commonmark-java 0.24.0

Added

• SourceSpan on nodes now have a getInputIndex to get the index within the
  original input string (in addition to the existing line/column indexes).
  This is useful when looking up the input source: It can now be done using
  substring instead of having to split the input into lines first (#348)
• Configurable line break rendering for TextContentRenderer via lineBreakRendering
  on the builder; e.g. LineBreakRendering.SEPARATE_BLOCKS will render an empty
  line between blocks (#344)

Changed

• Adopted small changes from OpenJDK vendoring to make updates easier for them (#343)

Fixed

• Enable overriding of built-in node rendering for TextContentRenderer (#346)

commonmark-java 0.23.0

Added

• New extension for footnotes!
  • Syntax:

    Main text[^1]
    
    [^1]: Additional text in a footnote
    

  • Inline footnotes like ^[inline footnote] are also supported when enabled
    via an option in FootnotesExtension.Builder
  • Use class FootnotesExtension in artifact commonmark-ext-footnotes (#332)
• New option omitSingleParagraphP in HtmlRenderer.Builder for not using <p>
  tags for when a document only has one paragraph (#150)
• Support for custom link processing during inline parsing (e.g. [foo]),
  see Parser.Builder#linkProcessor
• Support for extending inline parsing with custom inline content parsers. See
  Parser.Builder#customInlineContentParserFactory. This allows users/extensions
  to hook into inline parsing on a deeper level than before (e.g. with delimiter
  processors). It can be used to add support for math/latex formulas or other inline
  syntax. (#321)

Changed

• The default DefaultUrlSanitizer now also allows data as a protocol. Use the
  constructor with a list to customize this. (#329)
• LinkReferenceDefinition now extends Block (it was extending Node
  directly before)
• MarkdownRenderer: Don't escape = text if it's the first node in a block (#335)

Fixed

• Fix parsing of link reference definitions with incorrect title syntax (followed
  by characters other than space/tab). In that case, the title was set to the
  partially-parsed title and the source spans were wrong. (#315)
• Fix source spans of blocks with lazy continuation lines (#337)
• MarkdownRenderer: Preserve thematic break literals (#331)

commonmark-java 0.22.0

Added

• New MarkdownRenderer for rendering nodes to Markdown (CommonMark)!
  Note that while care is taken to produce equivalent Markdown, some differences
  in the original Markdown (if parsed) are not preserved, such as:
  • The type of heading used
  • The type of link used (reference links will be rendered as inline links)
  • Whether special characters are escaped or not
  • Leading and trailing whitespace
• Modular JAR (JPMS): All artifacts now include module descriptors (module-info)
  so jlink can be used; the old Automatic-Module-Name manifest entries were removed
• New package org.commonmark.parser.beta containing classes that are not part of
  the stable API but are exported from the module because they might be useful for
  extension parsers
• New package org.commonmark.text for text related utilities that are useful for
  both parsing and rendering
• TableCell now has getWidth returning the number of dash and colon characters
  in the delimiter row, useful for rendering proportional width tables (#296)
• ThematicBreak now has getLiteral containing the string that was used in the
  source when parsing (#309)
• ListItem now has getMarkerIndent and getContentIndent for retrieving the
  space between the start of the line and the marker/content
• Deprecated a some properties of BulletList, OrderedList, FencedCodeBlock
  and replaced with nullable ones because they might not be set when constructing
  these nodes manually instead of via parsing

Changed

• Java 11 or later is now required (dropping support for Java 8)
• Update to CommonMark spec 0.31.2

Fixed

• Fix LinkReferenceDefinition having null SourceSpan when title is present
  and parsing with source spans option enabled (#310)

commonmark-java 0.21.0

Added

• GitHub strikethrough: With the previous version we adjusted the
  extension to also accept the single tilde syntax. But if you use
  another extension that uses the single tilde syntax, you will get a
  conflict. To avoid that, StrikethroughExtension can now be
  configured to require two tildes like before, see Javadoc. (#271)

commonmark-java 0.20.0

Fixed

• GitHub tables: A single pipe (optional whitespace) now ends a table
  instead of crashing or being treated as an empty row, for consistency
  with GitHub (#255).
• GitHub strikethrough: A single tilde now also works, and more than two
  tildes are not accepted anymore. This brings us in line with what
  GitHub actually does, which is a bit underspecified (#267)
• The autolink extension now handles source spans correctly (#209)

commonmark-java 0.19.0

Added

• YAML front matter extension: Limited support for single and double
  quoted string values (#260)

Changed

• Check argument of enabledBlockTypes when building parser instead of NPEing later

commonmark-java 0.18.2

Changed

• Test against Java 17
• Bundle LICENSE.txt with artifacts (in addition to Maven metadata)

commonmark-java 0.18.1

Fixed

• Fix tables with leading/trailing header pipes and trailing spaces (#244).
  This was a regression in 0.16.1 which is now fixed.

commonmark-java 0.18.0

Changed

• Update to CommonMark spec 0.30:
  • Add textarea to list of literal HTML block tags.
    Like script, style, and pre, textarea blocks can contain
    blank lines without the contents being interpreted as commonmark.
  • Fix case folding for link reference labels in some cases
    (e.g. ẞ and SS should match)
  • Allow lowercase ASCII in HTML declaration
  • Don't let type 7 HTML blocks interrupt lazy paragraphs either
• Preserve the original case for the label of LinkReferenceDefinition.
  Before, we used to store the normalized version (lowercase, collapsed whitespace).

commonmark-java 0.17.2

Changed

• Pass original instead of normalized label to InlineParserContext for lookup (#204).
  This allows custom contexts to change the lookup logic and have access to the original
  label content.
  In case you have a custom implementation of InlineParserContext, you might need to adjust
  it to do normalization.