=== 1.6.6.2 / 2015-01-23
==== Bug fixes
* Fixed installation issue affecting compiler arguments. (#1230)
=== 1.6.6.1 / 2015-01-22
Note that 1.6.6.0 was not released.
==== Features
* Unified Node and NodeSet implementations of #search, #xpath and #css.
* Added Node#lang and Node#lang=.
* bin/nokogiri passes the URI to parse() if an HTTP URL is given.
* bin/nokogiri now loads ~/.nokogirirc so user can define helper methods, etc.
* bin/nokogiri can be configured to use Pry instead of IRB by adding a couple of lines to ~/.nokogirirc. (#1198)
* bin/nokogiri can better handle urls from STDIN (aiding use of xargs). (#1065)
* JRuby 9K support.
==== Bug fixes
* DocumentFragment#search now matches against root nodes. (#1205)
* (MRI) More fixes related to handling libxml2 parse errors during DocumentFragment#dup. (#1196)
* (JRuby) Builder now handles namespace hrefs properly when there is a default ns. (#1039)
* (JRuby) Clear the XPath cache on attr removal. (#1109)
* `XML::Comment.new` argument types are now consistent and safe (and documented) across MRI and JRuby. (#1224)
* (MRI) Restoring support for Ruby 1.9.2 that was broken in v1.6.4.1 and v1.6.5. (#1207)
* Check if `zlib` is available before building `libxml2`. (#1188)
* (JRuby) HtmlSaxPushParser now exists. (#1147) (Thanks, Piotr Szmielew!)
=== 1.6.5 / 2014-11-26
==== Features
* Implement Slop#respond_to_missing?. (#1176)
* Optimized the XPath query generated by an `an+b` CSS query.
==== Bug fixes
* Capture non-parse errors from Document#dup in Document#errors. (#1196)
* (JRuby) Document#canonicalize parameters are now consistent with MRI. (#1189)
=== 1.6.4.1 / 2014-11-05
==== Bug fixes
* (MRI) Fix a bug where CFLAGS passed in are dropped. (#1188)
* Fix a bug where CSS selector :nth(n) did not work. (#1187)
=== 1.6.4 / 2014-11-04
==== Features
* (MRI) Bundled Libxml2 is upgraded to 2.9.2.
* (MRI) `nokogiri --version` will include a list of applied patches.
* (MRI) Nokogiri no longer prints messages directly to TTY while building the extension.
* (MRI) Detect and help user fix a missing /usr/include/iconv.h on OS X. (#1111)
* (MRI) Improve the iconv detection for building libxml2.
==== Bug fixes
* (MRI) Fix DocumentFragment#element_children (#1138).
* Fix a bug with CSS attribute selector without any prefix where "foo [bar]" was treated as "foo[bar]". (#1174)
=== 1.6.3.1 / 2014-07-21
==== Bug fixes
* Addressing an Apple Macintosh installation problem for GCC users. #1130 (Thanks, @zenspider!)
=== 1.6.3 / 2014-07-20
==== Features
* Added Node#document? and Node#processing_instruction?
==== Bug fixes
* [JRuby] Fix Ruby memory exhaustion vulnerability. #1087 (Thanks, @ocher)
* [MRI] Fix segfault during GC when using `libxml-ruby` and `nokogiri` together in multi-threaded environment. #895 (Thanks, @ender672!)
* Building on OSX 10.9 stock ruby 2.0.0 now works. #1101 (Thanks, @zenspider!)
* Node#parse now works again for HTML document nodes (broken in 1.6.2+).
* Processing instructions can now be added via Node#add_next_sibling.
=== 1.6.2.1 / 2014-05-13
==== Bug fixes
* Fix statically-linked libxml2 installation when using universal builds of Ruby. #1104
* Patching `mini_portile` to address the git dependency detailed in #1102.
* Library load fix to address segfault reported on some systems. #1097
=== 1.6.2 / 2014-05-12
==== Security Note
A set of security and bugfix patches have been backported from the libxml2 and libxslt repositories onto the version of 2.8.0 packaged with Nokogiri, including these notable security fixes:
* https://git.gnome.org/browse/libxml2/commit/?id=4629ee02ac649c27f9c0cf98ba017c6b5526070f
* CVE-2013-2877 https://git.gnome.org/browse/libxml2/commit/?id=e50ba8164eee06461c73cd8abb9b46aa0be81869
* CVE-2014-0191 https://git.gnome.org/browse/libxml2/commit/?id=9cd1c3cfbd32655d60572c0a413e017260c854df
It is recommended that you upgrade from 1.6.x to this version as soon as possible.
==== Compatibility Note
Now requires libxml >= 2.6.21 (was previously >= 2.6.17).
==== Features
* Add cross building of fat binary gems for 64-Bit Windows (x64-mingw32) and add support for native builds on Windows. #864, #989, #1072
* (MRI) Alias CP932 to Windows-31J if iconv does not support Windows-31J.
* (MRI) Nokogiri now links packaged libraries statically. To disable static linking, pass --disable-static to extconf.rb. #923
* (MRI) Fix a library path (LIBPATH) precedence problem caused by CRuby bug #9760.
* (MRI) Nokogiri automatically deletes directories of packaged libraries only used during build. To keep them for debugging purposes, pass --disable-clean to extconf.rb. #952
* (MRI) Nokogiri now builds libxml2 properly with iconv support on platforms where libiconv is installed outside the system default directories, such as FreeBSD.
* Add support for an-b in nth selectors. #886 (Thanks, Magnus Bergmark!)
* Add support for bare and multiple :not() functions in selectors. #887 (Thanks, Magnus Bergmark!)
* (MRI) Add an extconf.rb option --use-system-libraries, alternative to setting the environment variable NOKOGIRI_USE_SYSTEM_LIBRARIES.
* (MRI) Update packaged libraries: libxslt to 1.1.28, zlib to 1.2.8, and libiconv to 1.14, respectively.
* Nokogiri::HTML::Document#title= and #meta_encoding= now always add an element if not present, trying hard to find the best place to put it.
* Nokogiri::XML::DTD#html_dtd? and #html5_dtd? are added.
* Nokogiri::XML::Node#prepend_child is added. #664
* Nokogiri::XML::SAX::ParserContext#recovery is added. #453
* Fix documentation for XML::Node#namespace. #803#802 (Thanks, Hoylen Sue)
* Allow Nokogiri::XML::Node#parse from unparented non-element nodes. #407
==== Bugfixes
* Ensure :only-child pseudo class works within :not pseudo class. #858 (Thanks, Yamagishi Kazutoshi!)
* Don't call pkg_config when using bundled libraries in extconf.rb #931 (Thanks, Shota Fukumori!)
* Nokogiri.parse() does not mistake a non-HTML document like a RSS document as HTML document. #932 (Thanks, Yamagishi Kazutoshi!)
* (MRI) Perform a node type check before adding a child node to another. Previously adding a text node to another as a child could cause a SEGV. #1092
* (JRuby) XSD validation crashes in Java version. #373
* (JRuby) Document already has a root node error while using Builder. #646
* (JRuby) c14n tests are all passing on JRuby. #226
* Parsing empty documents raise SyntaxError in strict mode. #1005
* (JRuby) Make xpath faster by caching the xpath context. #741
* (JRuby) XML SAX push parser leaks memory on JRuby, but not on MRI. #998
* (JRuby) Inconsistent behavior aliasing the default namespace. #940
* (JRuby) Inconsistent behavior between parsing and adding namespaces. #943
* (JRuby) Xpath returns inconsistent result set on cloned document with namespaces and attributes. #1034
* (JRuby) Java-Implementation forgets element namespaces #902
* (JRuby) JRuby-Nokogiri does not recognise attributes inside namespaces #1081
* (JRuby) JRuby-Nokogiri has different comment node name #1080
* (JRuby) JAXPExtensionsProvider / Java 7 / Secure Processing #1070
1.6.0 / 2013-06-08
This release was based on v1.5.10 and 1.6.0.rc1, and contains changes
mentioned in both.
Deprecations
Remove pre 1.9 monitoring from Travis.
1.6.0.rc1 / 2013-04-14
This release was based on v1.5.9, and so does not contain any fixes mentioned
in the notes for v1.5.10.
Notes
mini_portile is now a runtime dependency
Ruby 1.9.2 and higher now required
Features
(MRI) Source code for libxml 2.8.0 and libxslt 1.2.26 is packaged with the
gem. These libraries are compiled at gem install time unless the environment
variable NOKOGIRI_USE_SYSTEM_LIBRARIES is set. VERSION_INFO (also `nokogiri
-v`) exposes whether libxml was compiled from packaged source, or the system
library was used.
(Windows) libxml upgraded to 2.8.0
Deprecations
Support for Ruby 1.8.7 and prior has been dropped
=== 1.5.10 / 2013-06-07
* Bugfixes
* (JRuby) Fix "null document" error when parsing an empty IO in jruby 1.7.3. #883
* (JRuby) Fix schema validation when XSD has DOCTYPE set to DTD. #861 (Thanks, Patrick Cheng!)
* (MRI) Fix segfault when there is no default subelement for an HTML node. #917
* Notes
* Use rb_ary_entry instead of RARRAY_PTR (you know, for Rubinius). #877 (Thanks, Dirkjan Bussink!)
* Fix TypeError when running tests. #900 (Thanks, Cédric Boutillier!)
=== 1.5.9 / 2013-03-21
* Bugfixes
* Ensure that prefixed attributes are properly namespaced when reparented. #869
* Fix for inconsistent namespaced attribute access for SVG nested in HTML. #861
* (MRI) Fixed a memory leak in fragment parsing if nodes are not all subsequently reparented. #856
=== 1.5.8 / 2013-03-19
* Bugfixes
* (JRuby) Fix EmptyStackException thrown by elements with xlink:href attributes and no base_uri #534, #805. (Thanks, Patrick Quinn and Brian Hoffman!)
* Fixes duplicate attributes issue introduced in 1.5.7. #865
* Allow use of a prefixed namespace on a root node using Nokogiri::XML::Builder #868
=== 1.5.7 / 2013-03-18
* Features
* Windows support for Ruby 2.0.
* Bugfixes
* SAX::Parser.parse_io throw an error when used with lower case encoding. #828
* (JRuby) Java Nokogiri is finally green (passes all tests) under 1.8 and 1.9 mode. High five everyone. #798, #705
* (JRuby) Nokogiri::XML::Reader broken (as a pull parser) on jruby - reads the whole XML document. #831
* (JRuby) JRuby hangs parsing "&". #837
* (JRuby) JRuby NPE parsing an invalid XML instruction. #838
* (JRuby) Node#content= incompatibility. #839
* (JRuby) to_xhtml doesn't print the last slash for self-closing tags in JRuby. #834
* (JRuby) Adding an EntityReference after a Text node mangles the entity in JRuby. #835
* (JRuby) JRuby version inconsistency: nil for empty attributes. #818
* CSS queries for classes (e.g., ".foo") now treat all whitespace identically. #854
* Namespace behavior cleaned up and made consistent between JRuby and MRI. #846, #801 (Thanks, Michael Klein!)
* (MRI) SAX parser handles empty processing instructions. #845
== 1.5.6 / unreleased
* Features
* Improved performance of XML::Document#collect_namespaces. #761 (Thanks, Juergen Mangler!)
* New callback SAX::Document#processing_instruction (Thanks, Kitaiti Makoto!)
* Node#native_content= allows setting unescaped node contant. #768
* XPath lookup with namespaces supports symbol keys. #729 (Thanks, Ben Langfeld.)
* XML::Node#[]= stringifies values. #729 (Thanks, Ben Langfeld.)
* bin/nokogiri will process a document from $stdin
* bin/nokogiri -e will execute a program from the command line
* bin/nokogiri --version will print the Xerces and NekoHTML versions when ran with JRuby.
* Bugfixes
* Nokogiri now detects XSLT transform errors. #731 (Thanks, Justin Fitzsimmons!)
* Don't throw an Error when trying to replace top-level text node in DocumentFragment. #775
* Raise an ArgumentError if an invalid encoding is passed to the SAX parser. #756 (Thanks, Bradley Schaefer!)
* [JRuby] space prior to xml preamble causes nokogiri to fail parsing. (fixed along with #748) #790
* [JRuby] Fixed the bug Nokogiri::XML::Node#content inconsistency between Java and C. #794, #797
* [JRuby] raises INVALID_CHARACTER_ERR exception when EntityReference name starts with '#'. #719
* [JRuby] doesn't coerce namespaces out of strings on a direct subclass of Node. #715
* [JRuby] Node#content now renders newlines properly. #737 (Thanks, Piotr Szmielew!)
* [JRuby] Unknown namespace are ignore when the recover option is used. #748
* [JRuby] XPath queries for namespaces should not throw exceptions when called twice in a row. #764
* [JRuby] More consistent (with libxml2) whitespace formatting when emitting XML. #771
* [JRuby] namespaced attributes broken when appending raw xml to builder. #770
* [JRuby] Nokogiri::XML::Document#wrap raises undefined method `length' for nil:NilClass when trying to << to a node. #781
* [JRuby] Fixed "bad file descriptor" bug when closing open file descriptors. #495
* [JRuby] JRuby/CRuby incompatibility for attribute decorators. #785
* [JRuby] Issues parsing valid XML with no internal subset in the DTD. #547, #811
* [JRuby] Issues parsing valid node content when it contains colons. #728
* [JRuby] Correctly parse the doc type of html documents. #733
* [JRuby] Include dtd in the xml output when a builder is used with create_internal_subset. #751
* [JRuby] builder requires textwrappers for valid utf8 in jruby, not in mri. #784
== 1.5.3 / 2012-06-01
* Features
* Support for "prefixless" CSS selectors ~, > and + like jQuery
supports. #621, #623. (Thanks, David Lee!)
* Attempting to improve installation on homebrew 0.9 (with regards
to iconv). Isn't package management convenient?
* Bugfixes
* Custom xpath functions with empty nodeset arguments cause a
segfault. #634.
* Nokogiri::XML::Node#css now works for XML documents with default
namespaces when the rule contains attribute selector without
namespace.
* Fixed marshalling bugs around how arguments are passed to (and
returned from) XSLT custom xpath functions. #640.
* Nokogiri::XML::Reader#outer_xml is broken in JRuby #617
* Nokogiri::XML::Attribute on JRuby returns a nil namespace #647
* Nokogiri::XML::Node#namespace= cannot set a namespace without a
prefix on JRuby #648
* [JRuby] 1.9 mode causes dead lock while running rake #571
* HTML::Document#meta_encoding does not raise exception on docs with
malformed content-type. #655
* Fixing segfault related to unsupported encodings in in-context
parsing on 1.8.7. #643
* [JRuby] Concurrency issue in XPath parsing. #682
== 1.5.2 / 2012-03-09
Repackaging of 1.5.1 with a gemspec that is compatible with older Rubies. #631, #632.
== 1.5.1 / 2012-03-09
* Features
* XML::Builder#comment allows creation of comment nodes.
* CSS searches now support namespaced attributes. #593
* Java integration feature is added. Now, XML::Document.wrap
and XML::Document#to_java methods are available.
* RelaxNG validator support in the `nokogiri` cli utility. #591 (thanks, Dan Radez!)
* Bugfixes
* Fix many memory leaks and segfault opportunities. Thanks, Tim Elliott!
* extconf searches homebrew paths if homebrew is installed.
* Inconsistent behavior of Nokogiri 1.5.0 Java #620
* Inheriting from Nokogiri::XML::Node on JRuby (1.6.4/5) fails #560
* XML::Attr nodes are not allowed to be added as node children, so an
exception is raised. #558
* No longer defensively "pickle" adjacent text nodes on
Node#add_next_sibling and Node#add_previous_sibling calls. #595.
* Java version inconsistency: it returns nil for empty attributes #589
* to_xhtml incorrectly generates <p /></p> when tag is empty #557
* Document#add_child now accepts a Node, NodeSet, DocumentFragment,
or String. #546.
* Document#create_element now recognizes namespaces containing
non-word characters (like "SOAP-ENV"). This is mostly relevant to
users of Builder, which calls Document#create_element for nearly
everything. #531.
* File encoding broken in 1.5.0 / jruby / windows #529
* Java version does not return namespace defs as attrs for ::HTML #542
* Bad file descriptor with Nokogiri 1.5.0 #495
* remove_namespace! doesn't work in pure java version #492
* The Nokogiri Java native build throws a null pointer exception
when ActiveSupport's .blank? method is called directly on a parsed
object. #489
* 1.5.0 Not using correct character encoding #488
* Raw XML string in XML Builder broken on JRuby #486
* Nokogiri 1.5.0 XML generation broken on JRuby #484
* Do not allow multiple root nodes. #550
* Fixes for custom XPath functions. #605, #606 (thanks, Juan Wajnerman!)
* Node#to_xml does not override :save_with if it is provided. #505
* Node#set is a private method [JRuby]. #564 (thanks, Nick Sieger!)
* C14n cleanup and Node#canonicalize (thanks, Ivan Pirlik!) #563
* It fixes a critical bug in 1.4.6.
== 1.5.0 / 2011-07-01
* Notes
* See changelog from 1.4.7
* Features
* extracted sets of Node::SaveOptions into Node::SaveOptions::DEFAULT_{X,H,XH}TML (refactor)
* Bugfixes
* default output of XML on JRuby is no longer formatted due to
inconsistent whitespace handling. #415
* (JRuby) making empty NodeSets with null `nodes` member safe to operate on. #443
* Fix a bug in advanced encoding detection that leads to partially
duplicated document when parsing an HTML file with unknown
encoding.
* Add support for <meta charset="...">.
== 1.5.0 beta3 / 2010/12/02
* Notes
* JRuby performance tuning
* See changelog from 1.4.4
* Bugfixes
* Node#inner_text no longer returns nil. (JRuby) #264
== 1.5.0 beta2 / 2010/07/30
* Notes
* See changelog from 1.4.3
== 1.5.0 beta1 / 2010/05/22
* Notes
* JRuby support is provided by a new pure-java backend.
* Deprecations
* Ruby 1.8.6 is deprecated. Nokogiri will install, but official support is ended.
* LibXML 2.6.16 and earlier are deprecated. Nokogiri will refuse to install.
* FFI support is removed.
=== 1.4.7 / 2011-07-01
* Bugfixes
* Fix a bug in advanced encoding detection that leads to partially
duplicated document when parsing an HTML file with unknown
encoding. Thanks, Timothy Elliott (@ender672)! #478
=== 1.4.6 / 2011-06-19
* Notes
* This version is functionally identical to 1.4.5.
* Ruby 1.8.6 support has been restored.
=== 1.4.5 / 2011-06-15
* New Features
* Nokogiri::HTML::Document#title accessor gets and sets the document title.
* extracted sets of Node::SaveOptions into
Node::SaveOptions::DEFAULT_{X,H,XH}TML (refactor)
* Raise an exception if a string is passed to Nokogiri::XML::Schema#validate.
#406
* Bugfixes
* Node#serialize-and-friends now accepts a SaveOption object as the, erm,
save object.
* Nokogiri::CSS::Parser has-a Nokogiri::CSS::Tokenizer
* [JRUBY+FFI only] Weak references are now threadsafe. #355
* Make direct start_element() callback (currently used for
HTML::SAX::Parser) pass attributes in assoc array, just as
emulated start_element() callback does. rel. #356
* HTML::SAX::Parser should call back a block given to parse*() if any, just
as XML::SAX::Parser does.
* Add further encoding detection to HTML parser that libxml2 does not do.
* Document#remove_namespaces! now handles attributes with namespaces. #396
* XSLT::Stylesheet#transform no longer segfaults when handed a
non-XML::Document. #452
* XML::Reader no longer segfaults when under GC pressure. #439
=== 1.4.4 / Unreleased
* New Features
* XML::Node#children= sets the node's inner html (much like #inner_html=), but returns the reparent node(s).
* XSLT supports function extensions. #336
* XPath bind parameter substitution. #329
* XML::Reader node type constants. #369
* SAX Parser context provides line and column information
* Bugfixes
* XML::DTD#attributes returns an empty hash instead of nil when there are no attributes.
* XML::DTD#{keys,each} now work as expected. #324
* {XML,HTML}::DocumentFragment.{new,parse} no longer strip leading and trailing whitespace. #319
* XML::Node#{add_child,add_previous_sibling,add_next_sibling,replace} return a NodeSet when passed a string.
* Unclosed tags parsed more robustly in fragments. #315
* XML::Node#{replace,add_previous_sibling,add_next_sibling} edge cases fixed related to libxml's text node merging. #308
* Fixed a segfault when GC occurs during xpath handler argument marshalling. #345
* Added hack to Slop decorator to work with previously defined methods. #330
* Fix a memory leak when duplicating child nodes. #353
* Fixed off-by-one bug with nth-last-{child,of-type} CSS selectors when NOT using an+b notation. #354
* Fixed passing of non-namespace attributes to SAX::Document#start_element. #356
* Workaround for libxml2 in-context parsing bug. #362
* Fixed NodeSet#wrap on nodes within a fragment. #331
* Use lang/ruby/gem.mk instead of misc/rubygems/rubygem.mk.
* Update HOMEPAGE.
* Add CONFLICTS since it will install the same name commands both
ruby18 and ruby19 based packages.
Changes are too many to write here, please refer CHANGELOG.rdoc.
Nokogiri parses and searches XML/HTML very quickly, and also has correctly
implemented CSS3 selector support as well as XPath support.
Features:
* XPath support for document searching
* CSS3 selector support for document searching
* XML/HTML builder
* Drop in replacement for Hpricot (though not bug for bug)
