Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
tag: v1.4.5
Commits on Jun 16, 2011
  1. Mike Dalessio

    version bump to 1.4.5. also updating changelog release date to reflec…

    flavorjones authored
    …t my recent inability to concentrate.
  2. Mike Dalessio

    Updating Manifest.txt

    flavorjones authored
  3. Mike Dalessio
  4. Mike Dalessio

    Revert "rake-compiler's tendency to chdir means we should use expand_…

    flavorjones authored
    …path when adding files to the gem."
    
    This reverts commit a330f14.
Commits on Jun 4, 2011
  1. Mike Dalessio
  2. Mike Dalessio
  3. Mike Dalessio

    hoe's gemtest obsession and rake-compiler's chdir tendencies don't ag…

    flavorjones authored
    …ree, so I'm just going to give up and add `.gemtest` to the repo.
  4. Mike Dalessio

    If anyone actually uses rip, please to explain.

    flavorjones authored
    Conflicts:
    
    	Manifest.txt
  5. Mike Dalessio
Commits on Jun 3, 2011
  1. Mike Dalessio
  2. Mike Dalessio
Commits on May 24, 2011
  1. Mike Dalessio
Commits on May 21, 2011
  1. Mike Dalessio

    FFI port of cf062cb

    flavorjones authored
Commits on May 20, 2011
  1. Mike Dalessio
Commits on May 19, 2011
  1. Mike Dalessio
  2. Mike Dalessio
  3. Mike Dalessio

    FFI port of a0905d3 for #452.

    flavorjones authored
  4. Mike Dalessio

    getting ready for release

    flavorjones authored
  5. Aaron Patterson Mike Dalessio
  6. Aaron Patterson Mike Dalessio

    just yield rather than calling the block

    tenderlove authored flavorjones committed
  7. Mike Dalessio

    If the node doesn't have a document (e.g., Reader), then don't bother…

    flavorjones authored
    … registering a gc mark function. Partly fixes #439.
  8. Mike Dalessio
Commits on May 18, 2011
  1. Mike Dalessio

    changelog for #406

    flavorjones authored
  2. Nathan Long Mike Dalessio

    Nokogiri::XML::Schema#validate throws error if not given a

    nathanl authored flavorjones committed
    Nokogiri::XML::Document or the path of an existing file.
    
    Previously, passing in any other string would fail silently. For example, if
    rendered XML was passed in as a string, it would simply return an empty
    list of errors. Now an ArgumentError will be thrown.
Commits on May 4, 2011
  1. Mike Dalessio
Commits on Apr 6, 2011
  1. Mike Dalessio
Commits on Mar 9, 2011
  1. Mike Dalessio

    extracted sets of Node::SaveOptions into Node::SaveOptions::DEFAULT_{…

    flavorjones authored
    …X,H,XH}TML
    
    Conflicts:
    
    	CHANGELOG.rdoc
Commits on Mar 4, 2011
  1. Mike Dalessio
Commits on Feb 27, 2011
  1. Timothy Elliott Mike Dalessio

    Prevent buffer overflow in xml_io.c when io return string is too long

    ender672 authored flavorjones committed
    If an IO object returns a string that is longer than expected, io_read_callback() writes past the limit of the buffer and crashes the interpreter. This patch checks against the size of our buffer before copying the data over.
    
    I can't really think of a case where an IO object would do something this silly, but seems like a good practice anyways.
  2. Timothy Elliott Mike Dalessio

    Fix Memory Leak in xml_io.c

    ender672 authored flavorjones committed
    This example will use up all system memory:
    
      require 'nokogiri'
    
      class BadIO
        def read(*args)
          raise 'hell'
        end
      end
    
      100000000.times do
        Nokogiri::XML.parse(BadIO.new) rescue nil
      end
    
    The example above is obscure, but imagine an HTTP object raising TimeoutError. Nokogiri leaks memory every time this happens.
    
    I became suspicious of this when looking at ext/nokogiri/xml_io.c. io_read_callback() is invoked in the middle of some libxml operation, which may not behave well when a ruby exception is thrown and it gets interrupted. It seems that the safe way to handle this is to wrap the method call in rb_rescue.
    
    Note that the same applies to io_write_callback() in xml_io.c. Although I did not notice a memory leak with io_write_callback(), wrapping with rb_rescue() seems like the correct thing to do in both cases.
    
    A side effect of this fix is that IO errors will be silently discarded.
Commits on Feb 24, 2011
  1. Hampton Catlin Mike Dalessio

    Allow for node.search("..") to be executed as XPath

    hcatlin authored flavorjones committed
    instead of being converted to CSS. This seems like a reasonable
    bit of XPath and our users have tried it many, many times.
    
    Might want to consider search(".") as XPath instead of CSS
    also. However, this was left out.
Commits on Jan 26, 2011
  1. G. Ferguson Aaron Patterson

    "then" should be "the"

    gpherguson authored tenderlove committed
  2. G. Ferguson Aaron Patterson

    Minor spell check "your" should have been "you"

    gpherguson authored tenderlove committed
Commits on Jan 25, 2011
  1. Mike Dalessio

    FFI port of 45e6694 (rel. #356)

    flavorjones authored
  2. Mike Dalessio

    FFI port of 8c2a112 for #396

    flavorjones authored
Something went wrong with that request. Please try again.