title

Body
Right now, the JRuby gem does not ship with the DLLs. To work around this, I tried copying the DLLs from the Windows MRI gem, but then Nokogiri gives the following error: FFI::NotFoundError: Function 'calloc' not found in [exslt]
or cancel

Right now, the JRuby gem does not ship with the DLLs.

To work around this, I tried copying the DLLs from the Windows MRI gem, but then Nokogiri gives the following error:

FFI::NotFoundError: Function 'calloc' not found in [exslt]

Comments

jcfischer Thu Oct 01 01:23:29 -0700 2009 | link

We are having the same problems....

Comment
We are having the same problems....
or cancel

jojje Sun Oct 04 08:48:31 -0700 2009 | link

Same problems here which is unfortunate since the community really needs a cross-platform XML library. The developers of our project work on Windows, Linux as well as Solaris and deploy on Solaris. Nokogiri would be perfect in this regard if this bug was just fixed.

Please, this problem has existed for about a year and is a show-stopper for many projects. I know the developers time are limited and they do this in their spare time, but if wide usage of the library is of any interest, addressing this bug is a quick win to expand the use of Nokogiri.

Comment
Same problems here which is unfortunate since the community really needs a cross-platform XML library. The developers of our project work on Windows, Linux as well as Solaris and deploy on Solaris. Nokogiri would be perfect in this regard if this bug was just fixed. Please, this problem has existed for about a year and is a show-stopper for many projects. I know the developers time are limited and they do this in their spare time, but if wide usage of the library is of any interest, addressing this bug is a quick win to expand the use of Nokogiri.
or cancel

tenderlove Sun Oct 04 20:59:03 -0700 2009 | link

Hey everyone, this is a bug in JRuby. I've filed a ticket with them here:

http://jira.codehaus.org/browse/JRUBY-4052

Once they get it sorted out, I will close this ticket. :-)

Comment
Hey everyone, this is a bug in JRuby. I've filed a ticket with them here: http://jira.codehaus.org/browse/JRUBY-4052 Once they get it sorted out, I will close this ticket. :-)
or cancel

headius Mon Oct 05 12:09:10 -0700 2009 | link

I am confused about what it is we're not doing. If calloc is a (fairly) standard POSIX function, what is it we should be doing differently?

Comment
I am confused about what it is we're not doing. If calloc is a (fairly) standard POSIX function, what is it we should be doing differently?
or cancel

tenderlove Tue Oct 06 11:49:50 -0700 2009 | link

@headius I think the libc functions should be loaded by default. Wayne gives a workaround in the JIRA ticket, but I think it's unreasonable to require me to change my FFI code depending on the platform. The current code works on everything but windows.

Comment
@headius I think the libc functions should be loaded by default. Wayne gives a workaround in the JIRA ticket, but I think it's unreasonable to require me to change my FFI code depending on the platform. The current code works on everything but windows.
or cancel

jojje Sun Oct 11 06:28:31 -0700 2009 | link

Fix added to git://github.com/jojje/nokogiri.git for people to try having access to a Windows environment. Requires JRuby 1.4.0RC1 or higher due to some needed fixes regarding FFI.

Comment
Fix added to git://github.com/jojje/nokogiri.git for people to try having access to a Windows environment. Requires JRuby 1.4.0RC1 or higher due to some needed fixes regarding FFI.
or cancel

tenderlove Tue Oct 13 13:52:01 -0700 2009 | link

I applied this:

0555c69

I'm closing this ticket since it should be fixed for JRuby / Windows people with 1.4.0

Comment
I applied this: 0555c69e70083a6368b563f18da2c8d5416515cc I'm closing this ticket since it should be fixed for JRuby / Windows people with 1.4.0
or cancel

Parsed with GitHub Flavored Markdown
136.

1 vote

1.4.0
x

tenderlove
x

inner_html= dropping some elements
10 comments Created 2 months ago by bfolkens

Title

Body
I'm having some trouble on the following environment. The code below fails on a linux install but not on a macports install. Both environments are: $ nokogiri -v --- warnings: [] libxml: loaded: 2.7.3 binding: extension compiled: 2.7.3 nokogiri: 1.3.3 But on the linux environment, the following code: require 'test/unit' require 'rubygems' require 'nokogiri' class BugTest < Test::Unit::TestCase def test_should_parse_inner_text text = '<base><one>1</one><two>2</two></base>' doc = Nokogiri::XML(text) doc.search('base').each do |base_tag| base_tag.name = 'span' base_tag.inner_html = "#{base_tag.at('one').inner_text}/#{base_tag.at('two').inner_text}" end assert_equal '1/2', doc.to_html.strip end end Fails with: test_should_parse_inner_text(BugTest) [oo.rb:15]: <"1/2"> expected but was <"1">. Am I missing something obvious, or is this a bug? The above code is an abstraction from a larger project I'm working on, so I've tried to reduce it to the base of the issue. It passes the test on the MacPorts install (same version of libxml2 and nokogiri as on the Linux install). The Ruby -v on linux is: ruby 1.8.7 (2008-08-11 patchlevel 72) [i686-linux] And the MacPorts install is: ruby 1.8.7 (2008-08-11 patchlevel 72) [i686-darwin9]
or cancel
I'm having some trouble on the following environment. The code below fails on a linux install but not on a macports install. Both environments are:
```
$ nokogiri -v

--- 
warnings: []

libxml: 
 loaded: 2.7.3
 binding: extension
 compiled: 2.7.3
nokogiri: 1.3.3
```
But on the linux environment, the following code:
```
require 'test/unit'
require 'rubygems'
require 'nokogiri'

class BugTest < Test::Unit::TestCase
 def test_should_parse_inner_text
 text = '<base><one>1</one><two>2</two></base>'

 doc = Nokogiri::XML(text)
 doc.search('base').each do |base_tag|
 base_tag.name = 'span'
 base_tag.inner_html = "#{base_tag.at('one').inner_text}/#{base_tag.at('two').inner_text}"
 end

 assert_equal '1/2', doc.to_html.strip
 end
end
```
Fails with:
```
test_should_parse_inner_text(BugTest) [oo.rb:15]:
<"1/2"> expected but was
<"1">.
```
Am I missing something obvious, or is this a bug? The above code is an abstraction from a larger project I'm working on, so I've tried to reduce it to the base of the issue. It passes the test on the MacPorts install (same version of libxml2 and nokogiri as on the Linux install).

The Ruby -v on linux is:
```
ruby 1.8.7 (2008-08-11 patchlevel 72) [i686-linux]
```
And the MacPorts install is:
```
ruby 1.8.7 (2008-08-11 patchlevel 72) [i686-darwin9]
```
Comments
tenderlove Sat Sep 12 11:35:53 -0700 2009 | link

Strange. Are you sure nokogiri -v returns the same thing on the linux box? That seems crazy.

Comment
Strange. Are you *sure* nokogiri -v returns the same thing on the linux box? That seems crazy.
or cancel

bfolkens Sat Sep 12 11:49:30 -0700 2009 | link

Yeah - I thought I was going crazy so I even did a diff. Weirdest thing I've seen... AFAIK libxml2 doesn't really depend on much does it? Or is there something else that Nokogiri depends on that might be causing this? I tried libxml2 2.7.2 just in case, but still had the same problem.

Comment
Yeah - I thought I was going crazy so I even did a diff. Weirdest thing I've seen... AFAIK libxml2 doesn't really depend on much does it? Or is there something else that Nokogiri depends on that might be causing this? I tried libxml2 2.7.2 just in case, but still had the same problem.
or cancel

tenderlove Sat Sep 12 11:53:41 -0700 2009 | link

libxml2 only depends on iconv and zlib. Neither of those should cause this problem.

What linux are you running?

Comment
libxml2 only depends on iconv and zlib. Neither of those should cause this problem. What linux are you running?
or cancel

bfolkens Sat Sep 12 11:59:37 -0700 2009 | link

Gentoo (default/linux/x86/2008.0 profile) over the 2.6.18-xenU-ec2-v1.0 kernel

libxml2: 2.7.3-r2
libc: 2.8_p20080602-r1
zlib: 1.2.3-r1

Comment
Gentoo (default/linux/x86/2008.0 profile) over the 2.6.18-xenU-ec2-v1.0 kernel libxml2: 2.7.3-r2 libc: 2.8_p20080602-r1 zlib: 1.2.3-r1
or cancel

tenderlove Sat Sep 12 12:01:46 -0700 2009 | link

Okay. I'll get a gentoo box up and running. Might be a little while before I get this one to repro. :-(

Comment
Okay. I'll get a gentoo box up and running. Might be a little while before I get this one to repro. :-(
or cancel

bfolkens Sat Sep 12 12:09:46 -0700 2009 | link

Thanks a ton! In the meantime I'm trying to upgrade glibc and anything else that might be out of date, just to try some different versions of things.

Comment
Thanks a ton! In the meantime I'm trying to upgrade glibc and anything else that might be out of date, just to try some different versions of things.
or cancel

bfolkens Mon Sep 14 11:06:44 -0700 2009 | link

FWIW - The new glibc (2.9_p20081201-r2) didn't affect anything.

Comment
FWIW - The new glibc (2.9_p20081201-r2) didn't affect anything.
or cancel
bfolkens Mon Sep 14 19:06:53 -0700 2009 | link
Here's another take on it, if it helps at all:

require 'test/unit' require 'rubygems' require 'nokogiri' module Nokogiri::XML class Node include Test::Unit::Assertions def inner_html=(tags) children.each { |x| x.remove } assert_equal ['sup', 'sub'], document.fragment(tags).children.map {|n| n.name } document.fragment(tags).children.to_a.each do |node| add_child node end self end end end class BugTest < Test::Unit::TestCase def test_should_parse_inner_html text = '<base><one>1</one><two>2</two></base>' doc = Nokogiri::XML(text) base_tag = doc.at('base') base_tag.inner_html = "#{base_tag.at('one').inner_text}#{base_tag.at('two').inner_text}" assert_equal ['sup', 'sub'], base_tag.children.map {|n| n.name } end end

Successful return on the MacPorts install, and on the Linux install:

1) Failure: test_should_parse_inner_html(BugTest) [oo2.rb:12:in `inner_html=' oo2.rb:27:in `test_should_parse_inner_html']: <["sup", "sub"]> expected but was <["sup"]>.

In fact, even just this code returns only the first element and not the other:

Nokogiri::XML::DocumentFragment.parse("<one>1</one><two>2</two>")

Unless it's wrapped in another outer element like:

<x><one>1</one><two>2</two></x>

...then it returns the whole thing. And then obviously on the MacPorts install it returns an accurate copy regardless of the surrounding element.
Comment
Here's another take on it, if it helps at all: require 'test/unit' require 'rubygems' require 'nokogiri' module Nokogiri::XML class Node include Test::Unit::Assertions def inner_html=(tags) children.each { |x| x.remove } assert_equal ['sup', 'sub'], document.fragment(tags).children.map {|n| n.name } document.fragment(tags).children.to_a.each do |node| add_child node end self end end end class BugTest < Test::Unit::TestCase def test_should_parse_inner_html text = '<base><one>1</one><two>2</two></base>' doc = Nokogiri::XML(text) base_tag = doc.at('base') base_tag.inner_html = "#{base_tag.at('one').inner_text}#{base_tag.at('two').inner_text}" assert_equal ['sup', 'sub'], base_tag.children.map {|n| n.name } end end Successful return on the MacPorts install, and on the Linux install: 1) Failure: test_should_parse_inner_html(BugTest) [oo2.rb:12:in `inner_html=' oo2.rb:27:in `test_should_parse_inner_html']: <["sup", "sub"]> expected but was <["sup"]>. In fact, even just this code returns only the first element <one> and not the other: Nokogiri::XML::DocumentFragment.parse("<one>1</one><two>2</two>") Unless it's wrapped in another outer element like: <x><one>1</one><two>2</two></x> ...then it returns the whole thing. And then obviously on the MacPorts install it returns an accurate copy regardless of the surrounding element.
or cancel
bfolkens Mon Sep 14 20:01:00 -0700 2009 | link
I think I narrowed this down finally. For whatever reason, my local copy of Nokogiri (even though the gem was labeled 1.3.3) showed this diff from the copy on the linux machine (which was recently installed):

8,9c8,13 < @html_eh = node.kind_of? Nokogiri::HTML::DocumentFragment < --- > @klass = if node.kind_of?(Nokogiri::HTML::DocumentFragment) > Nokogiri::HTML::DocumentFragment > else > Nokogiri::XML::DocumentFragment > end > # 23,25c27,28 < regex = @html_eh ? %r{^\s*<#{Regexp.escape(name)}}i : < %r{^\s*<#{Regexp.escape(name)}} < --- > regex = (@klass == Nokogiri::HTML::DocumentFragment) ? %r{^\s*<#{Regexp.escape(name)}}i \ > : %r{^\s*<#{Regexp.escape(name)}}

So a fresh install on my MacPorts version now fails as well - lol - not quite the expected result. However, installing the gem from the master works great - so looks like this was already fixed ;)
Comment
I think I narrowed this down finally. For whatever reason, my local copy of Nokogiri (even though the gem was labeled 1.3.3) showed this diff from the copy on the linux machine (which was recently installed): 8,9c8,13 < @html_eh = node.kind_of? Nokogiri::HTML::DocumentFragment < --- > @klass = if node.kind_of?(Nokogiri::HTML::DocumentFragment) > Nokogiri::HTML::DocumentFragment > else > Nokogiri::XML::DocumentFragment > end > # 23,25c27,28 < regex = @html_eh ? %r{^\s*<#{Regexp.escape(name)}}i : < %r{^\s*<#{Regexp.escape(name)}} < --- > regex = (@klass == Nokogiri::HTML::DocumentFragment) ? %r{^\s*<#{Regexp.escape(name)}}i \ > : %r{^\s*<#{Regexp.escape(name)}} So a fresh install on my MacPorts version now fails as well - lol - not quite the expected result. However, installing the gem from the master works great - so looks like this was already fixed ;)
or cancel
tenderlove Mon Sep 14 21:07:45 -0700 2009 | link

Ugh. You're right. It fails against 1.3.3. I was checking against master. :-(

I guess I can stop fighting with VirtualBox now. Thanks for letting me know!

Comment
Ugh. You're right. It fails against 1.3.3. I was checking against master. :-( I guess I can stop fighting with VirtualBox now. Thanks for letting me know!
or cancel

Parsed with GitHub Flavored Markdown
135.

2 votes

Remove Node#collect_namespaces
3 comments Created 2 months ago by tenderlove

Title

Body
I think we should remove Node#collect_namespaces. Since namespace names are not unique, I don't know that this method is very useful.
or cancel

I think we should remove Node#collect_namespaces. Since namespace names are not unique, I don't know that this method is very useful.

Comments

flavorjones Mon Sep 14 15:36:38 -0700 2009 | link

+1

Comment
+1
or cancel

tenderlove Mon Sep 14 22:08:31 -0700 2009 | link

You're supposed to use the upvote button! ;-)

Although, I like the +1 better because it's not anonymous.

Comment
You're supposed to use the upvote button! ;-) Although, I like the +1 better because it's not anonymous.
or cancel

tenderlove Sun Oct 04 20:34:09 -0700 2009 | link

This was removed in c7eb4b2

Comment
This was removed in c7eb4b28896b220f338f4d203825feb9e73e3c8d
or cancel

Parsed with GitHub Flavored Markdown
134.

0 votes

1.4.0
x

tenderlove
x

Adding a node with a default namespace stores it as 'no-namespace' in the parent
2 comments Created 2 months ago by david

Title

Body
This works: doc = Nokogiri::XML("<element><child xmlns="woop:de:doo" /></element>") doc.at("//xmlns:child", 'xmlns' => 'woop:de:doo') #=> <child xmlns="woop:de:doo" /> This doesn't: doc = Nokogiri::XML::Document.new e = Nokogiri::XML::Node.new('element', doc) c = Nokogiri::XML::Node.new('child', doc) c.add_namespace(nil, 'woop:de:doo') e.add_child(c) doc.add_child(c) doc.at("//xmlns:child", 'xmlns' => 'woop:de:doo') #=> nil
or cancel
This works:
```
doc = Nokogiri::XML("<element><child xmlns="woop:de:doo" /></element>")
doc.at("//xmlns:child", 'xmlns' => 'woop:de:doo') #=> <child xmlns="woop:de:doo" />
```
This doesn't:
```
doc = Nokogiri::XML::Document.new
e = Nokogiri::XML::Node.new('element', doc)
c = Nokogiri::XML::Node.new('child', doc)
c.add_namespace(nil, 'woop:de:doo')
e.add_child(c)
doc.add_child(c)

doc.at("//xmlns:child", 'xmlns' => 'woop:de:doo') #=> nil
```
Comments
david Wed Sep 09 09:31:48 -0700 2009 | link
I'd also like to add that if you have a document like this:

<element> <c1 xmlns="one" /> <c2 xmlns="two" /> </element>

then

doc.root.collect_namespaces.inspect #=> {'xmlns' => 'two'}
Comment
I'd also like to add that if you have a document like this: <element> <c1 xmlns="one" /> <c2 xmlns="two" /> </element> then doc.root.collect_namespaces.inspect #=> {'xmlns' => 'two'}
or cancel
tenderlove Fri Sep 11 21:48:15 -0700 2009 | link

Yup. That is the danger of collect_namespaces. I think that method should be removed.

The first problem is fixed here: c6e5fa0

Comment
Yup. That is the danger of collect_namespaces. I think that method should be removed. The first problem is fixed here: c6e5fa0dedd8dd8ed1c7c0bdcf37c791b4dfc59c
or cancel

Parsed with GitHub Flavored Markdown
133.

0 votes

Cyrillyc chars
10 comments Created 2 months ago by romanvbabenko

Title

Body
<img src="http://rubyclub.com.ua/upload/2009/9/8/nokogiri.png"> I need original text like 'Пупкин' ruby -v ruby 1.8.7 (2008-08-11 patchlevel 72) [i486-linux] Linux lenny5/stable nokogiri -1.3.3.
or cancel
I need original text like 'Пупкин'

ruby -v
ruby 1.8.7 (2008-08-11 patchlevel 72) [i486-linux]

Linux lenny5/stable
nokogiri -1.3.3.

Comments
tenderlove Tue Sep 08 08:53:53 -0700 2009 | link

What encoding is the XML using?

Comment
What encoding is the XML using?
or cancel

romanvbabenko Tue Sep 08 09:02:40 -0700 2009 | link

$ enca staff.xml Universal transformation format 8 bits; UTF-8
Mixed line terminators

Comment
$ enca staff.xml Universal transformation format 8 bits; UTF-8 Mixed line terminators
or cancel
tenderlove Tue Sep 08 09:19:52 -0700 2009 | link
If you specify the encoding like this:

Nokogiri::XML('....', nil, 'UTF-8')

What does it return?
Comment
If you specify the encoding like this: Nokogiri::XML('....', nil, 'UTF-8') What does it return?
or cancel
romanvbabenko Tue Sep 08 09:38:47 -0700 2009 | link

text first post picture
<?xml version="1.0"?>
without encoding

Comment
text first post picture <?xml version="1.0"?> without encoding
or cancel
tenderlove Tue Sep 08 10:01:03 -0700 2009 | link
This works well for me:

require 'nokogiri' doc = Nokogiri::XML('<person last_name="Пупкин"></person>', nil, 'UTF-8') puts doc.at('person')['last_name']

How about you?
Comment
This works well for me: require 'nokogiri' doc = Nokogiri::XML('<person last_name="Пупкин"></person>', nil, 'UTF-8') puts doc.at('person')['last_name'] How about you?
or cancel
romanvbabenko Tue Sep 08 10:16:36 -0700 2009 | link

see thet
i need save to file readable text

Comment
<a href="http://romanvbabenko.com/system/assets/12/original/nokogiri-error.png">see thet</a> i need save to file readable text
or cancel
tenderlove Tue Sep 08 10:25:08 -0700 2009 | link
Try this:

require 'nokogiri' doc = Nokogiri::XML('<person last_name="Пупкин"></person>', nil, 'UTF-8') doc.encoding = 'UTF-8' puts doc.to_xml
Comment
Try this: require 'nokogiri' doc = Nokogiri::XML('<person last_name="Пупкин"></person>', nil, 'UTF-8') doc.encoding = 'UTF-8' puts doc.to_xml
or cancel
romanvbabenko Tue Sep 08 10:48:48 -0700 2009 | link

Oh. That is work fine. But, i see text in my file only after require 'multibyte'

http://github.com/mattetti/multibyte/tree/master

Comment
Oh. That is work fine. But, i see text in my file only after require 'multibyte' http://github.com/mattetti/multibyte/tree/master
or cancel

tenderlove Tue Sep 08 10:55:52 -0700 2009 | link

multibyte should not make a difference. Can you paste a short script somewhere that shows the problem?

Comment
multibyte should not make a difference. Can you paste a short script somewhere that shows the problem?
or cancel

tenderlove Fri Sep 11 21:09:12 -0700 2009 | link

I believe this is working as expected. Please reopen if I am incorrect.

Comment
I believe this is working as expected. Please reopen if I am incorrect.
or cancel

Parsed with GitHub Flavored Markdown
132.

0 votes

1.4.0
x

tenderlove
x

Segfault when searching with #at (1.3.3)
4 comments Created 2 months ago by wtn

Title

Body
/Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/xml/node.rb:591: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0] I think it happens whether I'm using OS X 10.5, 10.6, ruby 1.9.1, or 1.8.7 Also, it doesn't happen every time, but about a third of the time (I'm using Mechanize, perhaps the input is changing somewhat)
or cancel

/Library/Ruby/Gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/xml/node.rb:591: [BUG] Segmentation fault ruby 1.8.7 (2008-08-11 patchlevel 72) [universal-darwin10.0]

I think it happens whether I'm using OS X 10.5, 10.6, ruby 1.9.1, or 1.8.7
Also, it doesn't happen every time, but about a third of the time (I'm using Mechanize, perhaps the input is changing somewhat)

Comments

tenderlove Sun Sep 06 10:35:33 -0700 2009 | link

Could you possibly give me the HTML you're parsing? Or some sort of script so that I can reproduce it?

Comment
Could you possibly give me the HTML you're parsing? Or some sort of script so that I can reproduce it?
or cancel

wtn Sun Sep 06 15:39:32 -0700 2009 | link

Thanks. I discovered that this crashes about half the time when I paste it in irb, but when I run it as a script on the command line it doesn't crash.

I emailed you the script.

Comment
Thanks. I discovered that this crashes about half the time when I paste it in irb, but when I run it as a script on the command line it doesn't crash. I emailed you the script.
or cancel

tenderlove Fri Sep 11 21:08:43 -0700 2009 | link

I can't seem to reproduce this with Nokogiri 1.3.3. Could you possibly get me the document that makes this crash?

I suspect that the webpage is dynamic, and only one of the dynamic pages causes it to crash.

Comment
I can't seem to reproduce this with Nokogiri 1.3.3. Could you possibly get me the document that makes this crash? I suspect that the webpage is dynamic, and only one of the dynamic pages causes it to crash.
or cancel

tenderlove Mon Sep 14 22:15:21 -0700 2009 | link

Figured this out. There was a bug in Node#inspect. This commit fixed it:

928befa

Comment
Figured this out. There was a bug in Node#inspect. This commit fixed it: 928befa72b201f6c44818e286d079776f6be950c
or cancel

Parsed with GitHub Flavored Markdown
131.

0 votes

1.4.0
x

tenderlove
x

NodeSet#slice doesn't handle ranges beyond the end of the array
1 comment Created 2 months ago by zoozed

Title

Body
Looks like NodeSet#slice isn't handling range beyond the end of the array... #!/usr/bin/env ruby require 'rubygems' require 'nokogiri' xml = Nokogiri::XML('<?xml version="1.0" encoding="utf-8"?> <rss version="2.0"> <channel> <item><title>t1</title></item> <item><title>t2</title></item> </channel> </rss> ') items = (xml/:item) not_blowed_up = items[0, items.size] all_blowed_up = items[0, 100]
or cancel
Looks like NodeSet#slice isn't handling range beyond the end of the array...
```
#!/usr/bin/env ruby

require 'rubygems'
require 'nokogiri'

xml = Nokogiri::XML('<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0">
 <channel>
 <item><title>t1</title></item>
 <item><title>t2</title></item>
 </channel>
</rss>
')
items = (xml/:item)
not_blowed_up = items[0, items.size]
all_blowed_up = items[0, 100]
```
Comments

tenderlove Fri Sep 11 21:07:34 -0700 2009 | link

fixing node slices where the slice is larger than the node set length. closed by 9f04464

Comment
fixing node slices where the slice is larger than the node set length. closed by 9f04464ef9dbf526fa4bbd16fda132db7878503a
or cancel

Parsed with GitHub Flavored Markdown
130.

0 votes

racc requisite
3 comments Created 2 months ago by wtn

Title

Body
I think the reference to the racc tarball (v 1.4.5) should be removed from the gem compilation script because that version doesn't work on ruby 1.9. Instead of the user being directed to that tarball, the user should be told to install the racc gem (now 1.4.6 and compatable with 1.9).
or cancel
I think the reference to the racc tarball (v 1.4.5) should be removed from the gem compilation script because that version doesn't work on ruby 1.9. Instead of the user being directed to that tarball, the user should be told to install the racc gem (now 1.4.6 and compatable with 1.9).

Comments
tenderlove Tue Sep 01 13:16:10 -0700 2009 | link

I can't seem to find the code you are talking about. Can you send me a link? Or be more specific?

Thanks

Comment
I can't seem to find the code you are talking about. Can you send me a link? Or be more specific? Thanks
or cancel
wtn Tue Sep 01 14:08:53 -0700 2009 | link
Oh sorry, here's what I got on the console before I installed the racc gem:

localhost:home user$ sudo gem install tenderlove-nokogiri
Building native extensions. This could take a while...
ERROR: Error installing tenderlove-nokogiri:

ERROR: Failed to build gem native extension.

/usr/local/bin/ruby19 -rubygems /usr/local/lib/ruby19/gems/1.9.1/gems/rake-0.8.7/bin/rake RUBYARCHDIR=/usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/lib RUBYLIBDIR=/usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/lib Hoe.new {...} deprecated. Switch to Hoe.spec.
(in /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113) WARNING: HOE DEPRECATION: Add '>= 0' to the 'rake' dependency.
/usr/local/bin/ruby19 extconf.rb checking for xmlParseDoc() in -lxml2... yes
checking for xsltParseStylesheetDoc() in -lxslt... yes
checking for libxml/xmlversion.h in /usr/include/libxml2... yes
checking for libxslt/xslt.h in /usr/include... yes
checking for racc... no
need racc, get the tarball from http://i.loveruby.net/archive/racc/racc-1.4.5-all.tar.gz
extconf.rb failed Could not create Makefile due to some reason, probably lack of
necessary libraries and/or headers. Check the mkmf.log file for more
details. You may need configuration options.

Provided configuration options:

--with-opt-dir --without-opt-dir --with-opt-include --without-opt-include=${opt-dir}/include --with-opt-lib --without-opt-lib=${opt-dir}/lib --with-make-prog --without-make-prog --srcdir=. --curdir --ruby=/usr/local/bin/ruby19 --with-xml2lib --without-xml2lib --with-xsltlib --without-xsltlib

rake aborted!
Command failed with status (1): [/usr/local/bin/ruby19 extconf.rb...]
/usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/Rakefile:58:in `block (2 levels) in ' (See full trace by running task with --trace)

Gem files will remain installed in /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113 for inspection.
Results logged to /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/gem_make.out
Comment
Oh sorry, here's what I got on the console before I installed the racc gem: localhost:home user$ sudo gem install tenderlove-nokogiri Building native extensions. This could take a while... ERROR: Error installing tenderlove-nokogiri: ERROR: Failed to build gem native extension. /usr/local/bin/ruby19 -rubygems /usr/local/lib/ruby19/gems/1.9.1/gems/rake-0.8.7/bin/rake RUBYARCHDIR=/usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/lib RUBYLIBDIR=/usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/lib Hoe.new {...} deprecated. Switch to Hoe.spec. (in /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113) WARNING: HOE DEPRECATION: Add '>= 0' to the 'rake' dependency. /usr/local/bin/ruby19 extconf.rb checking for xmlParseDoc() in -lxml2... yes checking for xsltParseStylesheetDoc() in -lxslt... yes checking for libxml/xmlversion.h in /usr/include/libxml2... yes checking for libxslt/xslt.h in /usr/include... yes checking for racc... no need racc, get the tarball from http://i.loveruby.net/archive/racc/racc-1.4.5-all.tar.gz *** extconf.rb failed *** Could not create Makefile due to some reason, probably lack of necessary libraries and/or headers. Check the mkmf.log file for more details. You may need configuration options. Provided configuration options: --with-opt-dir --without-opt-dir --with-opt-include --without-opt-include=${opt-dir}/include --with-opt-lib --without-opt-lib=${opt-dir}/lib --with-make-prog --without-make-prog --srcdir=. --curdir --ruby=/usr/local/bin/ruby19 --with-xml2lib --without-xml2lib --with-xsltlib --without-xsltlib rake aborted! Command failed with status (1): [/usr/local/bin/ruby19 extconf.rb...] /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/Rakefile:58:in `block (2 levels) in <top (required)>' (See full trace by running task with --trace) Gem files will remain installed in /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113 for inspection. Results logged to /usr/local/lib/ruby19/gems/1.9.1/gems/tenderlove-nokogiri-0.0.0.20081021110113/gem_make.out
or cancel
tenderlove Tue Sep 01 14:58:55 -0700 2009 | link
Never install nokogiri from github. Always install it from rubyforge.

$ sudo gem install nokogiri

I will see about getting the gem removed from github. If it is hosted on github, that is a mistake.
Comment
Never install nokogiri from github. Always install it from rubyforge. $ sudo gem install nokogiri I will see about getting the gem removed from github. If it is hosted on github, that is a mistake.
or cancel
Parsed with GitHub Flavored Markdown
129.

0 votes

1.4.0
x

tenderlove
x

1.3.3 breaks RELAXNG.new(rng).validate(xml)
10 comments Created 2 months ago by DrusTheAxe

Title

Body
RELAXNG's validate() method changed in 1.3.3: * In 1.3.2, it returns an array of errors (and empty if succesful). * In 1.3.3, it always returns an empty array AND writes to stderr. This is probably due to 1.3.3's registering a libxml global error-handler. This is a critical breaking change on 2 counts: * validate() no longer returns errors ; impossible to programatically branch based on if the XML is valid or not * stderr may not exist; run this under IIS (where stderr is not set) and you fault NOTE: This was tested against Ruby 1.8.6 no Windows 7, but none of that should matter. At the very least a documented and clean workaround to alter the default error handler back to 1.3.2 behavior is critical; the docs are...thin...on the subject, and the only thing close to on point on the web are some dated forum messages in Feb'09 (the included code doesn't even parse correctly against 1.3.2 or 1.3.3). Below is a trivial test to repro the problem ----------------------------------------------------------------------------------------------- abort 'Usage: ruby relaxng_validate.rb <version>' if ARGV.empty? nokogiri_version = ARGV[0] require 'rubygems' gem 'nokogiri', nokogiri_version require 'nokogiri' puts "Nokogiri version #{Nokogiri::VERSION}" xml = <<EOXML <A/> EOXML schema = <<EOSCHEMA <?xml version="1.0" encoding="UTF-8"?> <grammar xmlns="http://relaxng.org/ns/structure/1.0"> <start> <ref name="A"/> </start> <define name="A"> <element name="A"> <interleave> <attribute name="B"/> <element name="C"> <text/> </element> <element name="D"> <element name="E"> <text/> </element> </element> </interleave> </element> </define> </grammar> EOSCHEMA puts 'Loading xml...' doc = Nokogiri::XML(xml) puts 'Loading schema...' relaxng = Nokogiri::XML::RelaxNG(schema) puts 'Validating xml against schema...' errors = relaxng.validate(doc) puts "Errors.size = #{errors.size}" errors.each { |error| puts " Error: #{error}" } puts 'Done.'
or cancel
RELAXNG's validate() method changed in 1.3.3:
In 1.3.2, it returns an array of errors (and empty if succesful). In 1.3.3, it always returns an empty array AND writes to stderr.

This is probably due to 1.3.3's registering a libxml global error-handler.

This is a critical breaking change on 2 counts:
validate() no longer returns errors ; impossible to programatically branch based on if the XML is valid or not stderr may not exist; run this under IIS (where stderr is not set) and you fault

NOTE: This was tested against Ruby 1.8.6 no Windows 7, but none of that should matter.

At the very least a documented and clean workaround to alter the default error handler back to 1.3.2 behavior is critical; the docs are...thin...on the subject, and the only thing close to on point on the web are some dated forum messages in Feb'09 (the included code doesn't even parse correctly against 1.3.2 or 1.3.3).

Below is a trivial test to repro the problem
```
abort 'Usage: ruby relaxng_validate.rb <version>' if ARGV.empty?
nokogiri_version = ARGV[0]
require 'rubygems'
gem 'nokogiri', nokogiri_version
require 'nokogiri'
puts "Nokogiri version #{Nokogiri::VERSION}"

xml = <<EOXML
<A/>
EOXML

schema = <<EOSCHEMA
<?xml version="1.0" encoding="UTF-8"?>
<grammar xmlns="http://relaxng.org/ns/structure/1.0">
 <start>
 <ref name="A"/>
 </start>
 <define name="A">
 <element name="A">
 <interleave>
 <attribute name="B"/>
 <element name="C">
 <text/>
 </element>
 <element name="D">
 <element name="E">
 <text/>
 </element>
 </element>
 </interleave>
 </element>
 </define>
</grammar>
EOSCHEMA

puts 'Loading xml...'
doc = Nokogiri::XML(xml)
puts 'Loading schema...'
relaxng = Nokogiri::XML::RelaxNG(schema)
puts 'Validating xml against schema...'
errors = relaxng.validate(doc)
puts "Errors.size = #{errors.size}"
errors.each { |error| puts " Error: #{error}" }

puts 'Done.'
```
Comments
DrusTheAxe Mon Aug 31 22:16:07 -0700 2009 | link
Oops. Run the command

ruby test.rb 1.3.2

and the output is

Nokogiri version 1.3.2 Loading xml... Loading schema... Validating xml against schema... Errors.size = 2 Error: Invalid sequence in interleave Error: Element A failed to validate content Done.

Run the command

ruby test.rb 1.3.3

and the output is

Nokogiri version 1.3.3 Loading xml... Loading schema... Validating xml against schema... element A: Relax-NG validity error : Invalid sequence in interleave element A: Relax-NG validity error : Element A failed to validate content Errors.size = 0 Done.
Comment
Oops. Run the command ruby test.rb 1.3.2 and the output is Nokogiri version 1.3.2 Loading xml... Loading schema... Validating xml against schema... Errors.size = 2 Error: Invalid sequence in interleave Error: Element A failed to validate content Done. Run the command ruby test.rb 1.3.3 and the output is Nokogiri version 1.3.3 Loading xml... Loading schema... Validating xml against schema... element A: Relax-NG validity error : Invalid sequence in interleave element A: Relax-NG validity error : Element A failed to validate content Errors.size = 0 Done.
or cancel
flavorjones Mon Aug 31 23:12:11 -0700 2009 | link

I cannot reproduce this on Ubuntu Linux 9.04.

Comment
I cannot reproduce this on Ubuntu Linux 9.04.
or cancel

flavorjones Mon Aug 31 23:17:45 -0700 2009 | link

Howard,

Can you please include some information about your platform, as well as the value of Nokogiri::VERSION_INFO (which is a hash) after loading Nokogiri 1.3.3 ?

Thanks much!

Comment
Howard, Can you please include some information about your platform, as well as the value of Nokogiri::VERSION_INFO (which is a hash) after loading Nokogiri 1.3.3 ? Thanks much!
or cancel
DrusTheAxe Mon Aug 31 23:34:54 -0700 2009 | link
Sure. Changing the line

puts "Nokogiri version #{Nokogiri::VERSION}"

to

puts "Ruby #{RUBY_PLATFORM} #{RUBY_VERSION}" puts "Nokogiri version #{Nokogiri::VERSION}" puts "Nokogiri VERSION_INFO: #{Nokogiri::VERSION_INFO.inspect}"

and run with

ruby test.rb 1.3.2

now shows at the top of the output

Ruby i386-mswin32 1.8.6 Nokogiri version 1.3.2 Nokogiri VERSION_INFO: {"nokogiri"=>"1.3.2", "warnings"=>[], "libxml"=>{"compiled"=>"2.7.3", "loaded"=>"2.7.3", "binding"=>"extension"}}

and when run with

ruby test.rb 1.3.3

outputs

Ruby i386-mswin32 1.8.6 Nokogiri version 1.3.3 Nokogiri VERSION_INFO: {"nokogiri"=>"1.3.3", "warnings"=>[], "libxml"=>{"compiled"=>"2.7.3", "loaded"=>"2.7.3", "binding"=>"extension"}}

As I said, I'm running 1.8.6 on Windows (XP and 7, same results).
The only difference in Nokogiri::VERSION_INFO appears to be the "nokogiri"=>"1.3.2" vs. "...1.3.3".

Nokogiri on Windows with --platform mswin32 has a prebuilt libxml, right?
Then I'd suspect the problem is in 1.3.3's registered libxml global error-handler
--and/or-- in the prebuilt libxml binary included in the Nokogiri 1.3.3 mswin32 gem.

P.S. I got Nokogiri via

get install Nokogiri --platform mswin32 --version 1.3.2 get install Nokogiri --platform mswin32 --version 1.3.3

to explicitly pull both versions down to my machine to repro the issue.

Anything else I can do to help?
Comment
Sure. Changing the line puts "Nokogiri version #{Nokogiri::VERSION}" to puts "Ruby #{RUBY_PLATFORM} #{RUBY_VERSION}" puts "Nokogiri version #{Nokogiri::VERSION}" puts "Nokogiri VERSION_INFO: #{Nokogiri::VERSION_INFO.inspect}" and run with ruby test.rb 1.3.2 now shows at the top of the output Ruby i386-mswin32 1.8.6 Nokogiri version 1.3.2 Nokogiri VERSION_INFO: {"nokogiri"=>"1.3.2", "warnings"=>[], "libxml"=>{"compiled"=>"2.7.3", "loaded"=>"2.7.3", "binding"=>"extension"}} and when run with ruby test.rb 1.3.3 outputs Ruby i386-mswin32 1.8.6 Nokogiri version 1.3.3 Nokogiri VERSION_INFO: {"nokogiri"=>"1.3.3", "warnings"=>[], "libxml"=>{"compiled"=>"2.7.3", "loaded"=>"2.7.3", "binding"=>"extension"}} As I said, I'm running 1.8.6 on Windows (XP and 7, same results). The only difference in Nokogiri::VERSION_INFO appears to be the "nokogiri"=>"1.3.2" vs. "...1.3.3". Nokogiri on Windows with --platform mswin32 has a prebuilt libxml, right? Then I'd suspect the problem is in 1.3.3's registered libxml global error-handler --and/or-- in the prebuilt libxml binary included in the Nokogiri 1.3.3 mswin32 gem. P.S. I got Nokogiri via get install Nokogiri --platform mswin32 --version 1.3.2 get install Nokogiri --platform mswin32 --version 1.3.3 to explicitly pull both versions down to my machine to repro the issue. Anything else I can do to help?
or cancel
flavorjones Tue Sep 01 00:45:30 -0700 2009 | link

Confirmed this issue on Windows XP with 1.3.3.

Comment
Confirmed this issue on Windows XP with 1.3.3.
or cancel
DrusTheAxe Tue Sep 01 18:17:24 -0700 2009 | link
A few questions:

Do you understand the source of the problem?

Do you have an ETA on a fix?

Do you have a workaround? (besides punting 1.3.3 and locking into 1.3.2)
Comment
A few questions: 1. Do you understand the source of the problem? 2. Do you have an ETA on a fix? 3. Do you have a workaround? (besides punting 1.3.3 and locking into 1.3.2)
or cancel
flavorjones Wed Sep 02 14:00:31 -0700 2009 | link

Howard -

Neither Aaron nor I will be able to spend the necessary time to debug this, since Windows is not a native platform for either of us. If you are under a time crunch, I'd advise backing down to 1.3.2.

And, if you have Windows expertise you'd like to lend to help us track the problem down, we'd love to have the help. Please remember, we're doing this in our spare time, for the benefit of humanity. That's you!

-m

Comment
Howard - Neither Aaron nor I will be able to spend the necessary time to debug this, since Windows is not a native platform for either of us. If you are under a time crunch, I'd advise backing down to 1.3.2. And, if you have Windows expertise you'd like to lend to help us track the problem down, we'd love to have the help. Please remember, we're doing this in our spare time, for the benefit of humanity. That's you! -m
or cancel

DrusTheAxe Thu Sep 03 22:53:07 -0700 2009 | link

Windows is native for me and I'm familiar with several XML libraries (Xerces, ElementTree, others), but new to Nokogiri and LibXML. I tried to hunt down how things were wired up, but didn't quite follow.

My suspicion is 1.3.3's wiring up of libxml's global error-handler, but i couldn't find such a thing. I did find what looks like 6-7 callbacks, but not sure which are which, and they look pretty cryptic anyway. Looks like there's some VB-Declare / .NET-P/Invoke / etc like native-ish-from-Ruby wiring ('pointer' and such), but I've only been doing Ruby for 4 months and haven't see that (in Ruby).

My suspicion is the newly registered global error-handler is dumping to stderr and not propogating the error messages up to Ruby like used to happen. That said, I'm not quite sure where to look. Not even sure it's Ruby code, could be native code, though the changelog comment said libxml-ruby (fwiw).

I'd be glad to help lend a hand, but at this point I've taken it as far as I can on my own. Looking for some pointers.

Comment
Windows is native for me and I'm familiar with several XML libraries (Xerces, ElementTree, others), but new to Nokogiri and LibXML. I tried to hunt down how things were wired up, but didn't quite follow. My suspicion is 1.3.3's wiring up of libxml's global error-handler, but i couldn't find such a thing. I did find what looks like 6-7 callbacks, but not sure which are which, and they look pretty cryptic anyway. Looks like there's some VB-Declare / .NET-P/Invoke / etc like native-ish-from-Ruby wiring ('pointer' and such), but I've only been doing Ruby for 4 months and haven't see that (in Ruby). My suspicion is the newly registered global error-handler is dumping to stderr and not propogating the error messages up to Ruby like used to happen. That said, I'm not quite sure *where* to look. Not even sure it's Ruby code, could be native code, though the changelog comment said libxml-ruby (fwiw). I'd be glad to help lend a hand, but at this point I've taken it as far as I can on my own. Looking for some pointers.
or cancel

tenderlove Sun Sep 06 14:47:31 -0700 2009 | link

Ugh. So, here is what I have learned so far. If I cross compile from the tag (REL_1.3.3), I cannot reproduce the problem. If I cross compile from HEAD, I cannot reproduce the problem. I can reproduce the problem with the released gem, and if I re-cross compile the released gem, I can reproduce the problem.

Also, I've noticed that the test suite does pick this up. Running the tests inside the released gem picks this up.

I am going to re-compile from the 1.3.3 release tag and replace the gems on rubyforge with the recompiled versions.

Comment
Ugh. So, here is what I have learned so far. If I cross compile from the tag (REL_1.3.3), I cannot reproduce the problem. If I cross compile from HEAD, I cannot reproduce the problem. I *can* reproduce the problem with the released gem, and if I re-cross compile the released gem, I can reproduce the problem. Also, I've noticed that the test suite *does* pick this up. Running the tests inside the released gem picks this up. I am going to re-compile from the 1.3.3 release tag and replace the gems on rubyforge with the recompiled versions.
or cancel

tenderlove Sun Sep 06 15:02:32 -0700 2009 | link

I've uploaded the new gems, so if you uninstall, then reinstall, everything should work.

I haven't found the root cause of this problem, but I'm closing this ticket because our tests exercise this behavior, and it is not reproducible from HEAD.

I need to find a less painful way of getting the tests running on windows. :-(

Comment
I've uploaded the new gems, so if you uninstall, then reinstall, everything should work. I haven't found the root cause of this problem, but I'm closing this ticket because our tests exercise this behavior, and it is not reproducible from HEAD. I need to find a less painful way of getting the tests running on windows. :-(
or cancel

Parsed with GitHub Flavored Markdown
128.

0 votes

Nokogiri Fail on Snow Leopard
2 comments Created 2 months ago by jwbynum

Title

Body
After installing nokogiri on Snow Leoaprd (using ARCHFLAG or not), it blows up with: dlopen(/usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle, 9): no suitable image found. Did find: /usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle: mach-o, but wrong architecture - /usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle
or cancel

After installing nokogiri on Snow Leoaprd (using ARCHFLAG or not), it blows up with:

dlopen(/usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle, 9): no suitable image found. Did find: /usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle: mach-o, but wrong architecture - /usr/local/lib/ruby/gems/1.8/gems/nokogiri-1.3.3/lib/nokogiri/nokogiri.bundle

Comments

tenderlove Mon Aug 31 10:18:26 -0700 2009 | link

Try uninstalling nokogiri and re-installing it.

Comment
Try uninstalling nokogiri and re-installing it.
or cancel

tenderlove Mon Aug 31 10:43:51 -0700 2009 | link

Turns out that ruby needed to be recompiled.

Closing. :-D

Comment
Turns out that ruby needed to be recompiled. Closing. :-D
or cancel

Parsed with GitHub Flavored Markdown
127.

0 votes

tenderlove
x

implement Nokogiri::XML::DTD#external_id and system_id
1 comment Created 2 months ago by tenderlove

Title

Body
implement Nokogiri::XML::DTD#external_id and system_id look at xmlDtdPtr
or cancel

implement Nokogiri::XML::DTD#external_id and system_id

look at xmlDtdPtr

Comments

tenderlove Sat Sep 12 14:06:51 -0700 2009 | link

adding DTD external id an system id. closed by 303b2b2

Comment
adding DTD external id an system id. closed by 303b2b29636e656b3dad66317c03033cc8e7a751
or cancel

Parsed with GitHub Flavored Markdown
126.

0 votes

1.4.0
x

tenderlove
x

Implement Nokogiri::XML::ElementDecl#content
1 comment Created 2 months ago by tenderlove

Title

Body
Implement Nokogiri::XML::ElementDecl#content look at tree.h struct _xmlElement, content member
or cancel

Implement Nokogiri::XML::ElementDecl#content

look at tree.h

struct _xmlElement, content member

Comments

tenderlove Sat Sep 12 17:14:18 -0700 2009 | link

updating changelog. closed by 1f658f0

Comment
updating changelog. closed by 1f658f04be49b62bbe4fd6eca8390d1809b17b3c
or cancel

Parsed with GitHub Flavored Markdown
125.

0 votes

strings returned by xpath expression "/text()" are bad formatted
1 comment Created 2 months ago by jney

Title

Body
Strings returned by Nokogiri::HTML(open(url)).xpath("//xpath_expression/text()").to_ary are such formatted that comparaison return false on identic strings. To avoid the problem i have to do it : Nokogiri::HTML(open(url)).xpath("//xpath_expression").collect(&:text)
or cancel

Strings returned by Nokogiri::HTML(open(url)).xpath("//xpath_expression/text()").to_ary are such formatted that comparaison return false on identic strings.
To avoid the problem i have to do it : Nokogiri::HTML(open(url)).xpath("//xpath_expression").collect(&:text)

Comments

tenderlove Sun Aug 30 10:45:37 -0700 2009 | link

Right, because it returns a Nokogiri::XML::Text node. That is different than a string.

Comment
Right, because it returns a Nokogiri::XML::Text node. That is different than a string.
or cancel

Parsed with GitHub Flavored Markdown
124.

0 votes

Nokogiri produces error "output error : unknown encoding" on certain pages
1 comment Created 2 months ago by mogman1

Title

Body
Every now and again Nokogiri will fail to process an HTML document, producing the error "output error : unknown encoding". Reference issue 122 as I strongly suspect that this is also related to the version of libxml2. This issue only comes up in my Windows XP environment and it works fine in my Linux environment (again, please see #122 as my environments have remained exactly the same). To see this, try the following: doc = open('http://businesslogos.com/resources_services.php') noko = Nokogiri::HTML(doc) That second line will generate the error. Most URLs are just fine, but the one above is an example of a URL that produces the error. I skimmed through the document looking for any bizarre characters but did not find anything obvious. If you upgrade the version of libxml2 to fix issue 122, that may very well fix this problem as well, but I wanted to alert you to it. Also, I want to reiterate that this is a problem that only shows up in the Windows environment and does not happen in the Linux envrionment, the exact opposite of the problem in issue 122 :-/ Sorry to lob two complaints in so short a time. I absolutely love the gem and think it's the best out there for this sort of thing, I am just trying to help make it even better!
or cancel

Every now and again Nokogiri will fail to process an HTML document, producing the error "output error : unknown encoding". Reference issue 122 as I strongly suspect that this is also related to the version of libxml2. This issue only comes up in my Windows XP environment and it works fine in my Linux environment (again, please see #122 as my environments have remained exactly the same). To see this, try the following:

doc = open('http://businesslogos.com/resources_services.php')
noko = Nokogiri::HTML(doc)

That second line will generate the error. Most URLs are just fine, but the one above is an example of a URL that produces the error. I skimmed through the document looking for any bizarre characters but did not find anything obvious. If you upgrade the version of libxml2 to fix issue 122, that may very well fix this problem as well, but I wanted to alert you to it. Also, I want to reiterate that this is a problem that only shows up in the Windows environment and does not happen in the Linux envrionment, the exact opposite of the problem in issue 122 :-/

Sorry to lob two complaints in so short a time. I absolutely love the gem and think it's the best out there for this sort of thing, I am just trying to help make it even better!

Comments

tenderlove Thu Aug 27 18:33:51 -0700 2009 | link

I believe this is a problem with the version of ICONV that nokogiri on windows is using. I will make sure it is upgraded in the next release, but beyond that, this is a problem I can't fix.

Comment
I believe this is a problem with the version of ICONV that nokogiri on windows is using. I will make sure it is upgraded in the next release, but beyond that, this is a problem I can't fix.
or cancel

Parsed with GitHub Flavored Markdown
123.

0 votes

1.4.0
x

tenderlove
x

DocumentFragment lacks detailed search
1 comment Created 2 months ago by samsm

Title

Body
fragment = 'hi' Nokogiri::HTML.fragment(fragment).search('#content').length # this returns zero Nokogiri::HTML(fragment).search('#content').length # this returns 1 Searching for an element ('p') does work, but using any CSS selector or XPath seems to always produce zero results. Non-fragment search works as I'd expect.
or cancel
```
fragment = 'hi'
Nokogiri::HTML.fragment(fragment).search('#content').length # this returns zero
Nokogiri::HTML(fragment).search('#content').length # this returns 1
```
Searching for an element ('p') does work, but using any CSS selector or XPath seems to always produce zero results. Non-fragment search works as I'd expect.
Comments

tenderlove Fri Aug 28 22:23:02 -0700 2009 | link

delegating DocumentFragment#css to the fragment children. closed by ed10f01

Comment
delegating DocumentFragment#css to the fragment children. closed by ed10f01b65d3ae03a12108e8daf530f88cd59fe7
or cancel

Parsed with GitHub Flavored Markdown
122.

0 votes

Nokogiri only returns HTML comment tags
6 comments Created 2 months ago by mogman1

Title

Body
Linux environment: CentOS 5.2 ruby 1.8.6 (2008-08-11 patchlevel 287) [x86_64-linux] rails 2.3.3 Nokogiri 1.3.3 Windows environment: Windows XP ruby 1.8.6 (2008-08-11 patchlevel 287) [i386-mswin32] rails 2.3.3 Nokogiri 1.3.3 This is a peculiar issue I discovered when I moved my application from my Windows XP development box to a linux box for production. When I run Nokogiri::HTML(open('http://www.some-url.com')) on my Windows environment, the entire HTML document is returned. However, when I do that same thing in the Linux environment, curiously only the HTML comment tags are returned. When I checked the temp file that open() generates, the entire HTML document was there, but for whatever reason Nokogiri only grabbed the comment tags. I checked and the same comment tags show inside the original document and in that order. Just for whatever reason, everything but comment tags are filtered out. My guess is that this is some weird bug having to do with the Linux version of Ruby working with Nokogiri, but otherwise I am thoroughly perplexed. I also tried using a previous version of Nokogiri (1.3.1) to see if that would work, but I got the same result.
or cancel

Linux environment:
CentOS 5.2
ruby 1.8.6 (2008-08-11 patchlevel 287) [x86_64-linux]
rails 2.3.3
Nokogiri 1.3.3

Windows environment:
Windows XP
ruby 1.8.6 (2008-08-11 patchlevel 287) [i386-mswin32]
rails 2.3.3
Nokogiri 1.3.3

This is a peculiar issue I discovered when I moved my application from my Windows XP development box to a linux box for production. When I run Nokogiri::HTML(open('http://www.some-url.com')) on my Windows environment, the entire HTML document is returned. However, when I do that same thing in the Linux environment, curiously only the HTML comment tags are returned.

When I checked the temp file that open() generates, the entire HTML document was there, but for whatever reason Nokogiri only grabbed the comment tags. I checked and the same comment tags show inside the original document and in that order. Just for whatever reason, everything but comment tags are filtered out.

My guess is that this is some weird bug having to do with the Linux version of Ruby working with Nokogiri, but otherwise I am thoroughly perplexed. I also tried using a previous version of Nokogiri (1.3.1) to see if that would work, but I got the same result.

Comments

mogman1 Tue Aug 25 08:58:50 -0700 2009 | link

It should be noted that I can pass in a string and Nokogiri parses things just fine. So while Nokogiri::HTML(open('http://www.some-url.com')) produces the strange behaviour, Nokogiri::HTML(open('http://www.some-url.com').read) works exactly as expected. So at least there is a work-around if someone else comes across this.

Comment
It should be noted that I can pass in a string and Nokogiri parses things just fine. So while Nokogiri::HTML(open('http://www.some-url.com')) produces the strange behaviour, Nokogiri::HTML(open('http://www.some-url.com').read) works exactly as expected. So at least there is a work-around if someone else comes across this.
or cancel

tenderlove Wed Aug 26 16:58:49 -0700 2009 | link

Can you run 'nokogiri -v' for me and add the output to the comments? I think this may be a bug in libxml2

Comment
Can you run 'nokogiri -v' for me and add the output to the comments? I think this may be a bug in libxml2
or cancel

mogman1 Thu Aug 27 10:04:51 -0700 2009 | link

nokogiri: 1.3.3
warnings: []

libxml:
compiled: 2.7.3 loaded: 2.7.3 binding: extension

Comment
nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.7.3 loaded: 2.7.3 binding: extension
or cancel

tenderlove Thu Aug 27 10:11:35 -0700 2009 | link

Is that from the linux environment, or the windows environment?

Comment
Is that from the linux environment, or the windows environment?
or cancel

mogman1 Thu Aug 27 11:03:20 -0700 2009 | link

Sorry, completely spaced there.

Linux:
nokogiri: 1.3.3
warnings: []

libxml:
compiled: 2.6.26 loaded: 2.6.26 binding: extension

Windows:
nokogiri: 1.3.3
warnings: []

libxml:
compiled: 2.7.3 loaded: 2.7.3 binding: extension

Comment
Sorry, completely spaced there. Linux: nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.6.26 loaded: 2.6.26 binding: extension Windows: nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.7.3 loaded: 2.7.3 binding: extension
or cancel

tenderlove Thu Aug 27 17:49:53 -0700 2009 | link

Okay. This is a bug in libxml2. If you upgrade your server to at least 2.6.32, the problem will go away.

Comment
Okay. This is a bug in libxml2. If you upgrade your server to at least 2.6.32, the problem will go away.
or cancel

Parsed with GitHub Flavored Markdown
121.

0 votes

REE
x

tcmalloc error parsing "<a></a>" fragment with REE
2 comments Created 2 months ago by henrik

Title

Body
Works fine with MRI, but not with REE: henrik@Nyx ~/Code$ which nokogiri /opt/ruby-enterprise-1.8.6-20090610/bin/nokogiri henrik@Nyx ~/Code$ nokogiri -v --- nokogiri: 1.3.3 warnings: [] libxml: compiled: 2.7.3 loaded: 2.7.3 binding: extension henrik@Nyx ~/Code$ which ruby /opt/ruby-enterprise-1.8.6-20090610/bin/ruby henrik@Nyx ~/Code$ ruby -rubygems -e 'require "nokogiri"; puts Nokogiri::HTML::DocumentFragment.parse("<a></a>")' src/tcmalloc.cc:186] Attempt to free invalid pointer: 0x201a90 Abort trap henrik@Nyx ~/Code$ Expected output is <a></a> without the error.
or cancel
Works fine with MRI, but not with REE:
```
henrik@Nyx ~/Code$ which nokogiri
/opt/ruby-enterprise-1.8.6-20090610/bin/nokogiri
henrik@Nyx ~/Code$ nokogiri -v
--- 
nokogiri: 1.3.3
warnings: []

libxml: 
 compiled: 2.7.3
 loaded: 2.7.3
 binding: extension
henrik@Nyx ~/Code$ which ruby
/opt/ruby-enterprise-1.8.6-20090610/bin/ruby
henrik@Nyx ~/Code$ ruby -rubygems -e 'require "nokogiri"; puts Nokogiri::HTML::DocumentFragment.parse("<a></a>")'
src/tcmalloc.cc:186] Attempt to free invalid pointer: 0x201a90
Abort trap
henrik@Nyx ~/Code$ 
```
Expected output is
```
<a></a>
```
without the error.
Comments

tenderlove Mon Sep 14 21:41:54 -0700 2009 | link

I'm not sure what to do about this. It's working for me:

http://skitch.com/aaron.patterson/b9ecb/terminal-bash-80x24

Comment
I'm not sure what to do about this. It's working for me: http://skitch.com/aaron.patterson/b9ecb/terminal-bash-80x24
or cancel

tenderlove Sun Oct 04 21:19:21 -0700 2009 | link

I can't repro this and the ticket hasn't been updated for almost a month. I will assume this is fixed on master.

Please reopen and update the ticket if it's still breaking against master. Thanks.

Comment
I can't repro this and the ticket hasn't been updated for almost a month. I will assume this is fixed on master. Please reopen and update the ticket if it's still breaking against master. Thanks.
or cancel

Parsed with GitHub Flavored Markdown
120.

0 votes

nokogiri-1.3.3 is not working with libxml2, libxslt in 1.3.1
1 comment Created 2 months ago by mao

Title

Body
I am use ubuntu 9.04 amd64, xml2, xslt and exslt libraries are installed. When I require nokogiri, I got following error: irb(main):003:0> require 'nokogiri' LoadError: Could not open any of [xml2, xslt, exslt] from /home/dan/installed/jruby/lib/ruby/1.8/ffi/library.rb:18:in `ffi_lib' from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri/ffi/libxml.rb:5 from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri/ffi/libxml.rb:31:in `require' from /home/dan/installed/jruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in `require' from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri.rb:10 from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri.rb:36:in `require' from /home/dan/installed/jruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:36:in `require' from (irb):4 I solved the problem and record them here: http://maodan520.spaces.live.com/blog/cns!E0C8D36B1650926A!237.entry But I think nokogiri is not working correct with neither libxml2.so nor libxslt.so, maybe it's a issue.
or cancel

I am use ubuntu 9.04 amd64, xml2, xslt and exslt libraries are installed. When I require nokogiri, I got following error:

irb(main):003:0> require 'nokogiri'
LoadError: Could not open any of [xml2, xslt, exslt]
from /home/dan/installed/jruby/lib/ruby/1.8/ffi/library.rb:18:in ffi_lib' from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri/ffi/libxml.rb:5 from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri/ffi/libxml.rb:31:inrequire'
from /home/dan/installed/jruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:31:in require' from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri.rb:10 from /home/dan/installed/jruby-1.3.1-rails/lib/ruby/gems/1.8/gems/nokogiri-1.3.3-java/lib/nokogiri.rb:36:inrequire'
from /home/dan/installed/jruby/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:36:in `require'
from (irb):4

I solved the problem and record them here:
http://maodan520.spaces.live.com/blog/cns!E0C8D36B1650926A!237.entry

But I think nokogiri is not working correct with neither libxml2.so nor libxslt.so, maybe it's a issue.

Comments

flavorjones Tue Aug 18 23:17:33 -0700 2009 | link

this problem was solved, i think. see #90 for followup.

Comment
this problem was solved, i think. see #90 for followup.
or cancel

Parsed with GitHub Flavored Markdown
119.

0 votes

NodeSet#search problems and NodeSet#+ is not different from NodeSet#&
1 comment Created 3 months ago by Serabe

Title

Body
Hi there: I've been tracking down some problems with my implementation for XPath and I think it is not my problem, but yours. Let me explain. In Nokogiri::XML::NodeSet#search you can see: each do |node| paths.each do |path| sub_set += send(path =~ /^(\.\/|\/)/ ? :xpath : :css, *(paths + [ns])) end end Ok, for each node you're iterating over paths and, for each path, you're calling either the xpath or the css method of **NodeSet**. And guess what, both of them iterates over each node... again. You can take a look at [search](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb#L70), [xpath](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb#L113) and [css](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb#L90) methods in [Nokogiri::XML::NodeSet](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb). The easiest way to fix this is deleting lines [75](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb#L75) and [80](http://github.com/tenderlove/nokogiri/blob/d30d0bfd5a07de643f919973cdee41759508b816/lib/nokogiri/xml/node_set.rb#L80). Just to explain why this is happening, my implementation is based on RubyArray and I used the op_plus function in it. By the way, if you're implementation of the plus operator removes duplicates I see no difference between + and &, so I would like to know what should I do in the java impl.
or cancel
Hi there:

I've been tracking down some problems with my implementation for XPath and I think it is not my problem, but yours. Let me explain. In Nokogiri::XML::NodeSet#search you can see:
```
each do |node|
  paths.each do |path|
    sub_set +=
      send(path =~ /^(\.\/|\/)/ ? :xpath : :css, *(paths + [ns]))
  end
end
```
Ok, for each node you're iterating over paths and, for each path, you're calling either the xpath or the css method of NodeSet. And guess what, both of them iterates over each node... again. You can take a look at search, xpath and css methods in Nokogiri::XML::NodeSet.

The easiest way to fix this is deleting lines 75 and 80.

Just to explain why this is happening, my implementation is based on RubyArray and I used the op_plus function in it. By the way, if you're implementation of the plus operator removes duplicates I see no difference between + and &, so I would like to know what should I do in the java impl.
Comments

tenderlove Wed Aug 12 11:44:31 -0700 2009 | link

making NodeSet more consistent with Set, adding NodeSet#| closed by 541cbeb

Comment
making NodeSet more consistent with Set, adding NodeSet#| closed by 541cbebe4ae13521f1e1c19baef8bdd65e5e2614
or cancel

Parsed with GitHub Flavored Markdown
118.

0 votes

1.4.0
x

tenderlove
x

Convert meta_encoding and meta_encoding= to ruby
1 comment Created 3 months ago by tenderlove

Title

Body
These methods need to be converted to Ruby. The current meta_encoding= method will call xmlFreeNode() on the old meta tag which will cause a segv: require 'nokogiri' doc = Nokogiri::HTML DATA.read node = doc.at('meta') puts node.name doc.meta_encoding = 'EUC-JP' p node __END__ <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <title>Hello World</title> </head> <body> <h1>Hello Again</h1> </body> </html>
or cancel
These methods need to be converted to Ruby. The current meta_encoding= method will call xmlFreeNode() on the old meta tag which will cause a segv:
```
require 'nokogiri'

doc = Nokogiri::HTML DATA.read
node = doc.at('meta')

puts node.name

doc.meta_encoding = 'EUC-JP'

p node

__END__
<html>
 <head>
 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
 <title>Hello World</title>
 </head>
 <body>
 <h1>Hello Again</h1>
 </body>
</html>
```
Comments

tenderlove Tue Oct 13 21:38:38 -0700 2009 | link

meta_encoding and meta_encoding= are implemented in ruby. closed by ceffd26

Comment
meta_encoding and meta_encoding= are implemented in ruby. closed by ceffd26b932b1b564cdd7ec9b107f7087e80f384
or cancel

Parsed with GitHub Flavored Markdown
117.

0 votes

1.4.0
x

tenderlove
x

Shouldn't inner_html convert to UTF8 the same way as inner_text?
4 comments Created 3 months ago by naofumi

Title

Body
There seems to be an inconsistency between how encoding conversion is applied with the inner_text, inner_html and to_html methods. With #inner_text, I think all output is automatically converted to UTF8. With #inner_html, encoding conversions are not applied. With #to_html, you can specify the desired encoding for the result with the :encoding option. I would prefer that the output for both #inner_html and #to_html are converted to UTF8 by default, but that you can override this with the :encoding option. At least, it would be nice to be able to pass the :encoding option to #inner_html.
or cancel
There seems to be an inconsistency between how encoding conversion is applied with the inner_text, inner_html and to_html methods.

With #inner_text, I think all output is automatically converted to UTF8.
With #inner_html, encoding conversions are not applied.
With #to_html, you can specify the desired encoding for the result with the :encoding option.

I would prefer that the output for both #inner_html and #to_html are converted to UTF8 by default, but that you can override this with the :encoding option.

At least, it would be nice to be able to pass the :encoding option to #inner_html.

Comments
naofumi Fri Aug 07 07:07:16 -0700 2009 | link
In order to provide an :encoding option for #inner_html, maybe the following example;
in nokogiri/xml/node.rb

def inner_html (\*args) children.map { |x| x.to_html(\*args) }.join end
in nokogiri/xml/node_set

def inner_html (\*args) collect{|j| j.inner_html(\*args)}.join('') end
Comment
In order to provide an :encoding option for #inner_html, maybe the following example; in nokogiri/xml/node.rb <code><pre> def inner_html (\*args) children.map { |x| x.to_html(\*args) }.join end </pre></code> in nokogiri/xml/node_set <code><pre> def inner_html (\*args) collect{|j| j.inner_html(\*args)}.join('') end </pre></code>
or cancel
tenderlove Thu Aug 27 18:16:15 -0700 2009 | link
I've added the ability to pass encoding to #inner_html. I don't want to automatically convert all documents to UTF-8 when calling #to_html. I think that would be bad for people processing documents in something besides UTF-8, and they want the final output to remain the specified encoding.

If you always want the output to be UTF-8, just tell the document that it should be encoded with UTF-8 like so:

doc = Nokogiri::HTML open('http://example.com/') doc.encoding = 'UTF-8' # Set the document encoding to UTF-8

After doing that inner_html and to_html will return UTF-8 documents.
Comment
I've added the ability to pass encoding to #inner_html. I don't want to automatically convert all documents to UTF-8 when calling #to_html. I think that would be bad for people processing documents in something besides UTF-8, and they want the final output to remain the specified encoding. If you always want the output to be UTF-8, just tell the document that it should be encoded with UTF-8 like so: doc = Nokogiri::HTML open('http://example.com/') doc.encoding = 'UTF-8' # Set the document encoding to UTF-8 After doing that inner_html and to_html will return UTF-8 documents.
or cancel
tenderlove Thu Aug 27 18:17:40 -0700 2009 | link

inner_html takes the same arguments as to_html. closed by ab9a8a0

Comment
inner_html takes the same arguments as to_html. closed by ab9a8a0a8f9d6e85764e97d988ebea8c2cd0021c
or cancel

naofumi Sun Aug 30 19:07:35 -0700 2009 | link

Thanks. Sounds good.

Comment
Thanks. Sounds good.
or cancel

Parsed with GitHub Flavored Markdown
116.

0 votes

nokogiri-1.3.3 introduces dependency on st.h -- error: st.h: No such file or directory
1 comment Created 3 months ago by TylerRick

Title

Body
1.3.2 installs fine but I can't seem to build/install 1.3.3. I'm running Ubuntu 9.04. What is st.h and how do I get it to find it? Thanks! <pre> > sudo gem1.9 install nokogiri -v 1.3.2 Building native extensions. This could take a while... Successfully installed nokogiri-1.3.2 1 gem installed Installing ri documentation for nokogiri-1.3.2... Installing RDoc documentation for nokogiri-1.3.2... > sudo gem1.9 install nokogiri -v 1.3.3 Building native extensions. This could take a while... ERROR: Error installing nokogiri: ERROR: Failed to build gem native extension. /usr/bin/ruby1.9 extconf.rb install nokogiri -v 1.3.3 checking for iconv.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes checking for libxml/parser.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes checking for libxslt/xslt.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes checking for libexslt/exslt.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes checking for xmlParseDoc() in -lxml2... yes checking for xsltParseStylesheetDoc() in -lxslt... yes checking for exsltFuncRegister() in -lexslt... yes checking for xmlRelaxNGSetParserStructuredErrors()... yes checking for xmlRelaxNGSetParserStructuredErrors()... yes checking for xmlRelaxNGSetValidStructuredErrors()... yes checking for xmlSchemaSetValidStructuredErrors()... yes checking for xmlSchemaSetParserStructuredErrors()... yes creating Makefile make cc -I. -I/usr/include/libxml2 -I/usr/include -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_XMLRELAXNGSETPARSERSTRUCTUREDERRORS -DHAVE_XMLRELAXNGSETPARSERSTRUCTUREDERRORS -DHAVE_XMLRELAXNGSETVALIDSTRUCTUREDERRORS -DHAVE_XMLSCHEMASETVALIDSTRUCTUREDERRORS -DHAVE_XMLSCHEMASETPARSERSTRUCTUREDERRORS -I/usr/include/libxml2 -I/usr/include -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses -fPIC -g -DXP_UNIX -O3 -Wall -Wcast-qual -Wwrite-strings -Wconversion -Wmissing-noreturn -Winline -o xml_reader.o -c xml_reader.c In file included from /usr/include/ruby-1.9.0/ruby.h:15, from ./nokogiri.h:6, from ./xml_reader.h:4, from xml_reader.c:1: /usr/include/ruby-1.9.0/ruby/ruby.h: In function ‘rb_type’: /usr/include/ruby-1.9.0/ruby/ruby.h:973: warning: conversion to ‘int’ from ‘VALUE’ may alter its value In file included from ./nokogiri.h:81, from ./xml_reader.h:4, from xml_reader.c:1: ./xml_document.h:5:16: error: st.h: No such file or directory xml_reader.c: In function ‘attribute_nodes’: xml_reader.c:171: warning: cast discards qualifiers from pointer target type xml_reader.c: In function ‘attribute_at’: xml_reader.c:199: warning: conversion to ‘int’ from ‘long int’ may alter its value xml_reader.c: In function ‘from_memory’: xml_reader.c:466: warning: conversion to ‘int’ from ‘long int’ may alter its value xml_reader.c:474: warning: conversion to ‘int’ from ‘long int’ may alter its value xml_reader.c: In function ‘from_io’: xml_reader.c:506: warning: conversion to ‘int’ from ‘long int’ may alter its value make: *** [xml_reader.o] Error 1 </pre>
or cancel
1.3.2 installs fine but I can't seem to build/install 1.3.3. I'm running Ubuntu 9.04.

What is st.h and how do I get it to find it?

Thanks!
```
> sudo gem1.9 install nokogiri -v 1.3.2
Building native extensions. This could take a while...
Successfully installed nokogiri-1.3.2
1 gem installed
Installing ri documentation for nokogiri-1.3.2...
Installing RDoc documentation for nokogiri-1.3.2...

> sudo gem1.9 install nokogiri -v 1.3.3
Building native extensions. This could take a while...
ERROR: Error installing nokogiri:
 ERROR: Failed to build gem native extension.

/usr/bin/ruby1.9 extconf.rb install nokogiri -v 1.3.3
checking for iconv.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes
checking for libxml/parser.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes
checking for libxslt/xslt.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes
checking for libexslt/exslt.h in /opt/local/include/,/opt/local/include/libxml2,/opt/local/include,/opt/local/include,/opt/local/include/libxml2,/usr/local/include,/usr/local/include/libxml2,/usr/include,/usr/include/libxml2,/usr/include,/usr/include/libxml2... yes
checking for xmlParseDoc() in -lxml2... yes
checking for xsltParseStylesheetDoc() in -lxslt... yes
checking for exsltFuncRegister() in -lexslt... yes
checking for xmlRelaxNGSetParserStructuredErrors()... yes
checking for xmlRelaxNGSetParserStructuredErrors()... yes
checking for xmlRelaxNGSetValidStructuredErrors()... yes
checking for xmlSchemaSetValidStructuredErrors()... yes
checking for xmlSchemaSetParserStructuredErrors()... yes
creating Makefile

make
cc -I. -I/usr/include/libxml2 -I/usr/include -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -DHAVE_XMLRELAXNGSETPARSERSTRUCTUREDERRORS -DHAVE_XMLRELAXNGSETPARSERSTRUCTUREDERRORS -DHAVE_XMLRELAXNGSETVALIDSTRUCTUREDERRORS -DHAVE_XMLSCHEMASETVALIDSTRUCTUREDERRORS -DHAVE_XMLSCHEMASETPARSERSTRUCTUREDERRORS -I/usr/include/libxml2 -I/usr/include -I/usr/include/ruby-1.9.0/x86_64-linux -I/usr/include/ruby-1.9.0 -I. -fPIC -fno-strict-aliasing -g -g -O2 -O2 -g -Wall -Wno-parentheses -fPIC -g -DXP_UNIX -O3 -Wall -Wcast-qual -Wwrite-strings -Wconversion -Wmissing-noreturn -Winline -o xml_reader.o -c xml_reader.c
In file included from /usr/include/ruby-1.9.0/ruby.h:15,
 from ./nokogiri.h:6,
 from ./xml_reader.h:4,
 from xml_reader.c:1:
/usr/include/ruby-1.9.0/ruby/ruby.h: In function ‘rb_type’:
/usr/include/ruby-1.9.0/ruby/ruby.h:973: warning: conversion to ‘int’ from ‘VALUE’ may alter its value
In file included from ./nokogiri.h:81,
 from ./xml_reader.h:4,
 from xml_reader.c:1:
./xml_document.h:5:16: error: st.h: No such file or directory
xml_reader.c: In function ‘attribute_nodes’:
xml_reader.c:171: warning: cast discards qualifiers from pointer target type
xml_reader.c: In function ‘attribute_at’:
xml_reader.c:199: warning: conversion to ‘int’ from ‘long int’ may alter its value
xml_reader.c: In function ‘from_memory’:
xml_reader.c:466: warning: conversion to ‘int’ from ‘long int’ may alter its value
xml_reader.c:474: warning: conversion to ‘int’ from ‘long int’ may alter its value
xml_reader.c: In function ‘from_io’:
xml_reader.c:506: warning: conversion to ‘int’ from ‘long int’ may alter its value
make: *** [xml_reader.o] Error 1
```
Comments

tenderlove Thu Aug 06 13:35:47 -0700 2009 | link

Ruby 1.9.0 is not supported. You should upgrade to 1.9.1-p129 or even the 1.9.2. 1.9.0 is too broken to be supported. :-(

Comment
Ruby 1.9.0 is not supported. You should upgrade to 1.9.1-p129 or even the 1.9.2. 1.9.0 is too broken to be supported. :-(
or cancel

Parsed with GitHub Flavored Markdown
115.

0 votes

1.4.0
x

tenderlove
x

"ArgumentError: NULL pointer given" on calling doc.meta_encoding
1 comment Created 3 months ago by arndtjenssen

Title

Body
Is thrown on (malformed) documents with missing encoding info on v1.3.2 Steps to reproduce: require 'nokogiri' require 'open-uri' doc = Nokogiri::HTML(open('http://www.europapress.es/valencia/noticia-moby-dick-john-huston-obrira-cap-setmana-filmoteca-destiu-valencia-20090731183544.html')) doc.meta_encoding results in: ArgumentError: NULL pointer given from (irb):55:in `meta_encoding' from (irb):55 from :0
or cancel
Is thrown on (malformed) documents with missing encoding info on v1.3.2
Steps to reproduce:
```
require 'nokogiri'
require 'open-uri'
doc = Nokogiri::HTML(open('http://www.europapress.es/valencia/noticia-moby-dick-john-huston-obrira-cap-setmana-filmoteca-destiu-valencia-20090731183544.html'))
doc.meta_encoding
```
results in:
```
ArgumentError: NULL pointer given
    from (irb):55:in `meta_encoding'
    from (irb):55
    from :0
```
Comments

tenderlove Thu Aug 06 21:24:55 -0700 2009 | link

returns nil when an HTML document does not declare a meta encoding tag. closed by d0e9312

Comment
returns nil when an HTML document does not declare a meta encoding tag. closed by d0e93128213a10376944bfe3943ce23547a33d14
or cancel

Parsed with GitHub Flavored Markdown
114.

0 votes

1.4.0
x

tenderlove
x

Error parsing provided css_path()
5 comments Created 3 months ago by alkanshel

Title

Body
Nokogiri occasionally has issues parsing the css_path it provides. A sample case is the following path (retrieved by running Node.css_path()): html > body > div > div:nth-of-type(2) > div > text():nth-of-type(2) Reasonably certain the cause is the 'text():nth-of-type(2)', which is generated from the corresponding .path of /html/body/div/form/table[3]/tr[11]/td/div/div/div/div/table/tr[2]/td[2]/span[2]/text(). The error message is 'Unexpected ':' in #<Nokigiri::CSS::...'
or cancel

Nokogiri occasionally has issues parsing the css_path it provides. A sample case is the following path (retrieved by running Node.css_path()):

html > body > div > div:nth-of-type(2) > div > text():nth-of-type(2)

Reasonably certain the cause is the 'text():nth-of-type(2)', which is generated from the corresponding .path of /html/body/div/form/table[3]/tr[11]/td/div/div/div/div/table/tr[2]/td[2]/span[2]/text().

The error message is 'Unexpected ':' in #<Nokigiri::CSS::...'

Comments

tenderlove Thu Aug 06 20:32:08 -0700 2009 | link

That XPath doesn't result in the CSS path you're showing. Can you give me a code example with reference HTML for me?

Comment
That XPath doesn't result in the CSS path you're showing. Can you give me a code example with reference HTML for me?
or cancel

alkanshel Fri Aug 07 09:12:34 -0700 2009 | link

Hmm. Okay, I'll check my code/data set and see if I can replicate the issue with better documentation.

Comment
Hmm. Okay, I'll check my code/data set and see if I can replicate the issue with better documentation.
or cancel

tenderlove Fri Aug 07 09:13:55 -0700 2009 | link

Thank you, I'd really appreciate it!

Comment
Thank you, I'd really appreciate it!
or cancel

alkanshel Fri Aug 07 15:48:20 -0700 2009 | link

Weird. I ran Nokogiri over the same data set and looked over the CSS paths generated, and I can't seem to replicate the issue. I'll have to chalk it up to solar radiation or PEBKAC, then. Sorry about that.

Comment
Weird. I ran Nokogiri over the same data set and looked over the CSS paths generated, and I can't seem to replicate the issue. I'll have to chalk it up to solar radiation or PEBKAC, then. Sorry about that.
or cancel

tenderlove Fri Aug 07 19:19:11 -0700 2009 | link

Haha! No problem. If you run across it again, please let me know!

Comment
Haha! No problem. If you run across it again, please let me know!
or cancel

Parsed with GitHub Flavored Markdown
113.

0 votes

1.4.0
x

ffi
x

tenderlove
x

Fix platform detection code
1 comment Created 3 months ago by tenderlove

Title

Body
We need to be able to tell when a user is running on windows, not by just the platform. Right now, the code looks at the platform when it needs to look at the OS. Switch to this (from Luis): RUBY_PLATFORM + RbConfig::CONFIG['host_os'] Similar approach is being used by mspec and the RubySpec to determine which API behavior should be checked for Java on Windows.
or cancel
We need to be able to tell when a user is running on windows, not by just the platform. Right now, the code looks at the platform when it needs to look at the OS. Switch to this (from Luis):
```
RUBY_PLATFORM + RbConfig::CONFIG['host_os']

Similar approach is being used by mspec and the RubySpec to determine
which API behavior should be checked for Java on Windows.
```
Comments

tenderlove Thu Aug 06 21:24:55 -0700 2009 | link

using host OS to figure out ENV["PATH"]. closed by 544b431

Comment
using host OS to figure out ENV["PATH"]. closed by 544b431e5ac6a12369ddc6fb01d2a4b06dbd534d
or cancel

Parsed with GitHub Flavored Markdown
112.

0 votes

1.4.0
x

ffi
x

tenderlove
x

Package libxml2 dll's with jruby gem
2 comments Created 3 months ago by tenderlove

Title

Body
Package libxml2 dll's with jruby gem so that the jruby gem can work on windows / jruby
or cancel
Package libxml2 dll's with jruby gem so that the jruby gem can work on windows / jruby

Comments
tenderlove Thu Jul 30 09:44:59 -0700 2009 | link
I forgot, people seem to get this error:

C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ffi/ffi.rb: 114:in `create _invoker': Function 'calloc' not found in [exslt] (FFI::NotFoundError) from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:50:in `attach_function' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:48:in `each' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:48:in `attach_function' from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri/ffi/libxml.rb:54 from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri/ffi/libxml.rb:31:in `require' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/ site_ruby/1.8/ru bygems/custom_require.rb:31:in `require' from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri.rb:10 from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri.rb:36:in `require' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/ site_ruby/1.8/ru bygems/custom_require.rb:36:in `require' from TestNokogiri.rb:3
Comment
I forgot, people seem to get this error: C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ffi/ffi.rb: 114:in `create _invoker': Function 'calloc' not found in [exslt] (FFI::NotFoundError) from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:50:in `attach_function' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:48:in `each' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/1.8/ ffi/library. rb:48:in `attach_function' from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri/ffi/libxml.rb:54 from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri/ffi/libxml.rb:31:in `require' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/ site_ruby/1.8/ru bygems/custom_require.rb:31:in `require' from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri.rb:10 from C:/Program Files/JRuby/jruby-1.3.1/lib/ruby/gems/1.8/gems/ nokogiri- 1.3.2-x86-mswin32/lib/nokogiri.rb:36:in `require' from C:/Program Files/JRuby/jruby-1.3.1/bin/../lib/ruby/ site_ruby/1.8/ru bygems/custom_require.rb:36:in `require' from TestNokogiri.rb:3
or cancel
tenderlove Wed Aug 12 11:45:21 -0700 2009 | link

This was fixed here: 70ad006

Comment
This was fixed here: 70ad006733c6d48d761a7e5c36c1b6f677c2aa60
or cancel

Parsed with GitHub Flavored Markdown
111.

0 votes

Windows Nokogiri 1.3.3 - ALWAYS LoadError: no such file to load -- nokogiri/1.9/nokogiri
1 comment Created 3 months ago by ocoolio

Title

Body
No matter how you set up, on every Windows system the newest Nokogiri (1.3.3) fails to load with the following error (both 1.9 and 1.8 Ruby): irb(main):001:0> require 'rubygems' => false irb(main):002:0> require 'nokogiri' LoadError: no such file to load -- nokogiri/1.9/nokogiri from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3-x86-mingw32/lib/nok ogiri/nokogiri.rb:1:in `require' from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3-x86-mingw32/lib/nok ogiri/nokogiri.rb:1:in `<top (required)>' from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri.rb:12: in `require' from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri.rb:12: 1.3.2 works great
or cancel
No matter how you set up, on every Windows system the newest Nokogiri (1.3.3) fails to load with the following error (both 1.9 and 1.8 Ruby):

irb(main):001:0> require 'rubygems'
=> false irb(main):002:0> require 'nokogiri'
LoadError: no such file to load -- nokogiri/1.9/nokogiri
```
 from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3-x86-mingw32/lib/nok
```
ogiri/nokogiri.rb:1:in `require'
```
 from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3-x86-mingw32/lib/nok
```
ogiri/nokogiri.rb:1:in `'
```
 from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri.rb:12:
```
in `require'
```
 from c:/Ruby/lib/ruby/gems/1.9.1/gems/nokogiri-1.3.3/lib/nokogiri.rb:12:
```
1.3.2 works great
Comments

tenderlove Tue Jul 28 13:20:50 -0700 2009 | link

Sorry about that. I uploaded the wrong gems. Please uninstall that version and try again.

Comment
Sorry about that. I uploaded the wrong gems. Please uninstall that version and try again.
or cancel

Parsed with GitHub Flavored Markdown
110.

0 votes

I have issues
1 comment Created 3 months ago by zenspider

Title

Body
boy... do I have issues. (ignore me... using it for developing omnifocus-github)
or cancel

boy... do I have issues.

(ignore me... using it for developing omnifocus-github)

Comments

jmhodges Tue Jul 28 01:54:30 -0700 2009 | link

Okay, bye!

Comment
Okay, bye!
or cancel

Parsed with GitHub Flavored Markdown
109.

1 vote

mdalessio
x

REE
x

document.rb:104: [BUG] object allocation during garbage collection phase
5 comments Created 3 months ago by darryl

Title

Body
require 'nokogiri' GC_HACK = false GC.disable if GC_HACK # will delay "error : Name is not from the document dictionnary" gc_count = 0 cycles = 0 loop do cycles = cycles + 1 if GC_HACK if gc_count > 10000 GC.enable GC.start p "gc start cycles: #{cycles}" sleep 10 gc_count = 0 GC.disable end gc_count = gc_count + 1 end p "cycles: #{cycles}" if cycles%1000 == 0 doc = Nokogiri::XML::Document.parse("<bad>blinky</bad>") doc.xpath('/bad').each{ |t| new_node = Nokogiri::XML::Node.new('bad', doc) new_node.content = 'clyde' t.replace(new_node) } end # spits out: # "element bad: error : Name is not from the document dictionnary 'bad'" # between 2 and 20 thousand times then dies with: #/opt/ruby-enterprise-1.8.6-20090610/lib/ruby/gems/1.8/gems/nokogiri-1.3.2/lib /nokogiri/xml/document.rb:104: [BUG] object allocation during garbage collection phase #ruby 1.8.6 (2008-08-11) [i686-linux] # #Aborted # setting GC_HACK = true will still give the warning but not # crash (or at least hasn't crashed yet at 250000 cycles :) ) # system: # ruby 1.8.6 (2008-08-11 patchlevel 287) [i686-linux] # Ruby Enterprise Edition 20090610 # libxml 2.7.3 # nokogiri (1.3.2) # gentoo (linux 2.6.29 SMP PREEMPT) # does not seem to happen on macos # might be a REE bug
or cancel
```
 require 'nokogiri'

 GC_HACK = false

 GC.disable if GC_HACK # will delay "error : Name is not from the document dictionnary"
 gc_count = 0
 cycles = 0
 loop do 
 cycles = cycles + 1
 if GC_HACK
 if gc_count > 10000
 GC.enable
 GC.start
 p "gc start cycles: #{cycles}"
 sleep 10
 gc_count = 0
 GC.disable
 end
 gc_count = gc_count + 1
 end
 p "cycles: #{cycles}" if cycles%1000 == 0

 doc = Nokogiri::XML::Document.parse("<bad>blinky</bad>")
 doc.xpath('/bad').each{ |t|
 new_node = Nokogiri::XML::Node.new('bad', doc)
 new_node.content = 'clyde'
 t.replace(new_node)
 }
 end

 # spits out: 
 # "element bad: error : Name is not from the document dictionnary 'bad'"
 # between 2 and 20 thousand times then dies with:

 #/opt/ruby-enterprise-1.8.6-20090610/lib/ruby/gems/1.8/gems/nokogiri-1.3.2/lib /nokogiri/xml/document.rb:104: [BUG] object allocation during garbage collection phase
 #ruby 1.8.6 (2008-08-11) [i686-linux]
 #
```

Comments

1.4.1x tenderlovex Ancestor search doesn't work with a css query.

Comments

1.4.1x tenderlovex Add the ":self" psuedo selector

Comments

Comments

Comments

Comments

Comments

Comments

1.4.1x tenderlovex Nokogiri::XML ignore my set encoding

Comments

Comments

1.4.0x Segmentation Fault on modified re-raise

Comments

Comments

Comments

c:0001 p:0000 s:0002 b:0002 l:0011a4 d:0011a4 TOP

Comments

Comments

Comments

Comments

Comments

REEx tcmalloc error

Comments

REEx extconf.rb have_func() always fails under Ruby Enterprise build system

Comments

1.4.0x Segfaults / memory corruption

Comments

ffix jrubyx Support for JRuby on Windows

Comments

1.4.0x tenderlovex inner_html= dropping some elements

Comments

Comments

1.4.0x tenderlovex Adding a node with a default namespace stores it as 'no-namespace' in the parent

Comments

Comments

1.4.0x tenderlovex Segfault when searching with #at (1.3.3)

Comments

1.4.0x tenderlovex NodeSet#slice doesn't handle ranges beyond the end of the array

Comments

Comments

1.4.0x tenderlovex 1.3.3 breaks RELAXNG.new(rng).validate(xml)

Below is a trivial test to repro the problem

Comments

Comments

tenderlovex implement Nokogiri::XML::DTD#external_id and system_id

Comments

1.4.0x tenderlovex Implement Nokogiri::XML::ElementDecl#content

Comments

Comments

Comments

1.4.0x tenderlovex DocumentFragment lacks detailed search

Comments

Comments

REEx tcmalloc error parsing "<a><b></a>" fragment with REE

Comments

Comments

Comments

1.4.0x tenderlovex Convert meta_encoding and meta_encoding= to ruby

Comments

1.4.0x tenderlovex Shouldn't inner_html convert to UTF8 the same way as inner_text?

Comments

Comments

1.4.0x tenderlovex "ArgumentError: NULL pointer given" on calling doc.meta_encoding

Comments

1.4.0x tenderlovex Error parsing provided css_path()

Comments

1.4.0x ffix tenderlovex Fix platform detection code

Comments

1.4.0x ffix tenderlovex Package libxml2 dll's with jruby gem

Comments

Comments

Comments

mdalessiox REEx document.rb:104: [BUG] object allocation during garbage collection phase

1.4.1
x

tenderlove
x

Ancestor search doesn't work with a css query.

1.4.1
x

tenderlove
x

Add the ":self" psuedo selector

1.4.1
x

tenderlove
x

Nokogiri::XML ignore my set encoding

1.4.0
x

Segmentation Fault on modified re-raise

REE
x

tcmalloc error

REE
x

extconf.rb have_func() always fails under Ruby Enterprise build system

1.4.0
x

Segfaults / memory corruption

ffi
x

jruby
x

Support for JRuby on Windows

1.4.0
x

tenderlove
x

inner_html= dropping some elements

1.4.0
x

tenderlove
x

Adding a node with a default namespace stores it as 'no-namespace' in the parent

1.4.0
x

tenderlove
x

Segfault when searching with #at (1.3.3)

1.4.0
x

tenderlove
x

NodeSet#slice doesn't handle ranges beyond the end of the array

1.4.0
x

tenderlove
x

1.3.3 breaks RELAXNG.new(rng).validate(xml)

tenderlove
x

implement Nokogiri::XML::DTD#external_id and system_id

1.4.0
x

tenderlove
x

Implement Nokogiri::XML::ElementDecl#content

1.4.0
x

tenderlove
x

DocumentFragment lacks detailed search

REE
x

tcmalloc error parsing "<a><b></a>" fragment with REE

1.4.0
x

tenderlove
x

Convert meta_encoding and meta_encoding= to ruby

1.4.0
x

tenderlove
x

Shouldn't inner_html convert to UTF8 the same way as inner_text?

1.4.0
x

tenderlove
x

"ArgumentError: NULL pointer given" on calling doc.meta_encoding

1.4.0
x

tenderlove
x

Error parsing provided css_path()

1.4.0
x

ffi
x

tenderlove
x

Fix platform detection code

1.4.0
x

ffi
x

tenderlove
x

Package libxml2 dll's with jruby gem

mdalessio
x

REE
x

document.rb:104: [BUG] object allocation during garbage collection phase