CHANGELOG.md

v1.39.1

2026-04-23

Added

• Title.prototype.isRedirect

Fixed

• AttributeToken.prototype.lint now reports data attributes with underscores (_) in their names, which will be sanitized by the MediaWiki parser
• RedirectTargetToken.innerText should not be the final redirect target when there are double redirects
• TranscludeToken.prototype.removeArg when removing an anonymous parameter
• LinkBaseToken.prototype.setLinkText when the link text contains list-like syntax such as * or #
• ListToken.indent setter when there are trailing spaces

Changed

• Setting the stage parameter of Parser.parse to Infinity now implicitly calls Token.prototype.buildLists after parsing to build the list items

Removed

• The old schema of Config.doubleUnderscore is no longer supported

v1.38.1

2026-04-05

Added

• New category of linting errors: blank image alt attribute which is an accessibility issue

Fixed

• nowrap is now a valid but obsolete attribute for <td> and <th> tags

Changed

• LanguageService.prototype.destroy now returns a promise that resolves when the cleanup of temporary files is complete, but usually it is not necessary to await the promise
• LanguageService.prototype.provideDiagnostics now always skips <score> extension tags containing LilyPond Scheme

v1.37.1

2026-03-31

Added

• The preset configuration for Moegirlpedia is updated to include Extension:Math

Removed

• LanguageService.prototype.provideCompletionItems no longer suggests LilyPond commands that do not start with a backslash (\) in <score> extension tags

v1.37.0

2026-03-26

Added

• NowikiToken.prototype.lint now reports invalid JSONC inside <maplink> and <mapframe>

Changed

• The executable getParserConfig now executes JavaScript modules fetched from the specified MediaWiki site in a sandboxed environment using the permission model

v1.36.0

2026-03-02

Added

• TranscludeToken.prototype.anonCount

Changed

• Improved performance for template parsing with many anonymous parameters

v1.35.2

2026-02-23

Fixed

• The lonely-http rule should ignore http://, RFC, PMID or ISBN following word characters

v1.35.1

2026-02-08

Added

• AstElement.prototype.json now takes an optional depth parameter to limit the depth of the output JSON
• Suggestion for invalid gallery links

Fixed

• AstText.prototype.lint when there are both < and unmatched ]
• QuoteToken.prototype.lint stops offering suggestions for lonely apostrophes next to italic apostrophes which can be ambiguous

v1.35.0

2026-01-22

Fixed

• The executable getParserConfig now checks whether language conversion is disabled for a MediaWiki site

Removed

• Support for Node.js < 20.19.0

v1.34.0

2026-01-13

Added

• New category of linting errors: invalid JSON inside <templatedata>

Changed

• Setting TagPairToken.selfClosing to true now removes the inner content of the tag

v1.33.0

2026-01-01

Added

• AttributesToken.prototype.css now inserts the style attribute if not present
• ParamLineToken.prototype.lint now reports unknown parameter names
• LanguageService.prototype.provideCompletionItems now suggests parameter names in <dynamicpagelist> and <inputbox>
• Parse and report extension tags inside <categorytree>
• Parse <pre> tags with format="wikitext"
• ImageParameterToken.prototype.lint now reports invalid values for the link parameter in gallery images

Fixed

• AttributeToken.prototype.setValue for an attribute without a value
• AttributesToken.classList when the class attribute contains multiple spaces between class names
• AttributesToken.prototype.sanitize should leave spaces between attributes
• Token.prototype.createElement for extension tags
• Token.prototype.expand for #contentmodel when the first argument is canonical

v1.32.0

2025-12-05

Added

• HTML conversion of <syntaxhighlight> with Prism highlighting
• CategoryToken.innerText
• Token.prototype.toHtml now supports category links at the end of the page
• Parser.callParserFunction now supports custom parser functions defined via Parser.setFunctionHook
• Title.prototype.getFileUrl that internally calls the filepath parser function to generate a Special:Redirect/file/ URL
• Parser.lint, Parser.print and Parser.toHtml

Fixed

• AttributeToken.prototype.lint should report invalid HTML attributes except for id for <math> and <chem> tags
• Issue with Token.prototype.expand when expanding #tag
• Issue with Token.prototype.expand when expanding parser functions with template arguments
• Token.prototype.cloneNode should copy the pageName property
• ImageParameterToken.link can now be undefined for invalid links in gallery images

Changed

• FileToken.prototype.toHtml now calls Title.prototype.getFileUrl to generate the src attribute

Removed

• The old schema of Config.parserFunction is no longer supported

v1.31.0

2025-11-13

Added

• Token.prototype.expand now supports more parser functions
• Parser.callParserFunction
• New category of linting errors: template arguments inside an extension tag
• Token.prototype.expand now supports transclusion of redirect pages

Fixed

• Duplicate linting errors reported by <ref> tags
• FileToken.prototype.toHtml for inline images with vertical alignment parameters
• LinkBaseToken.prototype.toHtml for self links

v1.30.0

2025-11-06

Added

• HeadingToken.prototype.section
• QuoteToken.prototype.findMatchingQuote and QuoteToken.prototype.getRange
• NoincludeToken.prototype.lint now reports useless attributes like IncludeToken.prototype.lint
• New category of linting errors: redirect-like text in list items and header-like text in plain text

Fixed

• Expanding custom parser functions defined via Parser.setFunctionHook

v1.29.2

2025-10-30

Added

• TranscludeToken.prototype.newAnonArg and TranscludeToken.prototype.setValue now support an optional parameter to append the new parameter on a new line

Fixed

• Parsing empty <translate></translate> tag pairs

v1.29.1

2025-10-28

Added

• LanguageService.prototype.provideCompletionItems now suggests behavior switches with full-width underscores (＿)

Fixed

• AstText.prototype.lint now ignores magic link-like texts in tokens where magic links are not parsed
• AttributeToken.prototype.escape should ignore extension tag attributes
• Disable AstElement.prototype.escape for some token types

Changed

• LanguageService.prototype.provideCompletionItems now suggests behavior switches when typing the first underscore (_ or ＿)

v1.29.0

2025-10-23

Added

• Token.pageName property that gets or sets the page name
• Token.prototype.expand now supports the int parser function
• ImageParameterToken.thumb property that gets the thumbnail filename from the manualthumb image parameter
• New category of linting errors: invalid thumbnail filenames in image parameters

Fixed

• Parser.templateDir can now have a nested file structure to support subpages

Changed

• Extension:Math is now linted via texvcjs instead of MathJax
• A copy of the MediaWiki-flavored GeoJSON schema is now included in the package

v1.28.1

2025-10-16

Added

• Parse the details attribute of a <ref> extension tag
• New categories of linting errors, including <ref> tags in internal or external links, and invalid URLs
• Parse behavior switches with full-width underscores (＿＿)

v1.28.0

2025-09-30

Added

• Better support for <tvar> tags from Extension:Translate
• The parameter option of the invalid-gallery rule now also reports width parameters of images with pxpx units

Fixed

• TranslateToken.prototype.setAttr and TranslateToken.prototype.toggleAttr

Changed

• The default severity of the both and selfClosing options of the unmatched-tag rule is now error instead of warning

v1.27.0

2025-09-20

Added

• Language conversion in the alt parameter of images
• FileToken.prototype.lint now also reports missing file extensions

Fixed

• The id attribute of a section header starting with underscores (_)
• AttributeToken.prototype.lint should report evil attribute values such as javascript:

Changed

• The rule severity can now be specified as a string: 'off', 'warning' or 'error', which are equivalent to 0, 1 and 2, respectively

v1.26.0

2025-09-14

Added

• Parser.lintConfig now supports more properties, including rules, computeEditInfo, fix and ignoreDisables

Fixed

• TeX macro detection in LanguageService.prototype.provideDiagnostics

v1.25.1

2025-08-27

Added

• Parser.configPaths that is used to specify additional paths to look for parser configuration files

v1.25.0

2025-08-19

Added

• Descriptions for the quick fixes are now localized

Changed

• The message keys are all changed to follow the Mediawiki convention

v1.24.1

2025-08-14

Added

• AstNode.prototype.remove now takes an optional ownLine parameter to remove the current line if it is empty
• Quick fix for the nested-link rule
• LanguageService.prototype.provideRefactoringAction that provides refactoring actions for the selected text
• LanguageService.prototype.resolveCodeAction that resolves fix-all code actions
• The tag-like rule now supports custom tags

Fixed

• Now <translate> extension tags in HTML comments or <includeonly> tags are not parsed
• The bold-header rule now ignores bold apostrophes or tags inside extension tags
• No HeadingToken should be parsed inside a <poem> tag
• Quick fixes for the url-encoding rule should ignore special characters including !<>[]{|}
• <tvar> extension tags inside the URL of an external link is now correctly parsed

Changed

• Update the <img> extension tag
• According to the Wikimedia Foundation's policy, the getParserConfig executable, the Parser.fetchConfig method and the LanguageService.prototype.setTargetWikipedia method now requires a user parameter to identify the user when accessing Wikimedia sites

v1.23.0

2025-08-01

Added

• HtmlToken.prototype.getRange that returns the range of the HTML tag pair
• New properties and methods of AstRange: childNodes, firstChild, lastChild, children, firstElementChild, lastElementChild, childElementCount, offsetHeight, offsetWidth, getClientRects, remove, before, after, replaceWith, append, prepend, removeChild, getRootNode, closest, querySelector, querySelectorAll, getElementByTypes, getElementById, getElementsByClassName, getElementsByTagName, escape

Fixed

• Updated tag attributes for the <ref> extension tag

v1.22.1

2025-07-25

Added

• LanguageService.prototype.provideDiagnostics now includes available quick fixes for Stylelint errors

Fixed

• AstRange.endIndex when the range is not empty

Changed

• The preset configuration for Moegirlpedia is updated to support the latest version of Extension:MoeImgTag

v1.22.0

2025-07-22

Added

• Now the rule configuration for Token.prototype.lint can be set via Parser.lintConfig
• Parser.getWMFSite that returns the name and host name of a MediaWiki sites hosted by Wikimedia
• LanguageService.prototype.setTargetWikipedia now supports other WMF-hosted sites
• Title.prototype.toSubjectPage, Title.prototype.toTalkPage, Title.prototype.toBasePage and Title.prototype.toRootPage now return the Title object itself for chaining
• TranscludeToken.prototype.getFrame that returns a basic JSON frame object for the #invoke parser function
• Parser.setFunctionHook that defines a custom parser function hook for the Token.prototype.expand method
• Parser.setHook that defines a custom extension tag hook for the Token.prototype.expand method
• Title.displayTitle

Fixed

• Parser.normalizeTitle now correctly handles titles with nested transclusions
• Conversion rules can have upper-case letters in their variant names

Changed

• The preset configurations are updated
• For Moegirlpedia, <img> tags are now parsed as extension tags (Extension:MoeImgTag) instead of HTML tags
• Lowered severity of invalid gallery image parameters and duplicate categories
• Lowered severity of unmatched HTML tags
• Lowered severity of plain-text apostrophes next to bold/italic apostrophes
• Lowered severity of unclosed apostrophes in section headers
• Lowered severity of fostered table content
• Lowered severity of double square brackets or curly braces
• Lowered severity of lonely http:// or https://

v1.21.3

2025-07-14

Fixed

• HTML conversion of <poem> containing <nowiki> or leading :
• HTML conversion of <h1> to <h6> without and id attribute
• HeadingToken as the first token of a <poem> tag does not need to be on a new line
• Sanitization of AttributeToken
• Use the legacy rgba() function instead of #rrggbbaa in LanguageService.prototype.provideColorPresentations for colors with an alpha channel
• NowikiToken.prototype.cloneNode should copy the name attribute if it exists

Changed

• The object parameter of Parser.createLanguageService is now optional

v1.21.2

2025-06-12

Added

• TranscludeToken.function
• Some time-related variables are now expanded, e.g., CURRENTMONTH and LOCALMONTH
• Parser.now property that specifies the current time for parsing

Fixed

• An issue with one-line definition lists (;:) when ; is followed by whitespace
• An occasional issue with multi-level indentation in a one-line definition list (;;dt::dd)

Changed

• TranscludeToken.prototype.print now highlights the modifiers (e.g., subst:) as magic words
• AttributeToken.prototype.toHtml now drops empty id attributes
• Lowered severity of unnecessary URL encoding in internal links

v1.21.1

2025-06-07

Added

• AstElement.prototype.print now adds a red wavy underline to invalid wikitext
• Token.prototype.expand now supports percent-encoded file names (e.g., %3F for ?)

Fixed

• Interaction between one-line definition lists (;:) and void HTML tags

v1.21.0

2025-05-29

Added

• Token.prototype.lint called on the root node now returns an optional output property, which is a string after all auto-fixes are applied
• TranscludeToken.module

v1.20.3

2025-05-12

Fixed

• Wrong message keys in simplified Chinese localization

Changed

• Lowered severity of useless fragments
• AstText.prototype.lint now ignores the group attribute of a <ref> or <references> tag
• Lowered severity of lonely language conversion brackets
• Lowered severity of an unexpected template argument

v1.20.2

2025-05-09

Changed

• Lowered severity of a duplicated image caption consisting of a template or a magic word

v1.20.1

2025-04-26

Fixed

• AttributesToken.name should be td, th or caption for table cells
• AttributeToken.tag should be td, th or caption for table cells

Changed

• Lowered severity of a lonely apostrophe followed by a closing bold/italic mark
• Lowered severity of a lonely bracket surrounding an internal or external link
• AstText.prototype.lint now ignores the name attribute of a <ref> tag
• Lowered severity of lonely RFC, PMID or ISBN
• Lowered severity from RedirectTargetToken.prototype.lint
• Lowered severity of multiline comments inside <gallery>
• Lowered severity of disallowed HTML tags

v1.20.0

2025-04-16

Added

• LanguageService.prototype.provideCompletionItems now supports extension tags <math> and <chem>
• LanguageService.prototype.provideDiagnostics now reports unknown macros in extension tags <math> and <chem>
• Accepted attribute names for extension tags <math>, <chem>, <phonos>, <section>, <syntaxhighlight>, <timeline> and <quiz>
• Parse the extension tag <hiero>
• Parse Extension:Translate

Fixed

• External links starting with // in <imagemap> are valid
• Internal links with empty alt text in <imagemap> are valid
• The executable getParserConfig for MediaWiki sites not hosted by Wikimedia
• More potential stack overflow issues
• Do not suggest more than 6 consecutive = when fixing unbalanced = in a section header
• A unicode number before a URL protocol prevents it from being parsed as an external link

v1.19.0

2025-04-07

Added

• LanguageService.prototype.provideDiagnostics now supports extension tags <math> and <chem> if MathJax is available

Fixed

• ImageParameterToken containing comments that do not break the parameter syntax

v1.18.4

2025-04-03

Added

• Parser.fetchConfig that executes the getParserConfig executable
• TranscludeToken.prototype.getPossibleValues now supports the #switch parser function

Fixed

• More cases where potential memory leak may occur
• Expansion of #switch
• TableToken.prototype.lint now handles arguments and a set of magic words recursively
• Stack overflow in TableToken.prototype.getLayout
• Issue with AstText.prototype.lint when the text contains certain unicode characters

Changed

• Lowered severity of fostered content of #invoke in TableToken.prototype.lint

v1.18.3

2025-03-24

Added

• AstElement.prototype.escape method that escapes = and | characters for building a template parameter

Fixed

• Stack overflow when there are too many tokens in a page
• TableToken.prototype.lint should not report appropriate HTML tags (<tr>, <td>, <th> and <caption>) as fostered-content

Changed

• ParameterToken.prototype.json now includes the duplicated property only if it is true
• TdToken.prototype.json now includes the rowspan and colspan properties only if they are greater than 1
• FileToken.prototype.lint now reports warnings instead of errors for duplicated caption parameters of an unknown file type

v1.18.2

2025-03-15

Added

• Config.functionHook property
• LanguageService.prototype.provideLinks now supports the filepath parser function
• New preset configuration jawiki for Japanese Wikipedia
• The executable getParserConfig now supports more MediaWiki sites with the CodeMirror extension installed

Fixed

• Only function hooks are allowed to take parameters
• Localized parser functions with a full-width colon ： instead of a half-width colon :

Changed

• LanguageService.prototype.resolveRenameLocation no longer supports magic words

v1.18.1

2025-03-13

Added

• LanguageService.prototype.setTargetWikipedia that updates the parser configuration for a specific Wikipedia language edition
• LanguageService.prototype.provideDocumentColors now supports color names in style attributes
• LanguageService.prototype.provideHover now supports HTML tags and attributes
• AttributeToken.prototype.lint now reports invalid attribute values of HTML tags

v1.18.0

2025-03-11

Added

• LanguageService.prototype.provideCompletionItems now provide documentations for suggested magic words
• LanguageService.prototype.provideDiagnostics and LanguageService.prototype.provideCompletionItems now support the <score> extension tag written in LilyPond
• New executable getParserConfig

Fixed

• Unhandled promise rejection in LanguageService if Stylelint is unavailable
• Manual upright parameters (e.g., upright=$1) of images are not treated as upright parameters
• The magic words #section, #section-x and #section-h are recognized as aliases of #lst, #lstx and #lsth, respectively

Changed

• The behavior switch __DISAMBIG__ now has the name DISAMBIGUATION, and the behavior switch __EXPECTED_UNCONNECTED_PAGE__ now has the name EXPECTEDUNCONNECTEDPAGE

v1.17.1

2025-03-08

Added

• Token.prototype.buildLists that builds all list-range tokens
• LanguageService.prototype.provideDiagnostics, LanguageService.prototype.provideHover and LanguageService.prototype.provideCompletionItems now work for embedded JSON in extension tags (templatedata, mapframe and maplink) and inline CSS
• LanguageService.prototype.provideFoldingRanges now works for embedded JSON in extension tags (templatedata, mapframe and maplink)
• Token.prototype.lint now reports errors and warnings for inline CSS
• AttributeToken.prototype.css that gets or sets the value of a CSS property
• AttributeToken.prototype.json now includes the tag property
• LanguageService.include property that configures the transclusion option for parsing

Fixed

• Token.prototype.findEnclosingHtml
• LanguageService.prototype.provideCompletionItems for attribute keys of extension tags, HTML tags and tables
• Various issues with Token.prototype.cloneNode
• Title.prototype.getUrl now ignores an empty fragment

Changed

• HtmlToken.prototype.findMatchingTag now returns undefined if there is no matching tag instead of throwing a syntax error

v1.16.6

2025-02-26

Fixed

• Potential memory leak when parsing a page with a large number of transclusions

Changed

• A LanguageService instance can provide both signature help and other services

v1.16.5

2025-02-25

Added

• English translation for JSDoc annotations
• LinkBaseToken.prototype.json now includes the fragment property
• CategoryToken.prototype.json now includes the sortKey property
• FileToken.prototype.json now includes the extension property
• ListToken.indent property
• ListToken.prototype.json now includes the indent property
• TdToken.prototype.json now includes the rowspan and colspan properties
• ArgToken.prototype.json now includes the default property
• ConverterRuleToken.prototype.json now includes the variant property
• GalleryToken.prototype.json now includes the widths and heights properties
• ParameterToken.prototype.json now includes the duplicated property

Fixed

• LanguageService.prototype.provideHover for variables without a parameter
• Token.prototype.toHtml when there are mixed apostrophes and <b>/<i> tags

v1.16.4

2025-02-21

Added

• TranscludeToken.prototype.getModule that returns the module name and the function name of a #invoke parser function
• LanguageService.prototype.provideCompletionItems now suggests parameter keys for the #invoke parser function

Fixed

• LanguageService.prototype.provideCompletionItems should not suggest argument names, link targets, template names, parameter keys or image widths at the cursor position
• LanguageService.prototype.provideCompletionItems should suggest with a leading : for the transclusion of a main namespace article
• AstElement.links should not return internal links inside the link parameter of an image twice
• AstNode.font for nodes following an external link
• AstNode.font when there are unmatched <b> or <i> tags
• LanguageService.prototype.provideRenameEdits should not rename the name attribute of a <ref> tag with a different group attribute

Changed

• LanguageService.prototype.provideDefinition now returns the Location of the <ref> inner content for any Position within a <ref> tag that has a name attribute
• LanguageService.prototype.provideCompletionItems inserts a > character after completing a closing tag when necessary
• LanguageService.prototype.provideCompletionItems inserts a = character after completing a template parameter key when necessary

v1.16.3

2025-02-16

Added

• LanguageService.prototype.provideInlayHints that computes inlay hints for template anonymous parameters
• AstNode.prototype.getLines that returns an array of the source text, the start and end positions of each line
• Signatures of more parser functions from extensions

Fixed

• LanguageService.prototype.provideDiagnostics now reports warnings
• LanguageService.prototype.resolveRenameLocation should ignore file and category links
• The outputs of wikiparse.Linter.prototype.codemirror and wikiparse.Linter.prototype.monaco now strictly follow the CodeMirror and Monaco Editor specifications, respectively

Changed

• The return values of AstElement.prototype.caretPositionfromIndex and its related methods are now different if the index is at the boundary between two tokens
• AstElement.prototype.elementFromIndex and its related methods now return Token instead of AstNode

v1.16.1

2025-02-10

Added

• Parser.createLanguageService will create a LanguageService instance that provides various functions
• Config.articlePath does not have to include '$1' now

Fixed

• "sideEffects": false is now added to package.json to facilitate tree-shaking
• {{名字空间}} and {{命名空間}} are now correctly recognized as {{NAMESPACE}}

v1.15.1

2025-01-30

Fixed

• < and > in external links

Changed

• Full HTML entity decoding via entities

Removed

• Interwiki prefixes are no longer available from the preset configuration files

v1.15.0

2025-01-15

Fixed

• Case-insensitive variables without a parameter, e.g., #language, and case-sensitive parser functions requiring a parameter, e.g., DEFAULTSORT

Changed

• The schema for parser configurations is changed, including changes for the parserFunction and doubleUnderscore properties and a new variable property

v1.14.1

2025-01-11

Added

• Named parameters are recognized for the #tag parser function

v1.14.0

2024-12-13

Added

• More auto-fix and suggestions for linting errors

Removed

• Removed category of linting errors: duplicated <ref> name

v1.13.9

2024-12-06

Added

• New category of linting errors: lonely RFC/PMID/ISBN
• LintError.Fix now has a new string property desc

Changed

• The preset configurations are updated

v1.13.7

2024-12-01

Fixed

• Token.prototype.lint should ignore <ref> tags with identical names but in different groups

v1.13.6

2024-11-29

Added

• New category of linting errors: duplicated <ref> name

v1.13.4

2024-11-17

Fixed

• TranscludeToken.name for some new parser functions, e.g., #dir

v1.13.3

2024-11-14

Fixed

• AttributeToken.prototype.lint now correctly reports invalid attribute names of <pre> tags

v1.13.2

2024-11-04

Fixed

• Parse and report extension tags inside <inputbox>/<dynamicpagelist>
• Parse transclusion inside <references>
• {{\n=a=\n}} is now parsed as Template:=a= instead of <p>{{</p><h1>a</h1><p>}}</p>

v1.13.1

2024-11-01

Fixed

• Case-insensitive <nowiki> inside a <pre> tag
• HTML comments inside a <inputbox> tag

Changed

• IncludeToken.prototype.lint now also reports useless attributes

v1.13.0

2024-10-24

Added

• New pseudo selectors :valid and :scope
• Pseudo selector :has() now supports relative selectors
• Parser.parse now accepts an array of token types as the third argument to specify the maximum stage of parsing

Fixed

• Token.prototype.expand now supports file names containing ꞉ as a substitute for : in a Windows file system
• Order of HTML/URL decoding in Title parsing

Changed

• AttributesToken.prototype.lint now reports warnings instead of errors for useless non-word characters

v1.12.7

2024-08-24

Fixed

• TranscludeToken.prototype.lint should not report # in an HTML entity as template page fragment

v1.12.6

2024-08-17

Fixed

• AttributeToken.prototype.lint for unknown extension tags, since v1.12.5

v1.12.5

2024-08-16

Added

• GalleryToken.prototype.toHtml
• GalleryToken.widths and GalleryToken.heights

Fixed

• AttributesToken.prototype.lint now correctly reports invalid attribute names containing {{!}}

v1.12.4

2024-08-07

Added

• HeadingToken.id
• New category of linting errors: duplicated id

Fixed

• HeadingToken.prototype.toHtml should skip language conversion for the id attribute
• HeadingToken.prototype.toHtml when there are more than 3 duplicated IDs
• Descendant combinator

Changed

• ArgToken.prototype.json no longer includes the default property

v1.12.3

2024-08-04

Fixed

• Fix a rare case of LinkToken.prototype.toHtml for the Media namespace

v1.12.2

2024-07-29

Added

• Title.prototype.getRedirection
• TranscludeToken.prototype.toHtml for nonexistent transcluded pages and transclusion loops

Fixed

• HeadingToken.prototype.toHtml now correctly handles headings with duplicated IDs
• Remove unnecessary implicit leading newline from Token.prototype.expand
• Remove unexpected leading space for fostered table content in TableToken.prototype.toHtml
• A valid title should not contain characters that are escaped as HTML entities twice (e.g., &)
• Token.prototype.solveConst for #switch with a fall-through default case
• Token.prototype.solveConst for #ifeq when comparing two numbers or strings containing HTML entities
• Token.prototype.toHtml for various token types that have a bold and/or italic style RedirectTargetToken.innerText should replace underscores with spaces
• Title.fragment should replace spaces with underscores
• Some Wikitext syntaxes at SOL, including headings and lists, are now allowed following behavior switches except __toc__
• <p> wrapping for TdToken.prototype.toHtml
• Media namespace redirects
• Token.prototype.toHtml when <blockquote> interferes with indent-<pre>
• AttributeToken.prototype.lint should also report reserved data attributes

Changed

• AstText.prototype.lint no longer reports lonely-http for http:// in ext-link-text
• Token.prototype.expand now removes all comments and expands ArgToken
• HeadingToken now includes trailing blank lines
• By turning Parser.redirects and Title.fragment into accessors, page titles and fragments are now automatically normalized
• The second arguments of AstText.prototype.deleteData and AstText.prototype.substringData are now optional
• The first item of the doubleUnderscore property of Config is now Record<string, string>

v1.11.1

2024-07-10

Fixed

• Lonely px should be considered as caption in ImageParameterToken

v1.11.0

2024-07-08

Added

• <p> wrapping for Token.prototype.toHtml
• Conversion from Wikitext lists to HTML
• ConverterToken.prototype.toHtml when there are variant flags
• FileToken.prototype.toHtml, FileToken.prototype.getFrame, FileToken.prototype.getHorizAlign and FileToken.prototype.getVertAlign
• ImageParameterToken.prototype.getUrl

Fixed

• ConverterFlagsToken.prototype.getEffectiveFlags when there are variant flags
• ExtLinkToken.prototype.toHtml when there are nested internal links
• ArgToken.prototype.cloneNode, AttributeToken.prototype.cloneNode, AttributesToken.prototype.cloneNode, GalleryToken.prototype.cloneNode and MagicLinkToken.prototype.cloneNode
• Implicit newline in Token.prototype.expand
• Token.prototype.expand for template redirects
• Title.prototype.getUrl for self links
• Definition list parsing since v1.10.0
• LinkBaseToken.prototype.toHtml when wrapped by apostrophes
• ExtLinkToken.prototype.toHtml and MagicLinkToken.prototype.toHtml for invalid URLs
• Title parsing with multiple consecutive spaces and/or underscores
• Self-closing <ref> tags inside a <references> tag
• HtmlToken should not be bold or italic
• ImageParameterToken.size when the parameter ends with pxpx
• ImageParameterToken.link when containing QuoteToken
• Image parameter parsing for link= followed by an invalid external link
• Framed images and manual thumbnails are always unscaled

Changed

• ListBaseToken.prototype.toHtml now returns an empty string
• ListBaseToken.prototype.getRange now modifies the AST and returns a plain token with the type list-range

Removed

• Token.prototype.redoQuotes method, which does not work as expected

v1.10.0

2024-06-26

Added

• AstNode.prototype.is
• Token.prototype.expand
• Parser.templateDir and Parser.templates properties
• Config.articlePath property
• Title.prototype.getUrl
• RedirectTargetToken.innerText and MagicLinkToken.innerText properties
• AstNode.prototype.toHtml method, which is an incomplete implementation of Wikitext-to-HTML conversion

Fixed

• SyntaxToken.prototype.lint now also reports errors from its children
• Parse the caption attribute of a <gallery> extension tag
• Fix LinkToken.prototype.pipeTrick
• Absolute path on Windows in Parser.config and Parser.i18n
• Token.prototype.cloneNode when applied to a root token
• Pseudo selector :any-link for RedirectTargetToken
• AttributesToken.prototype.lint now reports duplicated class attributes

Changed

• LinkBaseToken.prototype.setTarget now only adds leading : when necessary
• MagicLinkToken.prototype.lint now only reports the first occurrence of full-width punctuations in free-ext-link as an error
• TranscludeToken.prototype.lint now ignores the fragment of an invalid module name
• Token.prototype.solveConst now returns a copy of the original token
• MagicLinkToken.prototype.getUrl now returns a string for ISBN links
• Each ListToken now contains at most one ;
• AttributeToken.prototype.getValue now always trims leading and trailing whitespace

Removed

• RedirectTargetToken.prototype.text method overriding
• rbspan attribute of a <rt> tag

v1.9.3

2024-06-18

Added

• AstElement.length and AstText.length are now writable

Fixed

• Fix event handling for DoubleUnderscoreToken

Changed

• No longer report full-width punctuations in ext-link-url as an error in MagicLinkToken.prototype.lint

v1.9.2

2024-06-10

Fixed

• Fix event handling since v1.9.0
• Fix issues related to Parser.viewOnly

v1.9.1

2024-06-09

Fixed

• AstNode.font, AstNode.bold and AstNode.italic for external links which have a lower precedence than apostrophes
• Pseudo selector :any-link for FileToken
• Pseudo selector :lang()
• Pseudo selector :regex() for AttributesToken

Changed

• Token.prototype.normalizeTitle is now a private method
• Parser.getConfig is now available for public use

Removed

• Pseudo selectors :nth-child, :nth-last-child, :nth-of-type and :nth-last-of-type no longer support comma-delimited arguments

v1.9.0

2024-05-26

Added

• Magic links (RFC, PMID and ISBN) are now parsed as MagicLinkToken
• New property Parser.viewOnly, which helps to speed up AstElement.prototype.lint

Changed

• Improved performance for table parsing inside a template

v1.8.0

2024-05-20

Added

• New getters AstNode.font, AstNode.bold and AstNode.italic

Fixed

• Token.prototype.findEnclosingHtml now correctly handles self-closing tags
• \r is now automatically removed from EOL for CRLF (\r\n) line breaks

Changed

• Token.prototype.sections now returns an array of AstRange objects
• AstNode.nextVisibleSibling, AstNode.previousVisibleSibling, AstNode.prototype.destroy and AstNode.prototype.getLine were previously only available for AstElement, but now they are also available for AstText

v1.7.1

2024-05-13

Added

• Parse redirect pages

Fixed

• Allow an external image as ext-link-text
• Ignore duplicated parameters containing extension tags in TranscludeToken.prototype.lint
• Self-closing tags <noinclude/> and <includeonly/> are now valid

Changed

• For redirects with a fragment, the fragment is now ignored by Title.title

v1.6.2

2024-03-17

Added

• Parser tests are now available

Fixed

• Mimic the structure of CSS combinators
• Fully escape < and > when printing tag attributes
• Recognize invalid Title patterns
• Image size parameter with a suffix of pxpx is now also valid
• Fix conflicts between image parameters and external links
• Fix free external links led by a Unicode character

Changed

• Error messages are gradually translated into English

v1.6.1

2024-03-13

Added

• Linting errors can now be disabled using inline comments

Fixed

• Allow the image parameter page in PDF file links
• Disallow empty strings as the manual-upright image parameter
• Use parseInt() for the rowspan and colspan properties of TdToken

Changed

• The start position of a linting error associated with a TrToken is now one character ahead

v1.5.7

2024-03-12

Fixed

• TableToken.prototype.lint now ignores inconsistent table layout due to a large colspan value in the last column

Changed

• Lowered severity of ignored lines due to no image in GalleryToken.prototype.lint

v1.5.6

2024-03-07

Fixed

• Parsing a TdToken after another TdToken with a multiline HtmlToken in it
• ExtLinkToken when the link text starts with a token with higher precedence

Changed

• Title.fragment no longer decodes the deprecated MediaWiki-style fragment encoding

v1.5.5

2024-03-06

Fixed

• ListToken.prototype.getRange for <dt> items

Changed

• Attribute selectors now regard falsy values as non-existent attributes
• HeadingToken.innerText and TdToken.innerText now trim leading and trailing whitespace, while LinkToken.innerText trims leading whitespace only

Removed

• ImagemapToken no longer overrides the links getter

v1.5.4

2024-03-05

Fixed

• Ignore most invalid Wikitext syntax nested in a PreToken when linting

v1.5.3

2024-03-03

Added

• New getter FileToken.extension

Fixed

• DdToken is now correctly parsed when QuoteToken is present in the same line
• Improved ImageParameterToken validation for different file extensions
• Ignore incomplete Wikitext syntax in before and after attributes of a <choose> extension tag when linting

v1.5.1

2024-02-29

Added

• New property rules of the interface Parser

Fixed

• DdToken is now correctly parsed when HtmlToken is present in the same line

Changed

• filter in the style attribute is no longer reported as an error

Removed

• Token type aliases have been malfunctioning since v1.4.3, and are now removed

v1.5.0

2024-02-22

Added

• Auto-fix and suggestions for selected linting errors

Removed

• AstElement.prototype.echo method

v1.4.5

2024-02-12

Added

• New property rule of the interface LintError

v1.4.4

2024-02-04

Added

• New category of linting errors: inconsistent table layout

Fixed

• Inconsistent output from AttributesToken.prototype.print

Changed

• HtmlToken.prototype.print and ImageParameterToken.prototype.print do not differentiate invalid tokens any more
• HtmlToken.prototype.text now treats void tags as self-closing tags

v1.4.3

2024-01-31

Added

• New field range in the return value of AstElement.prototype.json

Fixed

• Missing type declarations for ExtLinkToken.firstChild and ConverterToken.firstChild are added

v1.4.2

2024-01-29

Fixed

• AstText.prototype.lint no longer reports warnings for matched single brace or bracket

v1.4.1

2024-01-27

Added

• New category of linting errors: <ref> or external links without specified text in anchors
• New preset configuration enwiki for English Wikipedia

Fixed

• Token.prototype.redoQuotes

Changed

• Reported duplicated categories do not need to be child nodes of the same parent node
• Better recognition of bracket pair ([]) that actually breaks the ExtLinkToken

v1.4.0

2024-01-23

Added

• Linting errors for LinkToken are now also reported from CategoryToken and FileToken where applicable
• New category of linting errors: duplicated categories, and bold in headers

Fixed

• Type declarations

v1.3.9

2024-01-22

Added

• New categories of linting errors, including obsolete HTML tags and attributes, internal links in external links, and unmatched apostrophes inside a header
• New properties for QuoteToken: bold and italic

Fixed

• Parser.isInterwiki now returns null if no interwiki prefix is found from the configuration
• Title starting with multiple colons is invalid

v1.3.8

2024-01-21

Changed

• AstElement.prototype.json now records AstText as an object instead of a plain string, and some getter properties including ArgToken.default, HeadingToken.level, HtmlToken.closing, HtmlToken.selfClosing, ParameterToken.anon, TableToken.closed, and TdToken.subtype are now supported

v1.3.7

2024-01-19

Fixed

• Trailing whitespace as the last ConverterRuleToken is now allowed

Changed

• Double pipes in TdToken will now report an 'error' instead of a 'warning'

v1.3.6

2024-01-17

Changed

• AttributesToken.prototype.setAttr now also accepts an attribute object as argument

v1.3.4

2024-01-16

Fixed

• Case-insensitive magic variables like server
• startIndex and endIndex of MagicLinkToken.prototype.lint

v1.3.3

2024-01-15

Fixed

• A case-sensitive parser function requires a colon after the function name
• External links like [[//xyz]]
• Invalid closing void HTML tag except </br>

v1.3.2

2024-01-15

Added

• Reporting unclosed IncludeToken
• Reporting lonely <onlyinclude>, <noinclude> and <includeonly>
• Reporting errors such as <pre> inside PreToken

v1.3.1

2024-01-14

Fixed

• Parsing an illegal template name, e.g., {{#if:{{<}}}}

v1.3.0

2024-01-11

Changed

• Ingoring 'http' in tag attribute values when linting

Removed

• excerpt field in LintError

v1.2.0

2024-01-04

Fixed

• Allowing < and > in the fragment of an internal link

Changed

• HiddenToken child nodes of GalleryToken are replaced by NoincludeToken
• The pseudo selector :invalid now reports invalid ImageParameterToken instead of redundant child nodes of ArgToken

Removed

• Pseudo selectors :read-only and :read-write

v1.1.6

2024-01-03

Fixed

• Fully localized message in LintError

v1.1.5

2023-12-19

Fixed

• Magic words ! and = are now required in parser configurations

v1.1.4

2023-12-18

Added

• New methods AstText.prototype.escape, ParameterToken.prototype.escape and MagicLinkToken.prototype.escape

Fixed

• Removing unexpected remove and insert events dispatched by Token.prototype.safeReplaceWith
• {{!}} and {{=}} can be part of an external link now

Changed

• Inserting a child node is now forbidden

v1.1.3

2023-12-16

Fixed

• Fixing Token.prototype.sections since v1.1.2

Changed

• Token.prototype.findEnclosingHtml now returns an AstRange object

v1.1.2

2023-12-13

Changed

• Token type converter-rule-noconvert now regarded as converter-rule-to
• Anonymous parameters will remain anonymous after ParameterToken.prototype.setValue

Removed

• HeadingToken.name property

v1.1.1

2023-12-11

Changed

• getAttr as the preferred method for attribute selector

v1.1.0

2023-12-11

Added

• New properties and methods for Title objects: extension, toSubjectPage, toTalkPage, isTalkPage, toBasePage, toRootPage

Fixed

• Wrapping the text after the last </onlyinclude> in a NoincludeToken
• Replacing remaining spaces in Title.title property with underscores
• AstRange now maintains its content after insertNode, deleteContents, extractContents and cloneContents methods

Changed

• startContainer and endContainer properties of AstRange must be siblings now

Removed

• AstRange.prototype.intersectsNode method

v1.0.3

2023-12-05

First TypeScript version