Skip to content

Latest commit

 

History

History
2657 lines (1596 loc) · 131 KB

CHANGELOG.md

File metadata and controls

2657 lines (1596 loc) · 131 KB
Raw

Nokogiri Changelog

Nokogiri follows Semantic Versioning, please see the README.md for details.

v1.next / unreleased

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.13.4. @flavorjones
  • [CRuby] Vendored libxslt is updated to v1.1.42. @flavorjones
  • [CRuby] Minimum supported version of libxml2 raised to v2.9.2 (released 2014-10-16) from v2.6.21. [#3232, #3287] @flavorjones
  • [JRuby] Minimum supported versino of Java raised to 8 (released 2014-03-18) from 7. [#3134] @flavorjones
  • [CRuby] Update to rake-compiler-dock v1.5.1 for building precompiled native gems. [#3216] @flavorjones

Notable changes

SAX Parsers

The XML and HTML4 SAX parsers have received a lot of attention in this release, and we've fixed multiple long-standing bugs with encoding and entity handling. In addition, libxml2 v2.13 has also made some underlying fixes and improvements to encoding and entity handling.

We're shipping these fixes in a minor release because we firmly believe the resulting behavior is correct and standards-compliant, however applications that have been depending on the buggy behavior may be impacted.

If your application relies on the SAX parsers, and in particular if you're SAX-parsing documents with parsed entities or incorrect encoding declarations, please read the changelog below carefully.

Fragment parsing

Document fragment parsing has been improved, particularly with respect to handling malformed fragments or fragments with implicit namespace prefixes. Namespace reconciliation still isn't where we want it to be, but it's an improvement.

HTML5 fragment parsing now allows the context node to be specified as a keyword argument to the HTML5::DocumentFragment.parse and .new methods, which in particular should allow for more flexible sanitization and support for the draft HTML Sanitizer API in downstream libraries.

Error handling

In scenarios where multiple errors could be reported by the underlying parser, the errors will be aggregated into a single Nokogiri::XML::SyntaxError that is raised. Previously only the final error reported by libxml2 was raised which was often misleading if it was only a warning and not the fatal error.

Schema validation

We've resolved many long-standing bugs in the various schema classes, validation methods, and their error reporting. Behavior is now consistent across schema types and input types, as well as parser backends (Xerces and libxml2).

Added

  • Introduce support for a new SAX callback XML::SAX::Document#reference, which is called to report some parsed XML entities when XML::SAX::ParserContext#replace_entities is set to the default value false. This is necessary functionality for some applications that were previously relying on incorrect entity error reporting which has been fixed (see below). For more information, read the docs for Nokogiri::XML::SAX::Document. [#1926] @flavorjones
  • XML::SAX::Parser#parse_memory and #parse_file now accept an optional encoding argument. When not provided, the parser will fall back to the encoding passed to the initializer, and then fall back to autodetection. [#3288] @flavorjones
  • XML::SAX::ParserContext.memory now accepts an optional encoding argument. When not provided, the encoding will be autodetected. [#3288] @flavorjones
  • New attributes XML::DocumentFragment#parse_options and HTML4::DocumentFragment#parse_options contain the options used to parse the document fragment. @flavorjones
  • [CRuby] Nokogiri::HTML5::Builder is similar to HTML4::Builder but returns an HTML5::Document. [#3119] @flavorjones
  • [CRuby] Attributes in an HTML5 document can be serialized individually, something that has always been supported by the HTML4 serializer. [#3125, #3127] @flavorjones
  • [CRuby] Introduce a compile-time option, --disable-xml2-legacy, to remove from libxml2 its dependencies on zlib and liblzma and disable implicit HTTP network requests. These all remain enabled by default, and are present in the precompiled native gems. This option is a precursor for removing these libraries in a future major release, but may be interesting for the security-minded who do not need features like automatic decompression and would like to remove these dependencies. You can read more and give feedback on these plans in #3168. [#3247] @flavorjones
  • [CRuby] If errors are returned from schema validation, a new attribute SyntaxError#path will contain the XPath path of the node that caused the validation failure. [#3316] @ryanong

Improved

  • Documentation has been improved for CSS.xpath_for. [#3224] @flavorjones
  • Documentation for the SAX parsing classes has been greatly improved, including encoding overrides and the complex entity-handling behavior. [#3265] @flavorjones
  • XML::Schema#read_memory and XML::RelaxNG#read_memory are now Ruby methods that call #from_document. Previously these were native functions, but they were buggy on both CRuby and JRuby (but worse on JRuby) and so this is now useful, comparable in performance, and simpler code that is easier to maintain. [#2113, #2115] @flavorjones
  • XML::SAX::ParserContext.io's encoding argument is now optional, and can now be an Encoding or an encoding name. When not provided will default to autodetecting the encoding. [#3288] @flavorjones
  • [CRuby] When compiling packaged libraries from source, allow users' AR and LD environment variables to set the archiver and linker commands, respectively. This augments the existing CC environment variable to set the compiler command. [#3165] @ziggythehamster
  • [CRuby] The HTML5 parse methods accept a :parse_noscript_content_as_text keyword argument which will emulate the parsing behavior of a browser which has scripting enabled. [#3178, #3231] @stevecheckoway
  • [CRuby] HTML5::DocumentFragment.parse and .new accept a :context keyword argument that is the parse context node or element name. Previously this could only be passed in as a positional argument to .new and not at all to .parse. @flavorjones
  • [CRuby] The update to libxml v2.13 improves "in context" fragment parsing recovery. We removed our hacky workaround for recovery that led to silently-degraded functionality when parsing fragments with parse errors. Specifically, malformed XML fragments that used implicit namespace prefixes will now "link up" to the namespaces in the parent document or node, where previously they did not. [#2092] @flavorjones
  • [CRuby] When multiple errors could be detected by the parser and there's no obvious document to save them in (for example, when parsing a document with the recovery parse option turned off), the libxml2 errors are aggregated into a single Nokogiri::XML::SyntaxError. Previously, only the last error recorded by libxml2 was raised, which might be misleading if it's merely a warning and not the fatal error preventing the operation. [#2562] @flavorjones
  • [CRuby] The SAX parser context and handler implementation has been simplified and now takes advantage of some of libxml2's default SAX handlers for entities and DTD management. [#3265] @flavorjones

Fixed

  • Node#clone, NodeSet#clone, and *::Document#clone all properly copy the metaclass of the original as expected. Previously, #clone had been aliased to #dup for these classes (since v1.3.0 in 2009). [#316, #3117] @flavorjones
  • CSS queries for pseudo-selectors that cannot be translated into XPath expressions now raise a more descriptive Nokogiri::CSS::SyntaxError when they are parsed. Previously, an invalid XPath expression was evaluated and a hard-to-understand XPath error was raised by the query engine. [#3193] @flavorjones
  • Schema#validate returns errors on empty and malformed files. Previously, it would return errors on empty/malformed Documents, but not when reading from files. [#642] @flavorjones
  • XML::Builder is now consistent with how it sets block scope. Previously, missing methods with blocks on dynamically-created nodes were always handled by invoking instance_eval(&block) on the Builder, even when the Builder was yielding self for all other missing methods with blocks. [#1041] @flavorjones
  • HTML4::DocumentFragment.parse accepts IO input. Previously, it required a string and would raise a TypeError when passed an IO. [#2069] @sharvy
  • [CRuby] libgumbo (the HTML5 parser) treats reaching max-depth as EOF. This addresses a class of issues when the parser is interrupted in this way. [#3121] @stevecheckoway
  • [CRuby] Update node GC lifecycle to avoid a potential memory leak with fragments in libxml 2.13.0 caused by changes in xmlAddChild. [#3156] @flavorjones
  • [CRuby] libgumbo correctly prints nonstandard element names in error messages. [#3219] @stevecheckoway
  • [CRuby] SAX parsing no longer registers errors when encountering external entity references. [#1926] @flavorjones
  • [JRuby] Fixed entity reference serialization, which rendered both the reference and the replacement text. Incredibly nobody noticed this bug for over a decade. [#3272] @flavorjones
  • [JRuby] Fixed some bugs in how Node#attributes handles attributes with namespaces. [#2677, #2679] @flavorjones
  • [JRuby] Fix Schema#validate to only return the most recent Document's errors. Previously, if multiple documents were validated, this method returned the accumulated errors of all previous documents. [#1282] @flavorjones
  • [JRuby] Fix Schema#validate to not clobber the @errors instance variable. [#1282] @flavorjones
  • [JRuby] Empty documents fail schema validation as they should. [#783] @flavorjones
  • [JRuby] SAX parsing now respects the #replace_entities attribute, which defaults to false. Previously this flag defaulted to true and was completely ignored. [#614] @flavorjones
  • [JRuby] The SAX callback Document#start_element_namespace received a blank string for the URI when a namespace was not present. It now receives nil (as does the CRuby impl). [#3265] @flavorjones
  • [JRuby] Reader#outer_xml and #inner_xml encode entities properly. [#1523] @flavorjones

Changed

  • [CRuby] Nokogiri::XML::CData.new no longer accepts nil as the content argument, making CData behave like other character data classes (like Comment and Text). This change was necessitated by behavioral changes in the upcoming libxml 2.13.0 release. If you wish to create an empty CDATA node, pass an empty string. [#3156] @flavorjones
  • Internals:
    • The internal CSS::XPathVisitor class now accepts the xpath prefix and the context namespaces as constructor arguments. The prefix: and ns: keyword arguments to CSS.xpath_for cannot be specified if the visitor: keyword argument is also used. CSS::XPathVisitor now exposes #builtins, #doctype, #prefix, and #namespaces attributes. [#3225] @flavorjones
    • The internal CSS selector cache has been extracted into a distinct class, CSS::SelectorCache. Previously it was part of the CSS::Parser class. [#3226] @flavorjones
    • The internal Gumbo.parse and Gumbo.fragment methods now take keyword arguments instead of positional arguments. [#3199] @flavorjones

Deprecated

  • The undocumented and unused method Nokogiri::CSS.parse is now deprecated and will generate a warning. The AST returned by this method is private and subject to change and removal in future versions of Nokogiri. This method will be removed in a future version of Nokogiri.
  • Passing an options hash to CSS.xpath_for is now deprecated and will generate a warning. Use keyword arguments instead. This will become an error in a future version of Nokogiri.
  • Passing an options hash to HTML5::DocumentFragment.parse is now deprecated and will generate a warning. Use keyword arguments instead. This will become an error in a future version of Nokogiri.
  • Passing libxml2 encoding IDs to SAX::ParserContext methods is now deprecated and will generate a warning. The use of SAX::Parser::ENCODINGS is also deprecated. Use Encoding objects or encoding names instead.

v1.16.7 / 2024-07-27

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.9, which the upstream release notes state is a security release to address CVE-2024-40896. Nokogiri's maintainers believe this vulnerability does not affect users of Nokogiri, but we advise upgrading at your earliest convenience anyway.

v1.16.6 / 2024-06-13

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.8, which the release notes state is a bugfix release.

v1.16.5

Security

  • [CRuby] Vendored libxml2 is updated to address CVE-2024-34459. See GHSA-r95h-9x8f-r3f7 for more information.

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.7 from v2.12.6. (@flavorjones)

v1.16.4 / 2024-04-10

Dependencies

  • [CRuby] Vendored zlib in the precompiled native gems is updated to v1.3.1 from v1.3. Nokogiri is not affected by the minizip CVE patched in this version, but this update may satisfy some security scanners. Related, see this discussion about removing the compression libraries altogether in a future version of Nokogiri.

v1.16.3 / 2024-03-15

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.6 from v2.12.5. (@flavorjones)

Changed

  • [CRuby] XML::Reader sets the @encoding instance variable during reading if it is not passed into the initializer. Previously, it would remain nil. The behavior of Reader#encoding has not changed. This works around changes to how libxml2 reports the encoding used in v2.12.6.

v1.16.2 / 2024-02-04

Security

  • [CRuby] Vendored libxml2 is updated to address CVE-2024-25062. See GHSA-xc9x-jj77-9p9j for more information.

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.5 from v2.12.4. (@flavorjones)

v1.16.1 / 2024-02-03

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.12.4 from v2.12.3. (@flavorjones)

Fixed

  • [CRuby] XML::Reader defaults the encoding to UTF-8 if it's not specified in either the document or as a method parameter. Previously non-ASCII characters were serialized as NCRs in this case. [#2891] (@flavorjones)
  • [CRuby] Restored support for compilation by GCC versions earlier than 4.6, which was broken in v1.15.0 (540e9aee). [#3090] (@adfoster-r7)
  • [CRuby] Patched upstream libxml2 to allow parsing HTML5 in the context of a namespaced node (e.g., foreign content like MathML). [#3112, #3116] (@flavorjones)
  • [CRuby] Fixed a small memory leak in libgumbo (HTML5 parser) when the maximum tree depth limit is hit. [#3098, #3100] (@stevecheckoway)

v1.16.0 / 2023-12-27

Notable Changes

Ruby

This release introduces native gem support for Ruby 3.3.

This release ends support for Ruby 2.7, for which upstream support ended 2023-03-31.

Pattern matching

This version marks official support for the pattern matching API in XML::Attr, XML::Document, XML::DocumentFragment, XML::Namespace, XML::Node, and XML::NodeSet (and their subclasses), originally introduced as an experimental feature in v1.14.0. (@flavorjones)

Documentation on what can be matched:

Dependencies

Fixed

  • CSS nth pseudo-classes now handle spaces, e.g. "2n + 1". [#3018] (@fusion2004)
  • [CRuby] libgumbo no longer leaks memory when an incomplete tag is abandoned by the HTML5 parser. [#3036] (@flavorjones)

Removed

  • Removed Nokogiri::HTML5.get which was deprecated in v1.12.0. [#2278] (@flavorjones)
  • Removed the CSS-to-XPath utility modules XPathVisitorAlwaysUseBuiltins and XPathVisitorOptimallyUseBuiltins, which were deprecated in v1.13.0 in favor of XPathVisitor constructor args. [#2403] (@flavorjones)
  • Removed XML::Reader#attribute_nodes which was deprecated in v1.13.8 in favor of #attribute_hash. [#2598, #2599] (@flavorjones)
  • [CRuby] Removed the libxml/libxml2_path key from VersionInfo, used in the past for third-party library integration, in favor of the nokogiri/cppflags and nokogiri/ldflags keys. Please note that third-party library integration is not fully supported and may be deprecated soon, see #2746 for more context. [#2143] (@flavorjones)

Thank you!

The following people and organizations were kind enough to sponsor @flavorjones or the Nokogiri project during the development of v1.16.0:

  • Götz Görisch @GoetzGoerisch
  • Airbnb @airbnb
  • Maxime Gauthier @biximilien
  • Renuo AG @renuo
  • YOSHIDA Katsuhiko @kyoshidajp
  • Homebrew @Homebrew
  • Hiroshi SHIBATA @hsbt
  • @zzak
  • Evil Martians @evilmartians
  • Ajaya Agrawalla @ajaya
  • Modern Treasury @Modern-Treasury
  • Danilo Lessa Bernardineli @danlessa
  • matt marques @mestre-dos-magos
  • Quan Nguyen @qu8n
  • Harry Lascelles @hlascelles
  • Oleksandr Tyshchenko @altivi
  • Prowly @prowlycom
  • Better Stack Community @betterstack-community
  • Sentry @getsentry
  • Codecov @codecov
  • Typesense @typesense
  • Roy Boivin II @Yabbo
  • Frank Groeneveld @frenkel

We'd also like to thank @github who donate a ton of compute time for our CI pipelines!

1.15.6 / 2024-03-16

Security

  • [CRuby] Vendored libxml2 is updated to address CVE-2024-25062. See GHSA-xc9x-jj77-9p9j for more information.

Dependencies

1.15.5 / 2023-11-17

Dependencies

1.15.4 / 2023-08-11

Dependencies

Fixed

  • Fixed a typo in a HTML5 parser error message. [#2927] (@anishathalye)
  • [CRuby] ObjectSpace.memsize_of is now safe to call on Documents with complex DTDs. In previous versions, this debugging method could result in a segfault. [#2923, #2924]

1.15.3 / 2023-07-05

Fixed

  • Passing an object that is not a kind of XML::Node as the first parameter to CDATA.new now raises a TypeError. Previously this would result in either a segfault (CRuby) or a Java exception (JRuby). [#2920]
  • Passing an object that is not a kind of XML::Node as the first parameter to Schema.from_document now raises a TypeError. Previously this would result in either a segfault (CRuby) or a Java exception (JRuby). [#2920]
  • [CRuby] Passing an object that is not a kind of XML::Node as the second parameter to Text.new now raises a TypeError. Previously this would result in a segfault. [#2920]
  • [CRuby] Replacing a node's children via methods like Node#inner_html=, #children=, and #replace no longer defensively dups the node's next sibling if it is a Text node. This behavior was originally adopted to work around libxml2's memory management (see #283 and #595) but should not have included operations involving xmlAddChild(). [#2916]
  • [JRuby] Fixed NPE when serializing an unparented HTML node. [#2559, #2895] (@cbasguti)

1.15.2 / 2023-05-24

Dependencies

  • [JRuby] Vendored org.nokogiri:nekodtd is updated to v0.1.11.noko2. This is functionally equivalent to v0.1.11.noko1 but restores support for Java 8.

Fixed

  • [JRuby] Java 8 support is restored, fixing a regression present in v1.14.0..v1.14.4 and v1.15.0..v1.15.1. [#2887]

1.15.1 / 2023-05-19

Dependencies

Fixed

  • [CRuby] The libxml2 update fixes an encoding regression when push-parsing UTF-8 sequences. [#2882, upstream issue and commit]

1.15.0 / 2023-05-15

Notes

Ability to opt into system malloc and free

Since 2009, Nokogiri has configured libxml2 to use ruby_xmalloc et al for memory management. This has provided benefits for memory management, but comes with a performance penalty.

Users can now opt into using system malloc for libxml2 memory management by setting an environment variable:

# "default" here means "libxml2's default" which is system malloc
NOKOGIRI_LIBXML_MEMORY_MANAGEMENT=default

Benchmarks show that this setting will significantly improve performance, but be aware that the tradeoff may involve poorer memory management including bloated heap sizes and/or OOM conditions.

You can read more about this in the decision record at adr/2023-04-libxml-memory-management.md.

Dependencies

Added

  • Encoding objects may now be passed to serialization methods like #to_xml, #to_html, #serialize, and #write_to to specify the output encoding. Previously only encoding names (strings) were accepted. [#2774, #2798] (@ellaklara)
  • [CRuby] Users may opt into using system malloc for libxml2 memory management. For more detail, see note above or adr/2023-04-libxml-memory-management.md.

Changed

  • [CRuby] Schema.from_document now makes a defensive copy of the document if it has blank text nodes with Ruby objects instantiated for them. This prevents unsafe behavior in libxml2 from causing a segfault. There is a small performance cost, but we think this has the virtue of being "what the user meant" since modifying the original is surprising behavior for most users. Previously this was addressed in v1.10.9 by raising an exception.

Fixed

  • [CRuby] XSLT.transform now makes a defensive copy of the document if it has blank text nodes with Ruby objects instantiated for them and the template uses xsl:strip-spaces. This prevents unsafe behavior in libxslt from causing a segfault. There is a small performance cost, but we think this has the virtue of being "what the user meant" since modifying the original is surprising behavior for most users. Previously this would allow unsafe memory access and potentially segfault. [#2800]

Improved

  • Nokogiri::XML::Node::SaveOptions#inspect now shows the names of the options set in the bitmask, similar to ParseOptions. [#2767]
  • #inspect and pretty-printing are improved for AttributeDecl, ElementContent, ElementDecl, and EntityDecl.
  • [CRuby] The C extension now uses Ruby's TypedData API for managing all the libxml2 structs. Write barriers may improve GC performance in some extreme cases. [#2808] (@etiennebarrie and @byroot)
  • [CRuby] ObjectSpace.memsize_of reports a pretty good guess of memory usage when called on Nokogiri::XML::Document objects. [#2807] (@etiennebarrie and @byroot)
  • [CRuby] Users installing the "ruby" platform gem and compiling libxml2 and libxslt from source will now be using a modern config.guess and config.sub that supports new architectures like loongarch64. [#2831] (@zhangwenlong8911)
  • [CRuby] HTML5 parser:
  • [JRuby] Node#first_element_child now returns nil if there are only non-element children. Previously a null pointer exception was raised. [#2808, #2844]
  • Documentation for Nokogiri::XSLT now has usage examples including custom function handlers.

Deprecated

  • Passing a Nokogiri::XML::Node as the first parameter to CDATA.new is deprecated and will generate a warning. This parameter should be a kind of Nokogiri::XML::Document. This will become an error in a future version of Nokogiri.
  • Passing a Nokogiri::XML::Node as the first parameter to Schema.from_document is deprecated and will generate a warning. This parameter should be a kind of Nokogiri::XML::Document. This will become an error in a future version of Nokogiri.
  • Passing a Nokogiri::XML::Node as the second parameter to Text.new is deprecated and will generate a warning. This parameter should be a kind of Nokogiri::XML::Document. This will become an error in a future version of Nokogiri.
  • [CRuby] Calling a custom XPath function without the nokogiri namespace is deprecated and will generate a warning. Support for non-namespaced functions will be removed in a future version of Nokogiri. (Note that JRuby has never supported non-namespaced custom XPath functions.)

Thank you!

The following people and organizations were kind enough to sponsor @flavorjones or the Nokogiri project during the development of v1.15.0:

  • Götz Görisch (@GoetzGoerisch)
  • Airbnb (@airbnb)
  • Kyohei Nanba (@kyo-nanba)
  • Maxime Gauthier (@biximilien)
  • @renuo
  • @dbootyfvrt
  • YOSHIDA Katsuhiko (@kyoshidajp)
  • Homebrew (@Homebrew)
  • Hiroshi SHIBATA (@hsbt)
  • PuLLi (@the-pulli)
  • SiteLog GmbH (@sitelog-gmbh)
  • @zzak
  • Evil Martians (@evilmartians)
  • Ajaya Agrawalla (@ajaya)
  • Modern Treasury (@Modern-Treasury)
  • Danilo Lessa Bernardineli (@danlessa)

We'd also like to thank @github who donate a ton of compute time for our CI pipelines!

1.14.5 / 2023-05-24

Note

To ensure that JRuby users on Java 8 can apply the security changes from v1.14.4, we're cutting this release on the v1.14.x branch. We don't expect to make any more v1.14.x releases.

(The changes in this release are incorporated into the v1.15.x release branch at v1.15.2.)

Dependencies

  • [JRuby] Vendored org.nokogiri:nekodtd is updated to v0.1.11.noko2. This is functionally equivalent to v0.1.11.noko1 but restores support for Java 8.

Fixed

  • [JRuby] Java 8 support is restored, fixing a regression introduced in v1.14.0. [#2887]

1.14.4 / 2023-05-11

Dependencies

  • [JRuby] Vendored Xalan-J is updated to v2.7.3. This is the first Xalan release in nine years, and it was done to address CVE-2022-34169.

    The Nokogiri maintainers wish to stress that Nokogiri users were not vulnerable to this CVE, as we explained in GHSA-qwq9-89rg-ww72, and so upgrading is really at the discretion of users.

    This release was cut primarily so that JRuby users of v1.14.x can avoid vulnerability scanner alerts on earlier versions of Xalan-J.

1.14.3 / 2023-04-11

Security

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.10.4 from v2.10.3.

1.14.2 / 2023-02-13

Fixed

  • Calling NodeSet#to_html on an empty node set no longer raises an encoding-related exception. This bug was introduced in v1.14.0 while fixing #2649. [#2784]

1.14.1 / 2023-01-30

Fixed

  • Serializing documents now works again with pseudo-IO objects that don't support IO's encoding API (like rubyzip's Zip::OutputStream). This was a regression in v1.14.0 due to the fix for #752 in #2434, and was not completely fixed by #2753. [#2773]
  • [CRuby] Address compiler warnings about void* casting and old-style C function definitions.

1.14.0 / 2023-01-12

Notable Changes

Ruby

This release introduces native gem support for Ruby 3.2. (Also see "Technical note" under "Changed" below.)

This release ends support for:

Faster, more reliable installation: Native Gem for aarch64-linux (aka linux/arm64/v8)

This version of Nokogiri ships official native gem support for the aarch64-linux platform, which should support AWS Graviton and other ARM64 Linux platforms. Please note that glibc >= 2.29 is required for aarch64-linux systems, see Supported Platforms for more information.

Faster, more reliable installation: Native Gem for arm-linux (aka linux/arm/v7)

This version of Nokogiri ships experimental native gem support for the arm-linux platform. Please note that glibc >= 2.29 is required for arm-linux systems, see Supported Platforms for more information.

Pattern matching

This version introduces an experimental pattern matching API for XML::Attr, XML::Document, XML::DocumentFragment, XML::Namespace, XML::Node, and XML::NodeSet (and their subclasses).

Some documentation on what can be matched:

We welcome feedback on this API at #2360.

Dependencies

CRuby

  • Vendored libiconv is updated to v1.17

JRuby

  • This version of Nokogiri uses jar-dependencies to manage most of the vendored Java dependencies. nokogiri -v now outputs maven metadata for all Java dependencies, and Nokogiri::VERSION_INFO also contains this metadata. [#2432]
  • HTML parsing is now provided by net.sourceforge.htmlunit:neko-htmlunit:2.61.0 (previously Nokogiri used a fork of org.cyberneko.html:nekohtml)
  • Vendored Jing is updated from com.thaiopensource:jing:20091111 to nu.validator:jing:20200702VNU.
  • New dependency on net.sf.saxon:Saxon-HE:9.6.0-4 (via nu.validator:jing:20200702VNU).

Added

  • Node#wrap and NodeSet#wrap now also accept a Node type argument, which will be duped for each wrapper. For cases where many nodes are being wrapped, creating a Node once using Document#create_element and passing that Node multiple times is significantly faster than re-parsing markup on each call. [#2657]
  • [CRuby] Invocation of custom XPath or CSS handler functions may now use the nokogiri namespace prefix. Historically, the JRuby implementation required this namespace but the CRuby implementation did not support it. It's recommended that all XPath and CSS queries use the nokogiri namespace going forward. Invocation without the namespace is planned for deprecation in v1.15.0 and removal in a future release. [#2147]
  • HTML5::Document#quirks_mode and HTML5::DocumentFragment#quirks_mode expose the quirks mode used by the parser.

Improved

Functional

Performance

  • Serialization of HTML5 documents and fragments has been re-implemented and is ~10x faster than previous versions. [#2596, #2569]
  • Parsing of HTML5 documents is ~90% faster thanks to additional compiler optimizations being applied. [#2639]
  • Compare Encoding objects rather than compare their names. This is a slight performance improvement and is future-proof. [#2454] (@casperisfine)

Error handling

  • Document#canonicalize now raises an exception if inclusive_namespaces is non-nil and the mode is inclusive, i.e. XML_C14N_1_0 or XML_C14N_1_1. inclusive_namespaces can only be passed with exclusive modes, and previously this silently failed.
  • Empty CSS selectors now raise a clearer Nokogiri::CSS::SyntaxError message, "empty CSS selector". Previously the exception raised from the bowels of racc was "unexpected '$' after ''". [#2700]
  • [CRuby] XML::Reader parsing errors encountered during Reader#attribute_hash and Reader#namespaces now raise an XML::SyntaxError. Previously these methods would return nil and users would generally experience NoMethodErrors from elsewhere in the code.
  • Prefer ruby_xmalloc to malloc within the C extension. [#2480] (@Garfield96)

Installation

  • Avoid compile-time conflict with system-installed gumbo.h on OpenBSD. [#2464]
  • Remove calls to vasprintf in favor of platform-independent rb_vsprintf
  • Installation from source on systems missing libiconv will once again generate a helpful error message (broken since v1.11.0). [#2505]
  • [CRuby+OSX] Compiling from source on MacOS will use the clang option -Wno-unknown-warning-option to avoid errors when Ruby injects options that clang doesn't know about. [#2689]

Fixed

  • SAX::Parser's encoding attribute will not be clobbered when an alternative encoding is passed into SAX::Parser#parse_io. [#1942] (@kp666)
  • Serialized HTML4::DocumentFragment will now be properly encoded. Previously this empty string was encoded as US-ASCII. [#2649]
  • Node#wrap now uses the parent as the context node for parsing wrapper markup, falling back to the document for unparented nodes. Previously the document was always used.
  • [CRuby] UTF-16-encoded documents longer than ~4000 code points now serialize properly. Previously the serialized document was corrupted when it exceeded the length of libxml2's internal string buffer. [#752]
  • [CRuby] The HTML5 parser now correctly handles text at the end of form elements.
  • [CRuby] HTML5::Document#fragment now always uses body as the parsing context. Previously, fragments were parsed in the context of the associated document's root node, which allowed for inconsistent parsing. [#2553]
  • [CRuby] Nokogiri::HTML5::Document#url now correctly returns the URL passed to the constructor method. Previously it always returned nil. [#2583]
  • [CRuby] HTML5 encoding detection is now case-insensitive with respect to meta tag charset declaration. [#2693]
  • [CRuby] HTML5 fragment parsing in context of an annotation-xml node now works. Previously this rarely-used path invoked rb_funcall with incorrect parameters, resulting in an exception, a fatal error, or potentially a segfault. [#2692]
  • [CRuby] HTML5 quirks mode during fragment parsing more closely matches document parsing. [#2646]
  • [JRuby] Fixed a bug with adding the same namespace to multiple nodes via #add_namespace_definition. [#1247]
  • [JRuby] NodeSet#[] now raises a TypeError if passed an invalid parameter type. [#2211]

Deprecated

  • Nokogiri.install_default_aliases is deprecated in favor of Nokogiri::EncodingHandler.install_default_aliases. This is part of a private API and is probably not called by anybody, but we'll go through a deprecation cycle before removal anyway. [#2643, #2446]

Changed

  • [CRuby+OSX] Technical note: On MacOS Ruby 3.2, the symbols from libxml2 and libxslt are no longer exported. Ruby 3.2 adopted new features from the Darwin toolchain that make it challenging to continue to support this rarely-used binary API. A future minor release of Nokogiri may remove these symbols (and others) entirely. Feedback from downstream gem maintainers is welcome at #2746, where you'll also be able to read deeper context on this decision.

Thank you!

The following people and organizations were kind enough to sponsor @flavorjones or the Nokogiri project during the development of v1.14.0:

  • Götz Görisch (@GoetzGoerisch)
  • Airbnb (@airbnb)
  • Kyohei Nanba (@kyo-nanba)
  • Maxime Gauthier (@biximilien)
  • @renuo
  • @dbootyfvrt
  • YOSHIDA Katsuhiko (@kyoshidajp)
  • Homebrew (@Homebrew)
  • David Vrensk (@dvrensk)
  • Alex Daragiu (@daragiu)
  • Github (@github)
  • Julian Joseph (@Julian88Tex)
  • Charles Simon-Meunier (@csimonmeunier)
  • Ben Slaughter (@benSlaughter)
  • Garen Torikian (@gjtorikian)
  • Frank Groeneveld (@frenkel)
  • Hiroshi SHIBATA (@hsbt)

1.13.10 / 2022-12-07

Security

  • [CRuby] Address CVE-2022-23476, unchecked return value from xmlTextReaderExpand. See GHSA-qv4q-mr5r-qprj for more information.

Improvements

  • [CRuby] XML::Reader#attribute_hash now returns nil on parse errors. This restores the behavior of #attributes from v1.13.7 and earlier. [#2715]

1.13.9 / 2022-10-18

Security

Dependencies

  • [CRuby] Vendored libxml2 is updated to v2.10.3 from v2.9.14.
  • [CRuby] Vendored libxslt is updated to v1.1.37 from v1.1.35.
  • [CRuby] Vendored zlib is updated from 1.2.12 to 1.2.13. (See LICENSE-DEPENDENCIES.md for details on which packages redistribute this library.)

Fixed

  • [CRuby] Nokogiri::XML::Namespace objects, when compacted, update their internal struct's reference to the Ruby object wrapper. Previously, with GC compaction enabled, a segmentation fault was possible after compaction was triggered. [#2658] (@eightbitraptor and @peterzhu2118)
  • [CRuby] Document#remove_namespaces! now defers freeing the underlying xmlNs struct until the Document is GCed. Previously, maintaining a reference to a Namespace object that was removed in this way could lead to a segfault. [#2658]

1.13.8 / 2022-07-23

Deprecated

  • XML::Reader#attribute_nodes is deprecated due to incompatibility between libxml2's xmlReader memory semantics and Ruby's garbage collector. Although this method continues to exist for backwards compatibility, it is unsafe to call and may segfault. This method will be removed in a future version of Nokogiri, and callers should use #attribute_hash instead. [#2598]

Improvements

  • XML::Reader#attribute_hash is a new method to safely retrieve the attributes of a node from XML::Reader. [#2598, #2599]

Fixed

  • [CRuby] Calling XML::Reader#attributes is now safe to call. In Nokogiri <= 1.13.7 this method may segfault. [#2598, #2599]

1.13.7 / 2022-07-12

Fixed

XML::Node objects, when compacted, update their internal struct's reference to the Ruby object wrapper. Previously, with GC compaction enabled, a segmentation fault was possible after compaction was triggered. [#2578] (@eightbitraptor)

1.13.6 / 2022-05-08

Security

  • [CRuby] Address CVE-2022-29181, improper handling of unexpected data types, related to untrusted inputs to the SAX parsers. See GHSA-xh29-r2w5-wx8m for more information.

Improvements

  • {HTML4,XML}::SAX::{Parser,ParserContext} constructor methods now raise TypeError instead of segfaulting when an incorrect type is passed.

1.13.5 / 2022-05-04

Security

Dependencies

  • [CRuby] Vendored libxml2 is updated from v2.9.13 to v2.9.14.

Improvements

  • [CRuby] The libxml2 HTML parser no longer exhibits quadratic behavior when recovering some broken markup related to start-of-tag and bare < characters.

Changed

  • [CRuby] The libxml2 HTML parser in v2.9.14 recovers from some broken markup differently. Notably, the XML CDATA escape sequence <![CDATA[ and incorrectly-opened comments will result in HTML text nodes starting with &lt;! instead of skipping the invalid tag. This behavior is a direct result of the quadratic-behavior fix noted above. The behavior of downstream sanitizers relying on this behavior will also change. Some tests describing the changed behavior are in test/html4/test_comments.rb.

1.13.4 / 2022-04-11

Security

Dependencies

  • [CRuby] Vendored zlib is updated from 1.2.11 to 1.2.12. (See LICENSE-DEPENDENCIES.md for details on which packages redistribute this library.)
  • [JRuby] Vendored Xerces-J (xerces:xercesImpl) is updated from 2.12.0 to 2.12.2.
  • [JRuby] Vendored nekohtml (org.cyberneko.html) is updated from a fork of 1.9.21 to 1.9.22.noko2. This fork is now publicly developed at https://github.com/sparklemotion/nekohtml

1.13.3 / 2022-02-21

Fixed

  • [CRuby] Revert a HTML4 parser bug in libxml 2.9.13 (introduced in Nokogiri v1.13.2). The bug causes libxml2's HTML4 parser to fail to recover when encountering a bare < character in some contexts. This version of Nokogiri restores the earlier behavior, which is to recover from the parse error and treat the < as normal character data (which will be serialized as &lt; in a text node). The bug (and the fix) is only relevant when the RECOVER parse option is set, as it is by default. [#2461]

1.13.2 / 2022-02-21

Security

  • [CRuby] Vendored libxml2 is updated from 2.9.12 to 2.9.13. This update addresses CVE-2022-23308.
  • [CRuby] Vendored libxslt is updated from 1.1.34 to 1.1.35. This update addresses CVE-2021-30560.

Please see GHSA-fq42-c5rg-92c2 for more information about these CVEs.

Dependencies

1.13.1 / 2022-01-13

Fixed

  • Fix Nokogiri::XSLT.quote_params regression in v1.13.0 that raised an exception when non-string stylesheet parameters were passed. Non-string parameters (e.g., integers and symbols) are now explicitly supported and both keys and values will be stringified with #to_s. [#2418]
  • Fix CSS selector query regression in v1.13.0 that raised an Nokogiri::XML::XPath::SyntaxError when parsing XPath attributes mixed into the CSS query. Although this mash-up of XPath and CSS syntax previously worked unintentionally, it is now an officially supported feature and is documented as such. [#2419]

1.13.0 / 2022-01-06

Notes

Ruby

This release introduces native gem support for Ruby 3.1. Please note that Windows users should use the x64-mingw-ucrt platform gem for Ruby 3.1, and x64-mingw32 for Ruby 2.6–3.0 (see RubyInstaller 3.1.0 release notes).

This release ends support for:

Faster, more reliable installation: Native Gem for ARM64 Linux

This version of Nokogiri ships experimental native gem support for the aarch64-linux platform, which should support AWS Graviton and other ARM Linux platforms. We don't yet have CI running for this platform, and so we're interested in hearing back from y'all whether this is working, and what problems you're seeing. Please send us feedback here: Feedback: Have you used the aarch64-linux native gem?

Publishing

This version of Nokogiri opts-in to the "MFA required to publish" setting on Rubygems.org. This and all future Nokogiri gem files must be published to Rubygems by an account with multi-factor authentication enabled. This should provide some additional protection against supply-chain attacks.

A related discussion about Trust exists at #2357 in which I invite you to participate if you have feelings or opinions on this topic.

Dependencies

  • [CRuby] Vendored libiconv is updated from 1.15 to 1.16. (Note that libiconv is only redistributed in the native windows and native darwin gems, see LICENSE-DEPENDENCIES.md for more information.) [#2206]
  • [CRuby] Upgrade mini_portile2 dependency from ~> 2.6.1 to ~> 2.7.0. ("ruby" platform gem only.)

Improved

  • {XML,HTML4}::DocumentFragment constructors all now take an optional parse options parameter or block (similar to Document constructors). [#1692] (@JackMc)
  • Nokogiri::CSS.xpath_for allows an XPathVisitor to be injected, for finer-grained control over how CSS queries are translated into XPath.
  • [CRuby] XML::Reader#encoding will return the encoding detected by the parser when it's not passed to the constructor. [#980]
  • [CRuby] Handle abruptly-closed HTML comments as recommended by WHATWG. (Thanks to tehryanx for reporting)
  • [CRuby] Node#line is no longer capped at 65535. libxml v2.9.0 and later support a new parse option, exposed as Nokogiri::XML::ParseOptions::PARSE_BIG_LINES, which is turned on by default in ParseOptions::DEFAULT_{XML,XSLT,HTML,SCHEMA} (Note that JRuby already supported large line numbers.) [#1764, #1493, #1617, #1505, #1003, #533]
  • [CRuby] If a cycle is introduced when reparenting a node (i.e., the node becomes its own ancestor), a RuntimeError is raised. libxml2 does no checking for this, which means cycles would otherwise result in infinite loops on subsequent operations. (Note that JRuby already did this.) [#1912]
  • [CRuby] Source builds will download zlib and libiconv via HTTPS. ("ruby" platform gem only.) [#2391] (@jmartin-r7)
  • [JRuby] Node#line behavior has been modified to return the line number of the node in the final DOM structure. This behavior is different from CRuby, which returns the node's position in the input string. Ideally the two implementations would be the same, but at least is now officially documented and tested. The real-world impact of this change is that the value returned in JRuby is greater by 1 to account for the XML prolog in the output. [#2380] (@dabdine)

Fixed

  • CSS queries on HTML5 documents now correctly match foreign elements (SVG, MathML) when namespaces are not specified in the query. [#2376]
  • XML::Builder blocks restore context properly when exceptions are raised. [#2372] (@ric2b and @rinthedev)
  • The Nokogiri::CSS::Parser cache now uses the XPathVisitor configuration as part of the cache key, preventing incorrect cache results from being returned when multiple XPathVisitor options are being used.
  • Error recovery from in-context parsing (e.g., Node#parse) now always uses the correct DocumentFragment class. Previously Nokogiri::HTML4::DocumentFragment was always used, even for XML documents. [#1158]
  • DocumentFragment#> now works properly, matching a CSS selector against only the fragment roots. [#1857]
  • XML::DocumentFragment#errors now correctly contains any parsing errors encountered. Previously this was always empty. (Note that HTML::DocumentFragment#errors already did this.)
  • [CRuby] Fix memory leak in Document#canonicalize when inclusive namespaces are passed in. [#2345]
  • [CRuby] Fix memory leak in Document#canonicalize when an argument type error is raised. [#2345]
  • [CRuby] Fix memory leak in EncodingHandler where iconv handlers were not being cleaned up. [#2345]
  • [CRuby] Fix memory leak in XPath custom handlers where string arguments were not being cleaned up. [#2345]
  • [CRuby] Fix memory leak in Reader#base_uri where the string returned by libxml2 was not freed. [#2347]
  • [JRuby] Deleting a Namespace from a NodeSet no longer modifies the href to be the default namespace URL.
  • [JRuby] Fix XHTML formatting of closing tags for non-container elements. [#2355]

Deprecated

  • Passing a Nokogiri::XML::Node as the second parameter to Node.new is deprecated and will generate a warning. This parameter should be a kind of Nokogiri::XML::Document. This will become an error in a future version of Nokogiri. [#975]
  • Nokogiri::CSS::Parser, Nokogiri::CSS::Tokenizer, and Nokogiri::CSS::Node are now internal-only APIs that are no longer documented, and should not be considered stable. With the introduction of XPathVisitor injection into Nokogiri::CSS.xpath_for there should be no reason to rely on these internal APIs.
  • CSS-to-XPath utility classes Nokogiri::CSS::XPathVisitorAlwaysUseBuiltins and XPathVisitorOptimallyUseBuiltins are deprecated. Prefer Nokogiri::CSS::XPathVisitor with appropriate constructor arguments. These classes will be removed in a future version of Nokogiri.

1.12.5 / 2021-09-27

Security

[JRuby] Address CVE-2021-41098 (GHSA-2rr5-8q37-2w7h).

In Nokogiri v1.12.4 and earlier, on JRuby only, the SAX parsers resolve external entities (XXE) by default. This fix turns off entity-resolution-by-default in the JRuby SAX parsers to match the CRuby SAX parsers' behavior.

CRuby users are not affected by this CVE.

Fixed

  • [CRuby] Document#to_xhtml properly serializes self-closing tags in libxml > 2.9.10. A behavior change introduced in libxml 2.9.11 resulted in emitting start and and tags (e.g., <br></br>) instead of a self-closing tag (e.g., <br/>) in previous Nokogiri versions. [#2324]

1.12.4 / 2021-08-29

Notable fix: Namespace inheritance

Namespace behavior when reparenting nodes has historically been poorly specified and the behavior diverged between CRuby and JRuby. As a result, making this behavior consistent in v1.12.0 introduced a breaking change.

This patch release reverts the Builder behavior present in v1.12.0..v1.12.3 but keeps the Document behavior. This release also introduces a Document attribute to allow affected users to easily change this behavior for their legacy code without invasive changes.

Compensating Feature in XML::Document

This release of Nokogiri introduces a new Document boolean attribute, namespace_inheritance, which controls whether children should inherit a namespace when they are reparented. Nokogiri::XML:Document defaults this attribute to false meaning "do not inherit," thereby making explicit the behavior change introduced in v1.12.0.

CRuby users who desire the pre-v1.12.0 behavior may set document.namespace_inheritance = true before reparenting nodes.

See https://nokogiri.org/rdoc/Nokogiri/XML/Document.html#namespace_inheritance-instance_method for example usage.

Fix for XML::Builder

However, recognizing that we want Builder-created children to inherit namespaces, Builder now will set namespace_inheritance=true on the underlying document for both JRuby and CRuby. This means that, on CRuby, the pre-v1.12.0 behavior is restored.

Users who want to turn this behavior off may pass a keyword argument to the Builder constructor like so:

Nokogiri::XML::Builder.new(namespace_inheritance: false)

See https://nokogiri.org/rdoc/Nokogiri/XML/Builder.html#label-Namespace+inheritance for example usage.

Downstream gem maintainers

Note that any downstream gems may want to specifically omit Nokogiri v1.12.0--v1.12.3 from their dependency specification if they rely on child namespace inheritance:

Gem::Specification.new do |gem|
  # ...
  gem.add_runtime_dependency 'nokogiri', '!=1.12.3', '!=1.12.2', '!=1.12.1', '!=1.12.0'
  # ...
end

Fixed

  • [JRuby] Fix NPE in Schema parsing when an imported resource doesn't have a systemId. [#2296] (@pepijnve)

1.12.3 / 2021-08-10

Fixed

  • [CRuby] Fix compilation of libgumbo on older systems with versions of GCC that give errors on C99-isms. Affected systems include RHEL6, RHEL7, and SLES12. [#2302]

1.12.2 / 2021-08-04

Fixed

  • [CRuby] Ensure that C extension files in non-native gem installations are loaded using require and rely on $LOAD_PATH instead of using require_relative. This issue only exists when deleting shared libraries that exist outside the extensions directory, something users occasionally do to conserve disk space. [#2300]

1.12.1 / 2021-08-03

Fixed

  • [CRuby] Fix compilation of libgumbo on BSD systems by avoiding GNU-isms. [#2298]

1.12.0 / 2021-08-02

Notable Addition: HTML5 Support (CRuby only)

HTML5 support has been added (to CRuby only) by merging Nokogumbo into Nokogiri. The Nokogumbo public API has been preserved, so this functionality is available under the Nokogiri::HTML5 namespace. [#2204]

Please note that HTML5 support is not available for JRuby in this version. However, we feel it is important to think about JRuby and we hope to work on this in the future. If you're interested in helping with HTML5 support on JRuby, please reach out to the maintainers by commenting on issue #2227.

Many thanks to Sam Ruby, Steve Checkoway, and Craig Barnes for creating and maintaining Nokogumbo and supporting the Gumbo HTML5 parser. They're now Nokogiri core contributors with all the powers and privileges pertaining thereto. 🙌

Notable Change: Nokogiri::HTML4 module and namespace

Nokogiri::HTML has been renamed to Nokogiri::HTML4, and Nokogiri::HTML is aliased to preserve backwards-compatibility. Nokogiri::HTML and Nokogiri::HTML4 parse methods still use libxml2's (or NekoHTML's) HTML4 parser in the v1.12 release series.

Take special note that if you rely on the class name of an object in your code, objects will now report a class of Nokogiri::HTML4::Foo where they previously reported Nokogiri::HTML::Foo. Instead of relying on the string returned by Object#class, prefer Class#=== or Object#is_a? or Object#instance_of?.

Future releases of Nokogiri may deprecate HTML methods or otherwise change this behavior, so please start using HTML4 in place of HTML.

Added

  • [CRuby] Nokogiri::VERSION_INFO["libxslt"]["datetime_enabled"] is a new boolean value which describes whether libxslt (or, more properly, libexslt) has compiled-in datetime support. This generally going to be true, but some distros ship without this support (e.g., some mingw UCRT-based packages, see msys2/MINGW-packages#8957). See #2272 for more details.

Changed

  • Introduce a new constant, Nokogiri::XML::ParseOptions::DEFAULT_XSLT, which adds the libxslt-preferred options of NOENT | DTDLOAD | DTDATTR | NOCDATA to ParseOptions::DEFAULT_XML.
  • Nokogiri.XSLT parses stylesheets using ParseOptions::DEFAULT_XSLT, which should make some edge-case XSL transformations match libxslt's default behavior. [#1940]

Fixed

  • [CRuby] Namespaced attributes are handled properly when their parent node is reparented into another document. Previously, the namespace may have gotten dropped. [#2228]
  • [CRuby] Reparented nodes no longer inherit their parent's namespace. Previously, a node without a namespace was forced to adopt its parent's namespace. [#1712, #425]

Improved

  • [CRuby] Speed up (slightly) the compile time of packaged libraries libiconv, libxml2, and libxslt by using autoconf's --disable-dependency-tracking option. ("ruby" platform gem only.)

Deprecated

  • Deprecating Nokogumbo's Nokogiri::HTML5.get. This method will be removed in a future version of Nokogiri.

Dependencies

  • [CRuby] Upgrade mini_portile2 dependency from ~> 2.5.0 to ~> 2.6.1. ("ruby" platform gem only.)

1.11.7 / 2021-06-02

Fixed

  • [CRuby] Backporting an upstream fix to XPath recursion depth limits which impacted some users of complex XPath queries. This issue is present in libxml 2.9.11 and 2.9.12. [#2257]

1.11.6 / 2021-05-26

Fixed

  • [CRuby] DocumentFragment#path now does proper error-checking to handle behavior introduced in libxml > 2.9.10. In v1.11.4 and v1.11.5, calling DocumentFragment#path could result in a segfault.

1.11.5 / 2021-05-19

Fixed

[Windows CRuby] Work around segfault at process exit on Windows when using libxml2 system DLLs.

libxml 2.9.12 introduced new behavior to avoid memory leaks when unloading libxml2 shared libraries (see libxml/!66). Early testing caught this segfault on non-Windows platforms (see #2059 and libxml@956534e) but it was incompletely fixed and is still an issue on Windows platforms that are using system DLLs.

We work around this by configuring libxml2 in this situation to use its default memory management functions. Note that if Nokogiri is not on Windows, or is not using shared system libraries, it will will continue to configure libxml2 to use Ruby's memory management functions. Nokogiri::VERSION_INFO["libxml"]["memory_management"] will allow you to verify when the default memory management functions are being used. [#2241]

Added

Nokogiri::VERSION_INFO["libxml"] now contains the key "memory_management" to declare whether libxml2 is using its default memory management functions, or whether it uses the memory management functions from ruby. See above for more details.

1.11.4 / 2021-05-14

Security

[CRuby] Vendored libxml2 upgraded to v2.9.12 which addresses:

Note that two additional CVEs were addressed upstream but are not relevant to this release. CVE-2021-3516 via xmllint is not present in Nokogiri, and CVE-2020-7595 has been patched in Nokogiri since v1.10.8 (see #1992).

Please see nokogiri/GHSA-7rrm-v45f-jp64 or #2233 for a more complete analysis of these CVEs and patches.

Dependencies

  • [CRuby] vendored libxml2 is updated from 2.9.10 to 2.9.12. (Note that 2.9.11 was skipped because it was superseded by 2.9.12 a few hours after its release.)

1.11.3 / 2021-04-07

Fixed

  • [CRuby] Passing non-Node objects to Document#root= now raises an ArgumentError exception. Previously this likely segfaulted. [#1900]
  • [JRuby] Passing non-Node objects to Document#root= now raises an ArgumentError exception. Previously this raised a TypeError exception.
  • [CRuby] arm64/aarch64 systems (like Apple's M1) can now compile libxml2 and libxslt from source (though we continue to strongly advise users to install the native gems for the best possible experience)

1.11.2 / 2021-03-11

Fixed

  • [CRuby] NodeSet may now safely contain Node objects from multiple documents. Previously the GC lifecycle of the parent Document objects could lead to nodes being GCed while still in scope. [#1952]
  • [CRuby] Patch libxml2 to avoid "huge input lookup" errors on large CDATA elements. (See upstream GNOME/libxml2#200 and GNOME/libxml2!100.) [#2132].
  • [CRuby+Windows] Enable Nokogumbo (and other downstream gems) to compile and link against nokogiri.so by including LDFLAGS in Nokogiri::VERSION_INFO. [#2167]
  • [CRuby] {XML,HTML}::Document.parse now invokes #initialize exactly once. Previously #initialize was invoked twice on each object.
  • [JRuby] {XML,HTML}::Document.parse now invokes #initialize exactly once. Previously #initialize was not called, which was a problem for subclassing such as done by Loofah.

Improved

  • Reduce the number of object allocations needed when parsing an HTML::DocumentFragment. [#2087] (@ashmaroli)
  • [JRuby] Update the algorithm used to calculate Node#line to be wrong less-often. The underlying parser, Xerces, does not track line numbers, and so we've always used a hacky solution for this method. [#1223, #2177]
  • Introduce --enable-system-libraries and --disable-system-libraries flags to extconf.rb. These flags provide the same functionality as --use-system-libraries and the NOKOGIRI_USE_SYSTEM_LIBRARIES environment variable, but are more idiomatic. [#2193] (@eregon)
  • [TruffleRuby] --disable-static is now the default on TruffleRuby when the packaged libraries are used. This is more flexible and compiles faster. (Note, though, that the default on TR is still to use system libraries.) [#2191, #2193] (@eregon)

Changed

  • Nokogiri::XML::Path is now a Module (previously it has been a Class). It has been acting solely as a Module since v1.0.0. See 8461c74.

1.11.1 / 2021-01-06

Fixed

  • [CRuby] If libxml-ruby is loaded before nokogiri, the SAX and Push parsers no longer call libxml-ruby's handlers. Instead, they defensively override the libxml2 global handler before parsing. [#2168]

1.11.0 / 2021-01-03

Notes

Faster, more reliable installation: Native Gems for Linux and OSX/Darwin

"Native gems" contain pre-compiled libraries for a specific machine architecture. On supported platforms, this removes the need for compiling the C extension and the packaged libraries. This results in much faster installation and more reliable installation, which as you probably know are the biggest headaches for Nokogiri users.

We've been shipping native Windows gems since 2009, but starting in v1.11.0 we are also shipping native gems for these platforms:

  • Linux: x86-linux and x86_64-linux -- including musl platforms like alpine
  • OSX/Darwin: x86_64-darwin and arm64-darwin

We'd appreciate your thoughts and feedback on this work at #2075.

Dependencies

Ruby

This release introduces support for Ruby 2.7 and 3.0 in the precompiled native gems.

This release ends support for:

Gems

  • Explicitly add racc as a runtime dependency. [#1988] (@voxik)
  • [MRI] Upgrade mini_portile2 dependency from ~> 2.4.0 to ~> 2.5.0 [#2005] (@alejandroperea)

Security

See note below about CVE-2020-26247 in the "Changed" subsection entitled "XML::Schema parsing treats input as untrusted by default".

Added

  • Add Node methods for manipulating "keyword attributes" (for example, class and rel): #kwattr_values, #kwattr_add, #kwattr_append, and #kwattr_remove. [#2000]
  • Add support for CSS queries a:has(> b), a:has(~ b), and a:has(+ b). [#688] (@jonathanhefner)
  • Add Node#value? to better match expected semantics of a Hash-like object. [#1838, #1840] (@MatzFan)
  • [CRuby] Add Nokogiri::XML::Node#line= for use by downstream libs like nokogumbo. [#1918] (@stevecheckoway)
  • nokogiri.gemspec is back after a 10-year hiatus. We still prefer you use the official releases, but main is pretty stable these days, and YOLO.

Performance

  • [CRuby] The CSS ~= operator and class selector . are about 2x faster. [#2137, #2135]
  • [CRuby] Patch libxml2 to call strlen from xmlStrlen rather than the naive implementation, because strlen is generally optimized for the architecture. [#2144] (@ilyazub)
  • Improve performance of some namespace operations. [#1916] (@ashmaroli)
  • Remove unnecessary array allocations from Node serialization methods [#1911] (@ashmaroli)
  • Avoid creation of unnecessary zero-length String objects. [#1970] (@ashmaroli)
  • Always compile libxml2 and libxslt with '-O2' [#2022, #2100] (@ilyazub)
  • [JRuby] Lots of code cleanup and performance improvements. [#1934] (@kares)
  • [CRuby] RelaxNG.from_document no longer leaks memory. [#2114]

Improved

  • [CRuby] Handle incorrectly-closed HTML comments as WHATWG recommends for browsers. [#2058] (Thanks to HackerOne user mayflower for reporting this)
  • {HTML,XML}::Document#parse now accept Pathname objects. Previously this worked only if the referenced file was less than 4096 bytes long; longer files resulted in undefined behavior because the read method would be repeatedly invoked. [#1821, #2110] (@doriantaylor and @phokz)
  • [CRuby] Nokogumbo builds faster because it can now use header files provided by Nokogiri. [#1788] (@stevecheckoway)
  • Add frozen_string_literal: true magic comment to all lib files. [#1745] (@oniofchaos)
  • [JRuby] Clean up deprecated calls into JRuby. [#2027] (@headius)

Fixed

  • HTML Parsing in "strict" mode (i.e., the RECOVER parse option not set) now correctly raises a XML::SyntaxError exception. Previously the value of the RECOVER bit was being ignored by CRuby and was misinterpreted by JRuby. [#2130]
  • The CSS ~= operator now correctly handles non-space whitespace in the class attribute. commit e45dedd
  • The switch to turn off the CSS-to-XPath cache is now thread-local, rather than being shared mutable state. [#1935]
  • The Node methods add_previous_sibling, previous=, before, add_next_sibling, next=, after, replace, and swap now correctly use their parent as the context node for parsing markup. These methods now also raise a RuntimeError if they are called on a node with no parent. [nokogumbo#160]
  • [JRuby] XML::Schema XSD validation errors are captured in XML::Schema#errors. These errors were previously ignored.
  • [JRuby] Standardize reading from IO like objects, including StringIO. [#1888, #1897]
  • [JRuby] Fix how custom XPath function namespaces are inferred to be less naive. [#1890, #2148]
  • [JRuby] Clarify exception message when custom XPath functions can't be resolved.
  • [JRuby] Comparison of Node to Document with Node#<=> now matches CRuby/libxml2 behavior.
  • [CRuby] Syntax errors are now correctly captured in Document#errors for short HTML documents. Previously the SAX parser used for encoding detection was clobbering libxml2's global error handler.
  • [CRuby] Fixed installation on AIX with respect to vasprintf. [#1908]
  • [CRuby] On some platforms, avoid symbol name collision with glibc's canonicalize. [#2105]
  • [Windows Visual C++] Fixed compiler warnings and errors. [#2061, #2068]
  • [CRuby] Fixed Nokogumbo integration which broke in the v1.11.0 release candidates. [#1788] (@stevecheckoway)
  • [JRuby] Fixed document encoding regression in v1.11.0 release candidates. [#2080, #2083] (@thbar)

Removed

  • The internal method Nokogiri::CSS::Parser.cache_on= has been removed. Use .set_cache if you need to muck with the cache internals.
  • The class method Nokogiri::CSS::Parser.parse has been removed. This was originally deprecated in 2009 in 13db61b. Use Nokogiri::CSS.parse instead.

Changed

XML::Schema input is now "untrusted" by default

Address CVE-2020-26247.

In Nokogiri versions <= 1.11.0.rc3, XML Schemas parsed by Nokogiri::XML::Schema were trusted by default, allowing external resources to be accessed over the network, potentially enabling XXE or SSRF attacks.

This behavior is counter to the security policy intended by Nokogiri maintainers, which is to treat all input as untrusted by default whenever possible.

Please note that this security fix was pushed into a new minor version, 1.11.x, rather than a patch release to the 1.10.x branch, because it is a breaking change for some schemas and the risk was assessed to be "Low Severity".

More information and instructions for enabling "trusted input" behavior in v1.11.0.rc4 and later is available at the public advisory.

HTML parser now obeys the strict or norecover parsing option

(Also noted above in the "Fixed" section) HTML Parsing in "strict" mode (i.e., the RECOVER parse option not set) now correctly raises a XML::SyntaxError exception. Previously the value of the RECOVER bit was being ignored by CRuby and was misinterpreted by JRuby.

If you're using the default parser options, you will be unaffected by this fix. If you're passing strict or norecover to your HTML parser call, you may be surprised to see that the parser now fails to recover and raises a XML::SyntaxError exception. Given the number of HTML documents on the internet that libxml2 would consider to be ill-formed, this is probably not what you want, and you can omit setting that parse option to restore the behavior that you have been relying upon.

Apologies to anyone inconvenienced by this breaking bugfix being present in a minor release, but I felt it was appropriate to introduce this fix because it's straightforward to fix any code that has been relying on this buggy behavior.

VersionInfo, the output of nokogiri -v, and related constants

This release changes the metadata provided in Nokogiri::VersionInfo which also affects the output of nokogiri -v. Some related constants have also been changed. If you're using VersionInfo programmatically, or relying on constants related to underlying library versions, please read the detailed changes for Nokogiri::VersionInfo at #2139 and accept our apologies for the inconvenience.

1.10.10 / 2020-07-06

Features

  • [MRI] Cross-built Windows gems now support Ruby 2.7 [#2029]. Note that prior to this release, the v1.11.x prereleases provided this support.

1.10.9 / 2020-03-01

Fixed

  • [MRI] Raise an exception when Nokogiri detects a specific libxml2 edge case involving blank Schema nodes wrapped by Ruby objects that would cause a segfault. Currently no fix is available upstream, so we're preventing a dangerous operation and informing users to code around it if possible. [#1985, #2001]
  • [JRuby] Change NodeSet#to_a to return a RubyArray instead of Object, for compilation under JRuby 9.2.9 and later. [#1968, #1969] (@headius)

1.10.8 / 2020-02-10

Security

[MRI] Pulled in upstream patch from libxml that addresses CVE-2020-7595. Full details are available in #1992. Note that this patch is not yet (as of 2020-02-10) in an upstream release of libxml.

1.10.7 / 2019-12-03

Fixed

  • [MRI] Ensure the patch applied in v1.10.6 works with GNU patch. [#1954]

1.10.6 / 2019-12-03

Fixed

  • [MRI] Fix FreeBSD installation of vendored libxml2. [#1941, #1953] (@nurse)

1.10.5 / 2019-10-31

Security

[MRI] Vendored libxslt upgraded to v1.1.34 which addresses three CVEs for libxslt:

  • CVE-2019-13117
  • CVE-2019-13118
  • CVE-2019-18197
  • CVE-2019-19956

More details are available at #1943.

Dependencies

  • [MRI] vendored libxml2 is updated from 2.9.9 to 2.9.10
  • [MRI] vendored libxslt is updated from 1.1.33 to 1.1.34

1.10.4 / 2019-08-11

Security

Address CVE-2019-5477 [#1915].

A command injection vulnerability in Nokogiri v1.10.3 and earlier allows commands to be executed in a subprocess by Ruby's Kernel.open method. Processes are vulnerable only if the undocumented method Nokogiri::CSS::Tokenizer#load_file is being passed untrusted user input.

This vulnerability appears in code generated by the Rexical gem versions v1.0.6 and earlier. Rexical is used by Nokogiri to generate lexical scanner code for parsing CSS queries. The underlying vulnerability was addressed in Rexical v1.0.7 and Nokogiri upgraded to this version of Rexical in Nokogiri v1.10.4.

This CVE's public notice is #1915

1.10.3 / 2019-04-22

Security

[MRI] Pulled in upstream patch from libxslt that addresses CVE-2019-11068. Full details are available in #1892. Note that this patch is not yet (as of 2019-04-22) in an upstream release of libxslt.

1.10.2 / 2019-03-24

Security

  • [MRI] Remove support from vendored libxml2 for future script macros. [#1871]
  • [MRI] Remove support from vendored libxml2 for server-side includes within attributes. [#1877]

Fixed

  • [JRuby] Fix node ownership in duplicated documents. [#1060]
  • [JRuby] Rethrow exceptions caught by Java SAX handler. [#1847, #1872] (@adjam)

1.10.1 / 2019-01-13

Added

  • [MRI] During installation, handle Xcode 10's new library path. [#1801, #1851] (@mlj and @deepj)
  • Avoid unnecessary creation of Procs in many methods. [#1776] (@chopraanmol1)

Fixed

  • CSS selector :has() now correctly matches against any descendant. Previously this selector matched against only direct children). [#350] (@Phrogz)
  • NodeSet#attr now returns nil if it's empty. Previously this raised a NoMethodError.
  • [MRI] XPath errors are no longer suppressed during XSLT::Stylesheet#transform. Previously these errors were suppressed which led to silent failures and a subsequent segfault. [#1802]

1.10.0 / 2019-01-04

Added

  • [MRI] Cross-built Windows gems now support Ruby 2.6 [#1842, #1850]

Dependencies

1.9.1 / 2018-12-17

Fixed

  • Fix a bug introduced in v1.9.0 where XML::DocumentFragment#dup no longer returned an instance of the callee's class, instead always returning an XML::DocumentFragment. This notably broke any subclass of XML::DocumentFragment including HTML::DocumentFragment as well as the Loofah gem's Loofah::HTML::DocumentFragment. [#1846]

1.9.0 / 2018-12-17

Security

  • [JRuby] Upgrade Xerces dependency from 2.11.0 to 2.12.0 to address upstream vulnerability CVE-2012-0881 [#1831] (Thanks @grajagandev for reporting.)

Improved

  • Decrease installation size by removing many unneeded files (e.g., /test) from the packaged gems. [#1719] (@stevecrozz)

Added

  • XML::Attr#value= allows HTML node attribute values to be set to either a blank string or an empty boolean attribute. [#1800]
  • Introduce XML::Node#wrap which does what XML::NodeSet#wrap has always done, but for a single node. [#1531] (@ethirajsrinivasan)
  • [MRI] Improve installation experience on macOS High Sierra (Darwin). [#1812, #1813] (@gpakosz and @nurse)
  • [MRI] Node#dup supports copying a node directly to a new document. See the method documentation for details.
  • [MRI] DocumentFragment#dup is now more memory-efficient, avoiding making unnecessary copies. [#1063]
  • [JRuby] NodeSet has been rewritten to improve performance! [#1795]

Fixed

  • NodeSet#each now returns self instead of zero. [#1822] (@olehif)
  • [MRI] Address a memory leak when using XML::Builder to create nodes with namespaces. [#1810]
  • [MRI] Address a memory leak when unparenting a DTD. [#1784] (@stevecheckoway)
  • [MRI] Use RbConfig::CONFIG instead of ::MAKEFILE_CONFIG to fix installations that use Makefile macros. [#1820] (@nobu)
  • [JRuby] Decrease large memory usage when making nested XPath queries. [#1749]
  • [JRuby] Fix failing tests on JRuby 9.2.x
  • [JRuby] Fix default namespaces in nodes reparented into a different document [#1774]
  • [JRuby] Fix support for Java 9. [#1759] (@Taywee)

Dependencies

  • [MRI] Upgrade mini_portile2 dependency from ~> 2.3.0 to ~> 2.4.0

1.8.5 / 2018-10-04

Security

[MRI] Pulled in upstream patches from libxml2 that address CVE-2018-14404 and CVE-2018-14567. Full details are available in #1785. Note that these patches are not yet (as of 2018-10-04) in an upstream release of libxml2.

Fixed

  • [MRI] Fix regression in installation when building against system libraries, where some systems would not be able to find libxml2 or libxslt when present. (Regression introduced in v1.8.3.) [#1722]
  • [JRuby] Fix node reparenting when the destination doc is empty. [#1773]

1.8.4 / 2018-07-03

Fixed

  • [MRI] Fix memory leak when creating nodes with namespaces. (Introduced in v1.5.7) [#1771]

1.8.3 / 2018-06-16

Security

[MRI] Behavior in libxml2 has been reverted which caused CVE-2018-8048 (loofah gem), CVE-2018-3740 (sanitize gem), and CVE-2018-3741 (rails-html-sanitizer gem). The commit in question is here:

https://github.com/GNOME/libxml2/commit/960f0e2

and more information is available about this commit and its impact here:

flavorjones/loofah#144

This release simply reverts the libxml2 commit in question to protect users of Nokogiri's vendored libraries from similar vulnerabilities.

If you're offended by what happened here, I'd kindly ask that you comment on the upstream bug report here:

https://bugzilla.gnome.org/show_bug.cgi?id=769760

More Security

[MRI] Vendored libxml2 upgraded to v2.9.8 which addresses CVE-2016-9318 [#1582].

Dependencies

  • [MRI] libxml2 is updated from 2.9.7 to 2.9.8

Added

  • Node#classes, #add_class, #append_class, and #remove_class are added.
  • NodeSet#append_class is added.
  • NodeSet#remove_attribute is a new alias for NodeSet#remove_attr.
  • NodeSet#each now returns an Enumerator when no block is passed (@park53kr)
  • [JRuby] General improvements in JRuby implementation (@kares)

Fixed

  • CSS attribute selectors now gracefully handle queries using integers. [#711]
  • Handle ASCII-8BIT encoding on fragment input [#553]
  • Handle non-string return values within Reader [#898]
  • [JRuby] Allow Node#replace to insert Comment and CDATA nodes. [#1666]
  • [JRuby] Stability and speed improvements to Node, Sax::PushParser, and the JRuby implementation [#1708, #1710, #1501]

1.8.2 / 2018-01-29

Security

[MRI] The update of vendored libxml2 from 2.9.5 to 2.9.7 addresses at least one published vulnerability, CVE-2017-15412. [#1714 has complete details]

Dependencies

  • [MRI] libxml2 is updated from 2.9.5 to 2.9.7
  • [MRI] libxslt is updated from 1.1.30 to 1.1.32

Added

  • [MRI] OpenBSD installation should be a bit easier now. [#1685] (@jeremyevans)
  • [MRI] Cross-built Windows gems now support Ruby 2.5

Fixed

  • Node#serialize once again returns UTF-8-encoded strings. [#1659]
  • [JRuby] made SAX parsing of characters consistent with C implementation [#1676] (@andrew-aladev)
  • [MRI] Predefined entities, when inspected, no longer cause a segfault. [#1238]

1.8.1 / 2017-09-19

Dependencies

  • [MRI] libxml2 is updated from 2.9.4 to 2.9.5.
  • [MRI] libxslt is updated from 1.1.29 to 1.1.30.
  • [MRI] optional dependency on the pkg-config gem has had its constraint loosened to ~> 1.1 (from ~> 1.1.7). [#1660]
  • [MRI] Upgrade mini_portile2 dependency from ~> 2.2.0 to ~> 2.3.0, which will validate checksums on the vendored libxml2 and libxslt tarballs before using them.

Fixed

  • NodeSet#first with an integer argument longer than the length of the NodeSet now correctly clamps the length of the returned NodeSet to the original length. [#1650] (@Derenge)
  • [MRI] Ensure CData.new raises TypeError if the content argument is not implicitly convertible into a string. [#1669]

1.8.0 / 2017-06-04

Dependencies

This release ends support for Ruby 2.1 on Windows in the x86-mingw32 and x64-mingw32 platform gems (containing pre-compiled DLLs). Official support ended for Ruby 2.1 on 2017-04-01.

Please note that this deprecation note only applies to the precompiled Windows gems. Ruby 2.1 continues to be supported (for now) in the default gem when compiled on installation.

Dependencies

  • [Windows] Upgrade iconv from 1.14 to 1.15 (unless --use-system-libraries)
  • [Windows] Upgrade zlib from 1.2.8 to 1.2.11 (unless --use-system-libraries)
  • [MRI] Upgrade rake-compiler dependency from 0.9.2 to 1.0.3
  • [MRI] Upgrade mini-portile2 dependency from ~> 2.1.0 to ~> 2.2.0
  • [JRuby] Removed support for jruby --1.8 code paths. [#1607] (@kares)
  • [MRI Windows] Retrieve zlib source from http://zlib.net/fossils to avoid deprecation issues going forward. See #1632 for details around this problem.

Added

  • NodeSet#clone is now an alias for NodeSet#dup [#1503] (@stephankaag)
  • Allow Processing Instructions and Comments as children of a document root. [#1033] (@windwiny)
  • [MRI] PushParser#replace_entities and #replace_entities= will control whether entities are replaced or not. [#1017] (@spraints)
  • [MRI] SyntaxError#to_s now includes line number, column number, and log level if made available by the parser. [#1304, #1637] (@spk and @ccarruitero)
  • [MRI] Cross-built Windows gems now support Ruby 2.4
  • [MRI] Support for frozen string literals. [#1413]
  • [MRI] Support for installing Nokogiri on a machine in FIPS-enabled mode [#1544]
  • [MRI] Vendored libraries are verified with SHA-256 hashes (formerly some MD5 hashes were used) [#1544]
  • [JRuby] (performance) remove unnecessary synchronization of class-cache [#1563] (@kares)
  • [JRuby] (performance) remove unnecessary cloning of objects in XPath searches [#1563] (@kares)
  • [JRuby] (performance) more performance improvements, particularly in XPath, Reader, XmlNode, and XmlNodeSet [#1597] (@kares)

Fixed

  • HTML::SAX::Parser#parse_io now correctly parses HTML and not XML [#1577] (Thanks for the test case, @gregors)
  • Support installation on systems with a lib64 site config. [#1562]
  • [MRI] on OpenBSD, do not require gcc if using system libraries [#1515] (@jeremyevans)
  • [MRI] XML::Attr.new checks type of Document arg to prevent segfaults. [#1477]
  • [MRI] Prefer xmlCharStrdup (and friends) to strdup (and friends), which can cause problems on some platforms. [#1517] (@jeremy)
  • [JRuby] correctly append a text node before another text node [#1318] (@jkraemer)
  • [JRuby] custom xpath functions returning an integer now work correctly [#1595] (@kares)
  • [JRuby] serializing (#to_html, #to_s, et al) a document with explicit encoding now works correctly. [#1281, #1440] (@kares)
  • [JRuby] XML::Reader now returns parse errors [#1586] (@kares)
  • [JRuby] Empty NodeSets are now decorated properly. [#1319] (@kares)
  • [JRuby] Merged nodes no longer results in Java exceptions during XPath queries. [#1320] (@kares)

1.7.2 / 2017-05-09

Security

[MRI] Upstream libxslt patches are applied to the vendored libxslt 1.1.29 which address CVE-2017-5029 and CVE-2016-4738.

For more information:

1.7.1 / 2017-03-19

Security

[MRI] Upstream libxml2 patches are applied to the vendored libxml 2.9.4 which address CVE-2016-4658 and CVE-2016-5131.

For more information:

1.7.0.1 / 2017-01-04

Fixed

  • Fix OpenBSD support. [#1569] (related to #1543)

1.7.0 / 2016-12-26

Added

  • Remove deprecation warnings in Ruby 2.4.0 [#1545] (@matthewd)
  • Support egcc compiler on OpenBSD [#1543] (@frenkel and @knu)

Dependencies

This release ends support for:

1.6.8.1 / 2016-10-03

Dependencies

Removes required dependency on the pkg-config gem. This dependency was introduced in v1.6.8 and, because it's distributed under LGPL, was objectionable to many Nokogiri users (#1488, #1496).

This version makes pkg-config an optional dependency. If it's installed, it's used; but otherwise Nokogiri will attempt to work around its absence.

1.6.8 / 2016-06-06

Security

[MRI] Bundled libxml2 is upgraded to 2.9.4, which fixes many security issues. Many of these had previously been patched in the vendored libxml 2.9.2 in the 1.6.7.x branch, but some are newer.

See these libxml2 email posts for more:

For a more detailed analysis, you may care to read Canonical's take on these security issues:

[MRI] Bundled libxslt is upgraded to 1.1.29, which fixes a security issue as well as many long-known outstanding bugs, some features, some portability improvements, and general cleanup.

See this libxslt email post for more:

Added

Several changes were made to improve performance:

  • [MRI] Simplify NodeSet#to_a with a minor speed-up. [#1397]
  • XML::Node#ancestors optimization. [#1297] (Bruno Sutic)
  • Use Symbol#to_proc where we weren't previously. [#1296] (Bruno Sutic)
  • XML::DTD#each uses implicit block calls. (@glaucocustodio)
  • Fall back to the pkg-config gem if we're having trouble finding the system libxml2. This should help many FreeBSD users. [#1417]
  • Set document encoding appropriately even on blank document. [#1043] (@batter)

Fixed

  • [JRuby] fix slow add_child [#692]
  • [JRuby] fix load errors when deploying to JRuby/Torquebox [#1114] (@atambo and @jvshahid)
  • [JRuby] fix NPE when inspecting nodes returned by NodeSet#drop [#1042] (@mkristian)
  • [JRuby] fix nil attribute node's namespace in reader [#1327] (@codekitchen)
  • [JRuby] fix Nokogiri munging unicode characters that require more than 2 bytes [#1113] (@mkristian)
  • [JRuby] allow unlinking an unparented node (#1112, #1152) (@esse)
  • [JRuby] allow Fragment parsing on a frozen string (#444, #1077)
  • [JRuby] HTML style tags are no longer encoded [#1316] (@tbeauvais)
  • [MRI] fix assertion failure while accessing attribute node's namespace in reader [#843] (@2potatocakes)
  • [MRI] fix issue with GCing namespace nodes returned in an xpath query. [#1155]
  • [MRI] Ensure C strings are null-terminated. [#1381]
  • [MRI] Ensure Rubygems is loaded before using mini_portile2 at installation. (#1393, #1411) (@JonRowe)
  • [MRI] Handling another edge case where the libxml-ruby gem's global callbacks were smashing the heap. [#1426]. (Thanks to @bbergstrom for providing an isolated test case)
  • [MRI] Ensure encodings are passed to Sax::Parser xmldecl callback. [#844]
  • [MRI] Ensure default ns prefix is applied correctly when reparenting nodes to another document. [#391] (@ylecuyer)
  • [MRI] Ensure Reader handles non-existent attributes as expected. [#1254] (@ccutrer)
  • [MRI] Cleanup around namespace handling when reparenting nodes. (#1332, #1333, #1444) (@cuttrer and @bradleybeddoes)
  • unescape special characters in CSS queries [#1303] (@twalpole)
  • consistently handle empty documents [#1349]
  • Update to mini_portile2 2.1.0 to address whitespace-handling during patching. [#1402]
  • Fix encoding of xml node namespaces.
  • Work around issue installing Nokogiri on overlayfs (commonly used in Docker containers). (#1370, #1405)

Notes

  • Removed legacy code remaining from Ruby 1.8.x support.
  • Removed legacy code remaining from REE support.
  • Removing hacky workarounds for bugs in some older versions of libxml2.
  • Handling C strings in a forward-compatible manner, see https://github.com/ruby/ruby/blob/v2_2_0/NEWS#L319

1.6.7.2 / 2016-01-20

This version pulls in several upstream patches to the vendored libxml2 and libxslt to address:

  • CVE-2015-7499

Ubuntu classifies this as "Priority: Low", RedHat classifies this as "Impact: Moderate", and NIST classifies this as "Severity: 5.0 (MEDIUM)".

MITRE record is https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2015-7499

1.6.7.1 / 2015-12-16

This version pulls in several upstream patches to the vendored libxml2 and libxslt to address:

  • CVE-2015-5312
  • CVE-2015-7497
  • CVE-2015-7498
  • CVE-2015-7499
  • CVE-2015-7500
  • CVE-2015-8241
  • CVE-2015-8242
  • CVE-2015-8317

See also http://www.ubuntu.com/usn/usn-2834-1/

1.6.7 / 2015-11-29

Added

This version supports native builds on Windows using the RubyInstaller DevKit. It also supports Ruby 2.2.x on Windows, as well as making several other improvements to the installation process on various platforms.

Security

This version also includes the security patches already applied in v1.6.6.3 and v1.6.6.4 to the vendored libxml2 and libxslt source. See #1374 and #1376 for details.

Added

  • Cross-built gems now have a proper ruby version requirement. [#1266]
  • Ruby 2.2.x is supported on Windows.
  • Native build is supported on Windows.
  • [MRI] libxml2 and libxslt config.guess files brought up to date. [#1326] (@hernan-erasmo)
  • [JRuby] fix error in validating files with jruby (#1355, #1361) (@twalpole)
  • [MRI, OSX] Patch to handle nonstandard location of iconv.h. (#1206, #1210, #1218, #1345) (@neonichu)

Fixed

  • [JRuby] reset the namespace cache when replacing the document's innerHtml [#1265] (@mkristian)
  • [JRuby] Document#parse should support IO objects that respond to #read. [#1124] (Jake Byman)
  • [MRI] Duplicate-id errors when setting the id attribute on HTML documents are now silenced. [#1262]
  • [JRuby] SAX parser cuts texts in pieces when square brackets exist. [#1261]
  • [JRuby] Namespaced attributes aren't removed by remove_attribute. [#1299]

1.6.6.4 / 2015-11-19

This version pulls in an upstream patch to the vendored libxml2 to address:

  • unclosed comment uninitialized access issue [#1376]

This issue was assigned CVE-2015-8710 after the fact. See http://seclists.org/oss-sec/2015/q4/616 for details.

1.6.6.3 / 2015-11-16

This version pulls in several upstream patches to the vendored libxml2 and libxslt to address:

  • CVE-2015-1819
  • CVE-2015-7941_1
  • CVE-2015-7941_2
  • CVE-2015-7942
  • CVE-2015-7942-2
  • CVE-2015-8035
  • CVE-2015-7995

See #1374 for details.

1.6.6.2 / 2015-01-23

Fixed

  • Fixed installation issue affecting compiler arguments. [#1230]

1.6.6.1 / 2015-01-22

Note that 1.6.6.0 was not released.

Added

  • 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.

Fixed

  • 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] (Piotr Szmielew)

1.6.5 / 2014-11-26

Added

  • Implement Slop#respond_to_missing?. [#1176]
  • Optimized the XPath query generated by an an+b CSS query.

Fixed

  • 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

Fixed

  • (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

Added

  • (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.

Fixed

  • (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

Fixed

  • Addressing an Apple Macintosh installation problem for GCC users. [#1130] (@zenspider)

1.6.3 / 2014-07-20

Added

  • Added Node#document? and Node#processing_instruction?

Fixed

  • [JRuby] Fix Ruby memory exhaustion vulnerability. [#1087] (@ocher)
  • [MRI] Fix segfault during GC when using libxml-ruby and nokogiri together in multi-threaded environment. [#895] (@ender672)
  • Building on OSX 10.9 stock ruby 2.0.0 now works. [#1101] (@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

Fixed

  • 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

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:

It is recommended that you upgrade from 1.6.x to this version as soon as possible.

Dependencies

Now requires libxml >= 2.6.21 (was previously >= 2.6.17).

Added

  • 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] (Magnus Bergmark)
  • Add support for bare and multiple :not() functions in selectors. [#887] (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] (Hoylen Sue)
  • Allow Nokogiri::XML::Node#parse from unparented non-element nodes. [#407]

Fixed

  • Ensure :only-child pseudo class works within :not pseudo class. [#858] (Yamagishi Kazutoshi)
  • Don't call pkg_config when using bundled libraries in extconf.rb [#931] (Shota Fukumori)
  • Nokogiri.parse() does not mistake a non-HTML document like a RSS document as HTML document. [#932] (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]

1.6.1 / 2013-12-14

Fixed

  • (JRuby) Fix out of memory bug when certain invalid documents are parsed.
  • (JRuby) Fix regression of billion-laughs vulnerability. [#586]

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

Added

  • (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

Dependencies

  • Support for Ruby 1.8.7 and prior has been dropped

1.5.11 / 2013-12-14

Fixed

  • (JRuby) Fix out of memory bug when certain invalid documents are parsed.
  • (JRuby) Fix regression of billion-laughs vulnerability. [#586]

1.5.10 / 2013-06-07

Fixed

  • (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. [#912] (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] (Dirkjan Bussink)
  • Fix TypeError when running tests. [#900] (Cédric Boutillier)

1.5.9 / 2013-03-21

Fixed

  • 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

Fixed

  • (JRuby) Fix EmptyStackException thrown by elements with xlink:href attributes and no base_uri. [#534, #805] (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

Added

  • Windows support for Ruby 2.0.

Fixed

  • 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] (Michael Klein)
  • (MRI) SAX parser handles empty processing instructions. [#845]

1.5.6 / 2012-12-19

Added

  • Improved performance of XML::Document#collect_namespaces. [#761] (Juergen Mangler)
  • New callback SAX::Document#processing_instruction (Kitaiti Makoto)
  • Node#native_content= allows setting unescaped node content. [#768]
  • XPath lookup with namespaces supports symbol keys. [#729] (Ben Langfeld.)
  • XML::Node#[]= stringifies values. [#729] (Ben Langfeld.)
  • bin/nokogiri will process a document from $stdin
  • bin/nokogiri -e will execute a program from the command line
  • (JRuby) bin/nokogiri --version will print the Xerces and NekoHTML versions.

Fixed

  • Nokogiri now detects XSLT transform errors. [#731] (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] (Bradley Schaefer)
  • Prefixed element inconsistency between CRuby and JRuby. [#712]
  • (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] (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]

1.5.5 / 2012-06-24

Added

  • Much-improved support for JRuby in 1.9 mode! Yay!

Fixed

  • Regression in JRuby Nokogiri add_previous_sibling (1.5.0 -> 1.5.1) [#691] (John Shahid)
  • JRuby unable to create HTML doc if URL arg provided [#674] (John Shahid)
  • JRuby raises NullPointerException when given HTML document is nil or empty string. [#699]
  • JRuby 1.9 error, uncaught throw 'encoding_found', has been fixed. [#673]
  • Invalid encoding returned in JRuby with US-ASCII. [#583]
  • XmlSaxPushParser raises IndexOutOfBoundsException when over 512 characters are given. [#567, #615]
  • When xpath evaluation returns empty NodeSet, decorating NodeSet's base document raises exception. [#514]
  • JRuby raises exception when xpath with namespace is specified. pull request [#681] (Piotr Szmielew)
  • JRuby renders nodes without their namespace when subclassing Node. [#695]
  • JRuby raises NAMESPACE_ERR (org.w3c.dom.DOMException) while instantiating RDF::RDFXML::Writer. [#683]
  • JRuby is not able to use namespaces in xpath. [#493]
  • JRuby's Entity resolving should be consistent with C-Nokogiri [#704, #647, #703]

1.5.4 / 2012-06-12

Added

  • The "nokogiri" script now has more verbose output when passed the --rng option. [#675] (Dan Radez)
  • Build support on hardened Debian systems that use -Werror=format-security. [#680].
  • Better build support for systems with pkg-config. [#584]
  • Better build support for systems with multiple iconv installations.

Fixed

  • Segmentation fault when creating a comment node for a DocumentFragment. [#677, #678]
  • Treat '.' as xpath in at() and search(). [#690]

Security

(MRI) Default parse options for XML documents were changed to not make network connections during document parsing, to avoid XXE vulnerability. [#693]

To re-enable this behavior, the configuration method nononet may be called, like this:

Nokogiri::XML::Document.parse(xml) { |config| config.nononet }

Insert your own joke about double-negatives here.

1.5.3 / 2012-06-01

Added

  • Support for "prefixless" CSS selectors ~, > and + like jQuery supports. [#621, #623] (David Lee)
  • Attempting to improve installation on homebrew 0.9 (with regards to iconv). Isn't package management convenient?

Fixed

  • 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

Added

  • 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] (Dan Radez)

Fixed

  • Fix many memory leaks and segfault opportunities. 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] (Juan Wajnerman)
  • Node#to_xml does not override :save_with if it is provided. [#505]
  • Node#set is a private method (JRuby). [#564] (Nick Sieger)
  • C14n cleanup and Node#canonicalize (Ivan Pirlik) [#563]

1.5.0 / 2011-07-01

Notes

  • See changelog from 1.4.7

Added

  • extracted sets of Node::SaveOptions into Node::SaveOptions::DEFAULT_{X,H,XH}TML (refactor)

Fixed

  • 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

Fixed

  • 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.

Dependencies

  • 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.

Removed

  • FFI support is removed.

1.4.7 / 2011-07-01

Fixed

  • Fix a bug in advanced encoding detection that leads to partially duplicated document when parsing an HTML file with unknown encoding. 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 / 2011-05-19

Added

  • 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]

Fixed

  • 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 / 2010-11-15

Added

  • 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

Fixed

  • 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]

1.4.3 / 2010-07-28

Added

  • XML::Reader#empty_element? returns true for empty elements. [#262]
  • Node#remove_namespaces! now removes namespace declarations as well. [#294]
  • NodeSet#at_xpath, NodeSet#at_css and NodeSet#> do what the corresponding methods of Node do.

Fixed

  • XML::NodeSet#{include?,delete,push} accept an XML::Namespace
  • XML::Document#parse added for parsing in the context of a document
  • XML::DocumentFragment#inner_html= works with contextual parsing! [#298, #281]
  • lib/nokogiri/css/parser.y Combined CSS functions + pseudo selectors fixed
  • Reparenting text nodes is safe, even when the operation frees adjacent merged nodes. [#283]
  • Fixed libxml2 versionitis issue with xmlFirstElementChild et al. [#303]
  • XML::Attr#add_namespace now works as expected. [#252]
  • HTML::DocumentFragment uses the string's encoding. [#305]
  • Fix the CSS3 selector translation rule for the general sibling combinator (a.k.a. preceding selector) that incorrectly converted "E ~ F G" to "//F//G[preceding-sibling::E]".

1.4.2 / 2010-05-22

Added

  • XML::Node#parse will parse XML or HTML fragments with respect to the context node.
  • XML::Node#namespaces returns all namespaces defined in the node and all ancestor nodes (previously did not return ancestors' namespace definitions).
  • Added Enumerable to XML::Node
  • Nokogiri::XML::Schema#validate now uses xmlSchemaValidateFile if a filename is passed, which is faster and more memory-efficient. [#219]
  • XML::Document#create_entity will create new EntityDecl objects. [#174]
  • JRuby FFI implementation no longer uses ObjectSpace._id2ref, instead using Charles Nutter's rocking Weakling gem.
  • Nokogiri::XML::Node#first_element_child fetch the first child node that is an ELEMENT node.
  • Nokogiri::XML::Node#last_element_child fetch the last child node that is an ELEMENT node.
  • Nokogiri::XML::Node#elements fetch all children nodes that are ELEMENT nodes.
  • Nokogiri::XML::Node#add_child, #add_previous_sibling, #before, #add_next_sibling, #after, #inner_html, #swap and #replace all now accept a Node, DocumentFragment, NodeSet, or a string containing markup.
  • Node#fragment? indicates whether a node is a DocumentFragment.

Fixed

  • XML::NodeSet is now always decorated (if the document has decorators). [#198]
  • XML::NodeSet#slice gracefully handles offset+length larger than the set length. [#200]
  • XML::Node#content= safely unlinks previous content. [#203]
  • XML::Node#namespace= takes nil as a parameter
  • XML::Node#xpath returns things other than NodeSet objects. [#208]
  • XSLT::StyleSheet#transform accepts hashes for parameters. [#223]
  • Pseudo selectors inside not() work. [#205]
  • XML::Builder doesn't break when nodes are unlinked. [#228] (vihai)
  • Encoding can be forced on the SAX parser. [#204] (Eugene Pimenov)
  • XML::DocumentFragment uses XML::Node#parse to determine children.
  • Fixed a memory leak in xml reader. [#244] (sdor)
  • Node#replace returns the new child node as claimed in the RDoc. Previously returned +self+.

Notes

  • The Windows gems now bundle DLLs for libxml 2.7.6 and libxslt 1.1.26. Prior to this release, libxml 2.7.3 and libxslt 1.1.24 were bundled.

1.4.1 / 2009-12-10

Added

  • Added Nokogiri::LIBXML_ICONV_ENABLED
  • Alias Node#[] to Node#attr
  • XML::Node#next_element added
  • XML::Node#> added for searching a nodes immediate children
  • XML::NodeSet#reverse added
  • Added fragment support to Node#add_child, Node#add_next_sibling, Node#add_previous_sibling, and Node#replace.
  • XML::Node#previous_element implemented
  • Rubinius support
  • The CSS selector engine now supports :has()
  • XML::NodeSet#filter() was added
  • XML::Node.next= and .previous= are aliases for add_next_sibling and add_previous_sibling. [#183]

Fixed

  • XML fragments with namespaces do not raise an exception (regression in 1.4.0)
  • Node#matches? works in nodes contained by a DocumentFragment. [#158]
  • Document should not define add_namespace() method. [#169]
  • XPath queries returning namespace declarations do not segfault.
  • Node#replace works with nodes from different documents. [#162]
  • Adding XML::Document#collect_namespaces
  • Fixed bugs in the SOAP4R adapter
  • Fixed bug in XML::Node#next_element for certain edge cases
  • Fixed load path issue with JRuby under Windows. [#160].
  • XSLT#apply_to will honor the "output method". (richardlehane)
  • Fragments containing leading text nodes with newlines now parse properly. [#178].

1.4.0 / 2009-10-30

Added

  • Node#at_xpath returns the first element of the NodeSet matching the XPath expression.
  • Node#at_css returns the first element of the NodeSet matching the CSS selector.
  • NodeSet#| for unions [#119] (Serabe)
  • NodeSet#inspect makes prettier output
  • Node#inspect implemented for more rubyish document inspecting
  • Added XML::DTD#external_id
  • Added XML::DTD#system_id
  • Added XML::ElementContent for DTD Element content validity
  • Better namespace declaration support in Nokogiri::XML::Builder
  • Added XML::Node#external_subset
  • Added XML::Node#create_external_subset
  • Added XML::Node#create_internal_subset
  • XML Builder can append raw strings (#141, patch from dudleyf)
  • XML::SAX::ParserContext added
  • XML::Document#remove_namespaces! for the namespace-impaired

Fixed

  • returns nil when HTML documents do not declare a meta encoding tag. [#115]
  • Uses RbConfig::CONFIG['host_os'] to adjust ENV['PATH'] [#113]
  • NodeSet#search is more efficient [#119] (Serabe)
  • NodeSet#xpath handles custom xpath functions
  • Fixing a SEGV when XML::Reader gets attributes for current node
  • Node#inner_html takes the same arguments as Node#to_html [#117]
  • DocumentFragment#css delegates to it's child nodes [#123]
  • NodeSet#[] works with slices larger than NodeSet#length [#131]
  • Reparented nodes maintain their namespace. [#134]
  • Fixed SEGV when adding an XML::Document to NodeSet
  • XML::SyntaxError can be duplicated. [#148]

Removed

  • Hpricot compatibility layer removed

1.3.3 / 2009-07-26

Added

  • NodeSet#children returns all children of all nodes

Fixed

  • Override libxml-ruby's global error handler
  • ParseOption#strict fixed
  • Fixed a segfault when sending an empty string to Node#inner_html= [#88]
  • String encoding is now set to UTF-8 in Ruby 1.9
  • Fixed a segfault when moving root nodes between documents. [#91]
  • Fixed an O(n) penalty on node creation. [#101]
  • Allowing XML documents to be output as HTML documents

Deprecations

  • Hpricot compatibility layer will be removed in 1.4.0

1.3.2 / 2009-06-22

Added

  • Nokogiri::XML::DTD#validate will validate your document

Fixed

  • Nokogiri::XML::NodeSet#search will search top level nodes. [#73]
  • Removed namespace related methods from Nokogiri::XML::Document
  • Fixed a segfault when a namespace was added twice
  • Made nokogiri work with Snow Leopard [#79]
  • Mailing list has moved to: http://groups.google.com/group/nokogiri-talk
  • HTML fragments now correctly handle comments and CDATA blocks. [#78]
  • Nokogiri::XML::Document#clone is now an alias of dup

Deprecations

  • Nokogiri::XML::SAX::Document#start_element_ns is deprecated, please switch to Nokogiri::XML::SAX::Document#start_element_namespace
  • Nokogiri::XML::SAX::Document#end_element_ns is deprecated, please switch to Nokogiri::XML::SAX::Document#end_element_namespace

1.3.1 / 2009-06-07

Fixed

  • extconf.rb checks for optional RelaxNG and Schema functions
  • Namespace nodes are added to the Document node cache

1.3.0 / 2009-05-30

Added

  • Builder changes scope based on block arity
  • Builder supports methods ending in underscore similar to tagz
  • Nokogiri::XML::Node#<=> compares nodes based on Document position
  • Nokogiri::XML::Node#matches? returns true if Node can be found with given selector.
  • Nokogiri::XML::Node#ancestors now returns an Nokogiri::XML::NodeSet
  • Nokogiri::XML::Node#ancestors will match parents against optional selector
  • Nokogiri::HTML::Document#meta_encoding for getting the meta encoding
  • Nokogiri::HTML::Document#meta_encoding= for setting the meta encoding
  • Nokogiri::XML::Document#encoding= to set the document encoding
  • Nokogiri::XML::Schema for validating documents against XSD schema
  • Nokogiri::XML::RelaxNG for validating documents against RelaxNG schema
  • Nokogiri::HTML::ElementDescription for fetching HTML element descriptions
  • Nokogiri::XML::Node#description to fetch the node description
  • Nokogiri::XML::Node#accept implements Visitor pattern
  • bin/nokogiri for easily examining documents (Yutaka HARA)
  • Nokogiri::XML::NodeSet now supports more Array and Enumerable operators: index, delete, slice, - (difference), + (concatenation), & (intersection), push, pop, shift, ==
  • Nokogiri.XML, Nokogiri.HTML take blocks that receive Nokogiri::XML::ParseOptions objects
  • Nokogiri::XML::Node#namespace returns a Nokogiri::XML::Namespace
  • Nokogiri::XML::Node#namespace= for setting a node's namespace
  • Nokogiri::XML::DocumentFragment and Nokogiri::HTML::DocumentFragment have a sensible API and a more robust implementation.
  • JRuby 1.3.0 support via FFI.

Fixed

  • Fixed a problem with nil passed to CDATA constructor
  • Fragment method deals with regular expression characters (Joel) [LH#73]
  • Fixing builder scope issues [LH#61, LH#74, LH#70]
  • Fixed a problem when adding a child could remove the child namespace [LH#78]
  • Fixed bug with unlinking a node then reparenting it. [#22]
  • Fixed failure to catch errors during XSLT parsing [#32]
  • Fixed a bug with attribute conditions in CSS selectors [#36]
  • Fixed intolerance of HTML attributes without values in Node#{before/after/inner_html=}. [#35]

1.2.3 / 2009-03-22

Fixed

  • Fixing bug where a node is passed in to Node#new
  • Namespace should be assigned on DocumentFragment creation. [LH#66]
  • Nokogiri::XML::NodeSet#dup works [#10]
  • Nokogiri::HTML returns an empty Document when given a blank string [#11]
  • Adding a child will remove duplicate namespace declarations [LH#67]
  • Builder methods take a hash as a second argument

1.2.2 / 2009-03-14

Added

  • Nokogiri may be used with soap4r. See XSD::XMLParser::Nokogiri
  • Nokogiri::XML::Node#inner_html= to set the inner html for a node
  • Nokogiri builder interface improvements
  • Nokogiri::XML::Node#swap swaps html for current node [LH#50]

Fixed

  • Fixed a tag nesting problem in the Builder API [LH#41]
  • Nokogiri::HTML.fragment will properly handle text only nodes [LH#43]
  • Nokogiri::XML::Node#before will prepend text nodes [LH#44]
  • Nokogiri::XML::Node#after will append text nodes
  • Nokogiri::XML::Node#search automatically registers root namespaces [LH#42]
  • Nokogiri::XML::NodeSet#search automatically registers namespaces
  • Nokogiri::HTML::NamedCharacters delegates to libxml2
  • Nokogiri::XML::Node#[] can take a symbol [LH#48]
  • vasprintf for windows updated. (Geoffroy Couprie)
  • Nokogiri::XML::Node#[]= should not encode entities [LH#55]
  • Namespaces should be copied to reparented nodes [LH#56]
  • Nokogiri uses encoding set on the string for default in Ruby 1.9
  • Document#dup should create a new document of the same type [LH#59]
  • Document should not have a parent method [LH#64]

1.2.1 / 2009-02-23

Fixed

  • Fixed a CSS selector space bug
  • Fixed Ruby 1.9 String Encoding (角谷さん)

1.2.0 / 2009-02-22

Added

  • CSS search now supports CSS3 namespace queries
  • Namespaces on the root node are automatically registered
  • CSS queries use the default namespace
  • Nokogiri::XML::Document#encoding get encoding used for this document
  • Nokogiri::XML::Document#url get the document url
  • Nokogiri::XML::Node#add_namespace add a namespace to the node [LH#38]
  • Nokogiri::XML::Node#each iterate over attribute name, value pairs
  • Nokogiri::XML::Node#keys get all attribute names
  • Nokogiri::XML::Node#line get the line number for a node (Dirkjan Bussink)
  • Nokogiri::XML::Node#serialize now takes an optional encoding parameter
  • Nokogiri::XML::Node#to_html, to_xml, and to_xhtml take an optional encoding
  • Nokogiri::XML::Node#to_str
  • Nokogiri::XML::Node#to_xhtml to produce XHTML documents
  • Nokogiri::XML::Node#values get all attribute values
  • Nokogiri::XML::Node#write_to writes the node to an IO object with optional encoding
  • Nokogiri::XML::ProcessingInstruction.new
  • Nokogiri::XML::SAX::PushParser for all your push parsing needs.

Fixed

  • Fixed Nokogiri::XML::Document#dup
  • Fixed header detection. (rubikitch)
  • Fixed a problem where invalid CSS would cause the parser to hang

Deprecations

  • Nokogiri::XML::Node.new_from_str will be deprecated in 1.3.0

Changed

  • Nokogiri::HTML.fragment now returns an XML::DocumentFragment [LH#32]

1.1.1

Added

  • Added XML::Node#elem?
  • Added XML::Node#attribute_nodes
  • Added XML::Attr
  • XML::Node#delete added.
  • XML::NodeSet#inner_html added.

Fixed

  • Not including an HTML entity for \r for HTML nodes.
  • Removed CSS::SelectorHandler and XML::XPathHandler
  • XML::Node#attributes returns an Attr node for the value.
  • XML::NodeSet implements to_xml

1.1.0

Added

  • Custom XPath functions are now supported. See Nokogiri::XML::Node#xpath
  • Custom CSS pseudo classes are now supported. See Nokogiri::XML::Node#css
  • Nokogiri::XML::Node#<< will add a child to the current node

Fixed

  • Mutex lock on CSS cache access
  • Fixed build problems with GCC 3.3.5
  • XML::Node#to_xml now takes an indentation argument
  • XML::Node#dup takes an optional depth argument
  • XML::Node#add_previous_sibling returns new sibling node.

1.0.7

Fixed

  • Fixed memory leak when using Dike
  • SAX parser now parses IO streams
  • Comment nodes have their own class
  • Nokogiri() should delegate to Nokogiri.parse()
  • Prepending rather than appending to ENV['PATH'] on windows
  • Fixed a bug in complex CSS negation selectors

1.0.6

Fixed

  • XPath Parser raises a SyntaxError on parse failure
  • CSS Parser raises a SyntaxError on parse failure
  • filter() and not() hpricot compatibility added
  • CSS searches via Node#search are now always relative
  • CSS to XPath conversion is now cached

1.0.5

Fixed

  • Added mailing list and ticket tracking information to the README.txt
  • Sets ENV['PATH'] on windows if it doesn't exist
  • Caching results of NodeSet#[] on Document

1.0.4

Fixed

  • Changed memory management from weak refs to document refs
  • Plugged some memory leaks
  • Builder blocks can call methods from surrounding contexts

1.0.3

Fixed

  • NodeSet now implements to_ary
  • XML::Document should not implement parent
  • More GC Bugs fixed. (Mike is AWESOME!)
  • Removed RARRAY_LEN for 1.8.5 compatibility. Thanks Shane Hanna.
  • inner_html fixed. (Thanks Yehuda!)

1.0.2

Fixed

  • extconf.rb should not check for frex and racc

1.0.1

Fixed

  • Made sure extconf.rb searched libdir and prefix so that ports libxml/ruby will link properly. Thanks lucsky!

1.0.0 / 2008-07-13

Added

  • Birthday!