Skip to content

Nokogiri Changelog

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

v1.17.1 / 2024-12-10

Fixed

  • Fixed a potential segfault when using Node#dup and DocumentFragment#dup. [#3359] @byroot @flavorjones
  • Node#dup and Node#clone now correctly decorate the new node with the document's Node decorators. [#3363] @flavorjones

v1.17.0 / 2024-12-08

Dependencies

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 context: keyword argument to the HTML5::DocumentFragment.parse and .new methods, which should allow for more flexible sanitization and future 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).

Keyword arguments

The following methods now accept keyword arguments in addition to positional arguments, and use ... parameter forwarding when possible: HTML4(), HTML4.fragment, HTML4.parse, HTML4::Document.parse, HTML4::DocumentFragment#initialize, HTML4::DocumentFragment.parse, HTML5(), HTML5.fragment, HTML5.parse, HTML5::Document.parse, HTML5::Document.read_io, HTML5::Document.read_memory, HTML5::DocumentFragment#initialize, HTML5::DocumentFragment.parse, XML(), XML.fragment, XML.parse, XML::Document.parse, XML::DocumentFragment#initialize, XML::DocumentFragment.parse, XML::Node#canonicalize, XML::Node.parse, XML::Reader(), XML::RelaxNG(), XML::RelaxNG.new, XML::RelaxNG.read_memory, XML::SAX::PushParser#initialize, XML::Schema(), XML::Schema.new, XML::Schema.read_memory, and XSLT().

Special thanks to those contributors who participated in the RubyConf 2024 Hack Day to work on #3323 to help modernize Nokogiri by adding keyword arguments and using parameter forwarding in many methods, and expanding some of the documentation! We intend to continue adding keyword argument support to more methods. #3323 #3324 #3326 #3327 #3329 #3330 #3332 #3333 #3334 #3335 #3336 #3342 #3355 #3356 @infews @matiasow @MattJones @mononoken @openbl @flavorjones

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 readonly attributes XML::DocumentFragment#parse_options and HTML4::DocumentFragment#parse_options return the options used to parse the document fragment. @flavorjones
  • New method XML::Reader.new is the primary constructor to which XML::Reader() forwards. Both methods now take url:, encoding:, and options: kwargs in addition to the previous calling convention of passing positional parameters. #3326 @infews @flavorjones
  • [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] 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 XML::RelaxNG, XML::Schema, XML::Reader, HTML5, HTML5::Document, HTML5::DocumentFragment, HTML4::Document, HTML4::DocumentFragment, XML, XML::Document, XML::DocumentFragment. #3355 @flavorjones
  • 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] 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
  • [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] When building from source on MacOS, environment variables AR and RANLIB are now respected when set instead of being overridden to /usr/bin/{ar,ranlib} (which is still the default). [#3338] @joshheinrichs-shopify

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] External entity references no long cause the SAX parser to register errors. [#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](https://github.com/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 libxml2 v2.13.0. 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 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.

Thank you!

The following people and organizations were kind enough to sponsor @flavorjones or the Nokogiri project during the development of v1.17.0:

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

v1.16.8 / 2024-12-02

Fixed

  • [CRuby] When serializing HTML5 documents, properly escape foreign content "style" elements. Normally, a "style" tag contains raw text that does not need entity-escaping, but when it appears in either SVG or MathML foreign content, the "style" tag is now correctly escaped when serialized. @flavorjones

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

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

Changed

  • [CRuby] XML::Reader sets the [@encoding](https://github.com/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

v1.16.1 / 2024-02-03

Dependencies

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:

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

Older releases

See misc/CHANGELOG-archive.md