Skip to content

VERSION.md

Latest commit

 

History

History
1211 lines (1014 loc) · 62.2 KB

VERSION.md

File metadata and controls

1211 lines (1014 loc) · 62.2 KB

flexmark-java

 

Version History

 

Future 0.50.0

  • Add: discourse for flexmark-java feature discussions
  • Fix: FormattingAppendable deprecate all conditional formatting related methods and methods only needed because of convoluted output construction.
  • Deprecate: all conditional formatting methods in FormattingAppendable
  • Break: make Java 8 minimum version and use JDK 8 for compilation
  • Fix: Factor out BasedSequenceImpl functionality that does not depend on BasedSequence and can be applied to any CharSequence into its own CharSequence interface with default implementations.
  • Add: LineFormattingAppendable to extend and eventually replace FormattingAppendable
    1. provide same append(...), line(), blankLine(), indent(), prefix(), etc. of FormattingAppendable but leading, trailing whitespace options, applied when current line is complete not during construction.
    2. add append(Collection<CharSequence>) for easy appending of child rendered results
    3. no callback methods and convoluted logic implementing indentations during content construction since these can be easily applied to individual lines once they are generated.
    4. provide implement List<CharSequence> interface for its content lines
    5. provide currentLine() for last content line which does not yet have an EOL
    6. provide currentLineIndex() as int for the index of the currentLine() value, will be equal to size() if the current line has no content, ie. end of output after previous line terminated with EOL.
    7. provide result() as Collection<CharSequence> where each char sequence represents a single line of output.
  • Add: FlexmarkHtmlParser options:

0.40.22

  • Fix: merge util tests from @James-Adam and fix bugs
  • Fix: change to MutableDataSet.set(DataKey<? extends T>, T)

0.40.20

  • Fix: #316, Github user extension incorrectly formats some text

    • Add: test to make sure previous character to @ is not isUnicodeIdentifierPart(), - nor .

  • Add: FlexmarkHtmlParser options:

  • Add: Remove reliance on YouTrack: IDEA-207453 and instead change resource file URL to path in SpecReader and add a message to all example tests with the file URL with :xxx where xxx is the line number of the spec example in the file.

    This will print the source location of the test in the console. Clicking the link goes right to the spec source file and line of the failed test if Awesome Console plugin is installed.

0.40.18

  • Add: EmbeddedAttributeProvider documentation and add it to the provider's list by default unless HtmlRenderer.EMBEDDED_NODE_PROVIDER is set to false. Add attributes to nodes in the AST by inserting a EmbeddedAttributeProvider.EmbeddedNodeAttributes node with the desired attributes. See: NodeInsertingPostProcessorSample.java for example.

  • Add: document that DocxRenderer emoji with GitHub preferred is not able to download images, compiling ImageUtils library with Java 8 eliminates the problem.

  • Add: resource file URL to SpecReader and add a message to all example tests with the file URL with :xxx where xxx is the line number of the spec example in the file.

    This will print the source location of the test in the console. Clicking the link goes right to the spec source file and line of the failed test. Unfortunately this URL will only work in IntelliJ when JetBrains add this to console to handle such file:// URLs: YouTrack: IDEA-207453

  • Fix: #310, PR: Change URL of GitHub CDN thanks to benelog

  • Fix: AsideExtension option keys to be dynamic data keys dependent on corresponding Parser block quote options for their defaults.

    ⚠️ This can potentially break code relying on versions of the extension before 0.40.18 because parsing rules can change depending on which block quote options are changed from their default values.

    To ensure independent options for aside blocks and block quotes, set aside options explicitly. The following will set all aside options to default values, independent from block quote options:

    .set(EXTEND_TO_BLANK_LINE, false)
.set(IGNORE_BLANK_LINE, false)
.set(ALLOW_LEADING_SPACE, true)
.set(INTERRUPTS_PARAGRAPH, true)
.set(INTERRUPTS_ITEM_PARAGRAPH, true)
.set(WITH_LEAD_SPACES_INTERRUPTS_ITEM_PARAGRAPH, true)

  • Deprecate: Parser.BLOCK_QUOTE_TO_BLANK_LINE, use more mnemonic Parser.BLOCK_QUOTE_EXTEND_TO_BLANK_LINE

  • Deprecate: CustomNode and CustomBlock. Block and Node should be used directly. The library aims to make no distinction between core and extension implementations, these classes add no useful information.

  • Deprecate: BaseSequence had old named functions which were misleading and duplicated under proper names:

    • countChars() -> countLeading()
    • countCharsNot() -> countLeadingNot()
    • countCharsReversed() -> countTrailing()
    • countNotCharsReversed() -> countTrailingNot()

    First of all they only counted leading characters which the name did not imply. Second they were duplicated.

    Add character counting functions:

    • countOfAny()
    • countOfAnyNot()

  • Fix: DefinitionExtension does not correctly set the child parse column, causing list items to be expecting 1 extra space for child item recognition.

  • Add: ExtensionConversion.NONE to suppress any output from corresponding element

  • Add: FlexmarkHtmlParser.EXT_MATH, default ExtensionConversion.HTML, for selecting <math> tag processing. For now only MARKDOWN does nothing useful. Later it will be used to convert math ml to GitLab math inline element.

0.40.16

0.40.14

  • Fix: AutolinkExtension removing leading Typographic nodes when the first link occurs in text following the typographic node.
  • Add: PDF converter sample with non-latin character set rendering information.
  • Fix: missing simple_smile emoji cheat sheet shortcut

0.40.12

  • Fix: #300, Typography extension breaks some auto links
  • Add: TypographicText interface to mark nodes which hold text which is replaced with typographic for rendering but treated as text for decoration processing. For now AutolinkExtension is the only one making use of it to prevent typographic smarts from breaking up a link and causing part of it to be left out of the URL.

0.40.10

0.40.8

0.40.6

  • Fix: MarkdownTable.appendTable(FormattingAppendable) to set FormattingAppendable.ALLOW_LEADING_WHITESPACE so indentation prefix is not eliminated.
  • Fix: FormattingAppendableImpl to not skip pending spaces if FormattingAppendable.ALLOW_LEADING_WHITESPACE is selected, these will be prefixed before indent.
  • Add: FlexmarkHtmlParser.SKIP_LINKS, default false. When true links are converted to text part of the link.
  • Add: DoNotAttributeDecorate interface to be implemented by all text nodes which do not result in span-like rendering of their text, such as TypographicSmarts, TypographicQuotes and SoftLineBreak but will erroneously create a break in the text and mess up attribute assignment to text spans.
  • Add: AttributesExtension.WRAP_NON_ATTRIBUTE_TEXT, default true. When false does not wrap nodes marked as DoNotAttributeDecorate in TextBase. When true, will allow intuitive attribute processing for text containing DoNotAttributeDecorate as single text span instead of one interrupted by the typographic smarts nodes.
  • Fix: AttributesExtension incorrectly parsed invalid empty sequences and absorbed the text instead of skipping it. ie. {}, { }, {#} and {.}
  • Add: AttributesExtension.USE_EMPTY_IMPLICIT_AS_SPAN_DELIMITER, default false. When set to true will treat {#} or {.}, without embedded spaced, as start attribute span delimiter to mark start of attribute assignment to text between {.} or {#} and the matching attributes element.

0.40.4

  • Fix: #295, CoreNodeFormatter does not descend into children on Link nodes
    • Link, LinkRef, Image and ImageRef node formatter renderer now descends into child nodes during all formatting, not just translation formatting.
  • Add: Formatter.OPTIMIZED_INLINE_RENDERING default false. When set to true will use previous rendering for links and images which appends the node characters without descending into child nodes.

0.40.2

0.40.0

  • Fix: Java9+ Compatibility, IntelliJ Migration contained in migrate 0_35_x to 0_40_0.xml, to use:

    • copy to IntelliJ application settings to migrations subdirectory
    • if you have the project which you want to migrate open, then close it
    • open the project in IntelliJ Ultimate or Community
    • update the flexmark-java dependency version to 0.40.0 (or later) and make sure the new library is downloaded/updated in the project.
    • use menu Refactor > Migrate...
    • select migrate flexmark-java 0.35.x to 0.40.0
    • press Run
    • in the refactoring preview tool window that opens hit Do Refactor

    ⚠️ In my projects, the IDE can miss some class migrations, especially on files which are not open. Re-running the migration tends to apply the migration. In cases where migration is not applied manual editing will be needed. In my projects Document, Node and Formatter were the problem classes:

    • com.vladsch.flexmark.ast.Document to com.vladsch.flexmark.util.ast.Document
    • com.vladsch.flexmark.ast.Node to com.vladsch.flexmark.util.ast.Node
    • com.vladsch.flexmark.formatter.internal.Formatter to com.vladsch.flexmark.formatter.Formatter

    ⚠️ flexmark-docx-converter and flexmark-pdf-converter cannot be used with Java 9+ modules because the underlying libraries used have package load conflicts that will be resolved in a later release.

  • Fix: overlapping packages in different modules, change:

    • com.vladsch.flexmark.ast.AllNodesVisitor to com.vladsch.flexmark.util.ast.AllNodesVisitor
    • com.vladsch.flexmark.ast.BlankLine to com.vladsch.flexmark.util.ast.BlankLine
    • com.vladsch.flexmark.ast.BlankLineContainer to com.vladsch.flexmark.util.ast.BlankLineContainer
    • com.vladsch.flexmark.ast.Block to com.vladsch.flexmark.util.ast.Block
    • com.vladsch.flexmark.ast.BlockContent to com.vladsch.flexmark.util.ast.BlockContent
    • com.vladsch.flexmark.ast.BlockNodeVisitor to com.vladsch.flexmark.util.ast.BlockNodeVisitor
    • com.vladsch.flexmark.ast.Content to com.vladsch.flexmark.util.ast.Content
    • com.vladsch.flexmark.ast.ContentNode to com.vladsch.flexmark.util.ast.ContentNode
    • com.vladsch.flexmark.ast.CustomBlock to com.vladsch.flexmark.util.ast.CustomBlock
    • com.vladsch.flexmark.ast.CustomNode to com.vladsch.flexmark.util.ast.CustomNode
    • com.vladsch.flexmark.ast.DescendantNodeIterable to com.vladsch.flexmark.util.ast.DescendantNodeIterable
    • com.vladsch.flexmark.ast.DescendantNodeIterator to com.vladsch.flexmark.util.ast.DescendantNodeIterator
    • com.vladsch.flexmark.ast.Node to com.vladsch.flexmark.util.ast.Node
    • com.vladsch.flexmark.ast.Document to com.vladsch.flexmark.util.ast.Document
    • com.vladsch.flexmark.ast.DoNotCollectText to com.vladsch.flexmark.util.ast.DoNotCollectText
    • com.vladsch.flexmark.ast.DoNotDecorate to com.vladsch.flexmark.util.ast.DoNotDecorate
    • com.vladsch.flexmark.ast.DoNotLinkDecorate to com.vladsch.flexmark.util.ast.DoNotLinkDecorate
    • com.vladsch.flexmark.ast.KeepTrailingBlankLineContainer to com.vladsch.flexmark.util.ast.KeepTrailingBlankLineContainer
    • com.vladsch.flexmark.ast.NodeAdaptedVisitor to com.vladsch.flexmark.util.ast.NodeAdaptedVisitor
    • com.vladsch.flexmark.ast.NodeAdaptingVisitHandler to com.vladsch.flexmark.util.ast.NodeAdaptingVisitHandler
    • com.vladsch.flexmark.ast.NodeAdaptingVisitor to com.vladsch.flexmark.util.ast.NodeAdaptingVisitor
    • com.vladsch.flexmark.ast.NodeIterable to com.vladsch.flexmark.util.ast.NodeIterable
    • com.vladsch.flexmark.ast.NodeIterator to com.vladsch.flexmark.util.ast.NodeIterator
    • com.vladsch.flexmark.ast.NodeRepository to com.vladsch.flexmark.util.ast.NodeRepository
    • com.vladsch.flexmark.ast.NodeVisitor to com.vladsch.flexmark.util.ast.NodeVisitor
    • com.vladsch.flexmark.ast.NodeVisitorBase to com.vladsch.flexmark.util.ast.NodeVisitorBase
    • com.vladsch.flexmark.ast.NonRenderingInline to com.vladsch.flexmark.util.ast.NonRenderingInline
    • com.vladsch.flexmark.ast.Package to com.vladsch.flexmark.util.ast.Package
    • com.vladsch.flexmark.ast.ReferenceNode to com.vladsch.flexmark.util.ast.ReferenceNode
    • com.vladsch.flexmark.ast.ReferencingNode to com.vladsch.flexmark.util.ast.ReferencingNode
    • com.vladsch.flexmark.ast.VisitHandler to com.vladsch.flexmark.util.ast.VisitHandler
    • com.vladsch.flexmark.ast.Visitor to com.vladsch.flexmark.util.ast.Visitor
    • com.vladsch.flexmark.IParse to com.vladsch.flexmark.util.IParse
    • com.vladsch.flexmark.IRender to com.vladsch.flexmark.util.IRender

  • Fix: API classes under internal package:

    • flexmark:

      • com.vladsch.flexmark.internal.inline.AsteriskDelimiterProcessor to com.vladsch.flexmark.parser.core.delimiter.AsteriskDelimiterProcessor

      • com.vladsch.flexmark.internal.inline.EmphasisDelimiterProcessor to com.vladsch.flexmark.parser.core.delimiter.EmphasisDelimiterProcessor

      • com.vladsch.flexmark.internal.Bracket to com.vladsch.flexmark.parser.core.delimiter.Bracket

      • com.vladsch.flexmark.internal.Delimiter to com.vladsch.flexmark.parser.core.delimiter.Delimiter

      • com.vladsch.flexmark.internal.ThematicBreakOptions to com.vladsch.flexmark.internal.ThematicBreakParser.ThematicBreakOptions

      • com.vladsch.flexmark.internal.HeadingOptions to com.vladsch.flexmark.internal.HeadingParser.HeadingOptions

      • com.vladsch.flexmark.internal.BlockParserTracker to com.vladsch.flexmark.parser.block.BlockParserTracker

      • com.vladsch.flexmark.internal.BlockQuoteParser to com.vladsch.flexmark.parser.core.BlockQuoteParser

      • com.vladsch.flexmark.internal.DocumentBlockParser to com.vladsch.flexmark.parser.core.DocumentBlockParser

      • com.vladsch.flexmark.internal.FencedCodeBlockParser to com.vladsch.flexmark.parser.core.FencedCodeBlockParser

      • com.vladsch.flexmark.internal.HeadingParser to com.vladsch.flexmark.parser.core.HeadingParser

      • com.vladsch.flexmark.internal.IndentedCodeBlockParser to com.vladsch.flexmark.parser.core.IndentedCodeBlockParser

      • com.vladsch.flexmark.internal.ListBlockParser to com.vladsch.flexmark.parser.core.ListBlockParser

      • com.vladsch.flexmark.internal.ListItemParser to com.vladsch.flexmark.parser.core.ListItemParser

      • com.vladsch.flexmark.internal.ParagraphParser to com.vladsch.flexmark.parser.core.ParagraphParser

      • com.vladsch.flexmark.internal.ThematicBreakParser to com.vladsch.flexmark.parser.core.ThematicBreakParser

      • com.vladsch.flexmark.internal.HtmlBlockParser to com.vladsch.flexmark.parser.core.HtmlBlockParser

      • com.vladsch.flexmark.internal.ReferencePreProcessorFactory to com.vladsch.flexmark.parser.core.ReferencePreProcessorFactory

      • com.vladsch.flexmark.internal.InlineParserImpl to com.vladsch.flexmark.parser.internal.InlineParserImpl

      • com.vladsch.flexmark.internal.BlockStartImpl to com.vladsch.flexmark.parser.internal.BlockStartImpl

      • com.vladsch.flexmark.internal.BlockContinueImpl to com.vladsch.flexmark.parser.internal.BlockContinueImpl

      • com.vladsch.flexmark.internal.LinkRefProcessorData to com.vladsch.flexmark.parser.internal.LinkRefProcessorData

      • com.vladsch.flexmark.internal.CommonmarkInlineParser to com.vladsch.flexmark.parser.internal.CommonmarkInlineParser

      • com.vladsch.flexmark.internal.MatchedBlockParserImpl to com.vladsch.flexmark.parser.internal.MatchedBlockParserImpl

      • com.vladsch.flexmark.internal.PostProcessorManager to com.vladsch.flexmark.parser.internal.PostProcessorManager

      • com.vladsch.flexmark.internal.DocumentParser to com.vladsch.flexmark.parser.internal.DocumentParser

      • com.vladsch.flexmark.internal.HtmlDeepParser to com.vladsch.flexmark.parser.internal.HtmlDeepParser

      • resource /com/vladsch/flexmark/internal/util/entities.properties to /com/vladsch/flexmark/util/html/entities.properties in flexmark-util module

    • flexmark-docx-converter

      • com.vladsch.flexmark.docx.converter.internal.DocxRenderer to com.vladsch.flexmark.docx.converter.DocxRenderer
      • com.vladsch.flexmark.docx.converter.internal.DocxRendererPhase to com.vladsch.flexmark.docx.converter.DocxRendererPhase
      • com.vladsch.flexmark.docx.converter.internal.NodeDocxRendererHandler to com.vladsch.flexmark.docx.converter.NodeDocxRendererHandler
      • com.vladsch.flexmark.docx.converter.internal.DocxRendererOptions to com.vladsch.flexmark.docx.converter.DocxRendererOptions

    • flexmark-formatter

      • com.vladsch.flexmark.formatter.internal.Formatter to com.vladsch.flexmark.formatter.Formatter
      • com.vladsch.flexmark.formatter.internal.FormattingPhase to com.vladsch.flexmark.formatter.FormattingPhase
      • com.vladsch.flexmark.formatter.internal.MarkdownWriter to com.vladsch.flexmark.formatter.MarkdownWriter
      • com.vladsch.flexmark.formatter.internal.NodeFormatter to com.vladsch.flexmark.formatter.NodeFormatter
      • com.vladsch.flexmark.formatter.internal.NodeFormatterContext to com.vladsch.flexmark.formatter.NodeFormatterContext
      • com.vladsch.flexmark.formatter.internal.NodeFormatterFactory to com.vladsch.flexmark.formatter.NodeFormatterFactory
      • com.vladsch.flexmark.formatter.internal.NodeFormattingHandler to com.vladsch.flexmark.formatter.NodeFormattingHandler
      • com.vladsch.flexmark.formatter.internal.NodeRepositoryFormatter to com.vladsch.flexmark.formatter.NodeRepositoryFormatter
      • com.vladsch.flexmark.formatter.internal.PhasedNodeFormatter to com.vladsch.flexmark.formatter.PhasedNodeFormatter
      • com.vladsch.flexmark.formatter.internal.NodeFormatterSubContext to com.vladsch.flexmark.formatter.NodeFormatterSubContext

0.35.8

  • Fix: missing test scope on flexmark-test-utils

0.35.6

  • Fix: MarkdownTable.fillMissingColumns() would add empty columns not setting the proper source offset, causing later MarkdownTable.addTrackedOffset(int) to fail
  • Fix: table extension parsing of table cells with inline elements containing table separator characters would parse correctly but not convert inlined table separators back to text, so these would not render.
  • Add: FlexmarkHtmlParser.SKIP_FENCED_CODE, default false. When enabled will convert fenced code to indented code in generated markdown.
  • Add: FlexmarkHtmlParser.SKIP_CHAR_ESCAPE, default false. When enabled will not escape special characters.

0.35.4

  • Add: html/ui helpers flexmark-util for use with Swing UI HTML generation

0.35.2

  • Fix: base64 encoding function in ImageUtils to remove \n from encoded string.

  • Fix: Aside extension to use new KeepTrailingBlankLineContainer marker for blocks which have a prefix which allows attribution of blank lines to the block.

  • Add: sample for converting inline nodes to text nodes in node post processor. Thanks to @markkolich.

  • Fix: ReplacedTextMapper can only handle a single replacement set, while AutoLinkExtension was applying a replacement on top of already replaced text.

    ⚠️ Now any functions which perform replacements using ReplacedTextMapper need to check if the passed in text mapper isModified() and if this is the case invoke startNestedReplacement(BasedSequence).

  • Add: log4j logging to DocumentParser, when debug is enabled parser will log resulting AST after parsing, paragraph pre-processing, block pre-processing and inline processing to isolate where a problem is introduced by an extension into the AST.

  • Breaking: ⚠️ Potentially breaking change for some code if parsing with Parser.BLANK_LINES_IN_AST enabled. Last blank line of blocks is now moved out to the parent block. Greatest effect is on list items which previously held on to their last blank line, except for the last item in the list. Now all trailing blank lines are moved out to the parent list. Therefore children of lists can be list items or blank lines, not just list items. This makes blank line attribution more consistent.

  • Fix: #287, ''flexmark-html-parser' The module has an mistake

  • Fix: empty table cells now contain a space so that the position of the cell's text in the file is not lost.

  • Fix: Formatter inline elements leaving embedded EOL sequences when Formatter.KEEP_SOFT_LINE_BREAKS is false.

  • Fix: trailing blank lines in block quotes are now left inside the block quote instead of moving them up to the document level if these blank lines are unambiguously attributable to the block quote by having a block quote prefix on the line. Unlike other block elements where a trailing blank line is not attributable to the block element, block quotes have explicit prefix which identifies the blank line as part of the block quote.

    ⚠️ trailing blank line nodes will be appended to the block quote even if Parser.BLANK_LINES_IN_AST is false to force block quote to span the blank lines as the only way to extend the block quote enclosed characters.

0.35.0

  • Test: navigation
    • Next/Prev Tab
  • Add: For next/prev tab table navigation in:
    • separator stop at start and end of cell for editing :
    • header/body stop at end of text to allow continued typing
    • caption navigate to end of text like a cell
  • Test: addTrackOffset() for caption before/inside/after location.
    • Add: isBeforeCaption(), isInsideCaption(), isAfterCaption()
  • Fix: change offset tracking to provide nullable Character and isDeleted. true - deleted, false - inserted/typed. Character null - unknown, else character typed, inserted or deleted. Then all internal changes to what is needed will be isolated inside the MarkdownTable class.
  • Add: table navigation helpers
  • Fix: TextCollectingVisitor to handle not just inline nodes but paragraphs and optionally other blocks. It now adds spaces between non-adjacent text nodes and adds a blank line between paragraphs.
  • Add: TableTextCollectingVisitor which will extract the table text (without separator) and add a blank line after the table.
  • Add: inline parser will no longer process block nodes marked as DoNotDecorate with inline parser extensions to prevent undesired inline processing of nodes such as a table separator by typographic extension.
  • Add: table manipulation functions to MarkdownTable to allow easy modification of a markdown table which can then be output as formatted markdown text. This class implements table formatting for the tables extension.
  • Add: to table formatter options:
    • FORMAT_TABLE_CAPTION_SPACES , default DiscretionaryText.AS_IS, how to handle spaces after [ and before ] in formatted output |
    • FORMAT_TABLE_INDENT_PREFIX , default "", adds arbitrary prefix to tables in formatted output |
    • FORMAT_TABLE_MANIPULATOR , default TableManipulator.NULL, interface invoked before table is appended to formatted output. Allows table manipulation |
    • FORMAT_CHAR_WIDTH_PROVIDER , default CharWidthProvider.NULL, interface used to provide actual characters widths so table formatting can take these into account |
  • Break: potentially breaking change, now FencedCode block with empty content returns SubSequence.NULL for content with no lines, previously returned empty string with start==end representing the position in the document where the content should have been. Affects only FencedCodeBlock and AsideBlock. Can useFencedCodeBlock.getOpeningMarker().getEndOffset() \+ 1 if you need the position where the content would have been. Similarly for AsideBlock
  • Fix: SegmentedSequence start and end offsets to always be correct when contained segments are all non-based characters and computed on creation.
  • Break: some breaking changes to FormattingAppendableImpl derived constructors. Now most arguments must be passed, no defaults. It was getting too messy to trace which constructors were used and what were the resulting options for the appendable.
  • Fix: with HtmlRenderer.FORMAT_FLAGS set to HtmlRenderer.PASS_THROUGH pre-formatted would incorrectly add indent spaces.
  • Fix: optimized HTML rendering with HtmlRenderer.FORMAT_FLAGS set to HtmlRenderer.PASS_THROUGH, no fancy output formatting other than line breaks but 7x faster than using standard fancy formatting output.
  • Fix: Emoji shortcut did not render emoji with UNICODE_ONLY image type option
  • Add: Formatter.KEEP_HARD_LINE_BREAKS, default true and Formatter.KEEP_SOFT_LINE_BREAKS default true. Setting to false will not output corresponding line break.
  • Add: Formatter.APPEND_TRANSFERRED_REFERENCES, default false. Setting to true will append all transferred references to the bottom of the formatted document.
  • Add: NodeRepository API to return referenced references from a given node.
  • Fix: Jira and YouTrack converters ignored HtmlRenderer.RECHECK_UNDEFINED_REFERENCES
  • Add: Base64 conversion to image utils

0.34.58

  • Fix: possible index out of bounds on segmented sequences
  • Add: HTML parser option to modify id attribute before output. Used to strip prefix user-content- from GitHub rendered docs. See OUTPUT_ID_ATTRIBUTE_REGEX
  • Fix: HTML parser to transfer last <a> with id/name attribute to parent.
  • Fix: HTML parser to strip out <a> tags with id attribute and no text when direct children of <body>. A kludge to strip out duplicate ids in GitHub rendered docs.
  • Fix: HTML parser to respect <ol> start attribute for first list item.
  • Fix: HTML parser to add *, ~ and ^ to escaped text characters so they don't generate erroneous markup.
  • Add: to TOC extension CASE_SENSITIVE_TOC_TAG option, default true, setting it to false allows [TOC] and [TOC]: # to be matched without case sensitivity. The syntax for toc options is still case sensitive.

0.34.56

0.34.53

  • Change: revert docx4j to 3.3.4 in docx4j-3 branch

0.34.52

  • Fix: intermittent NPE when including macro with table in table

0.34.51

  • Fix: intermittent NPE when including macro with table in table in docx4j-3 branch

0.34.50

  • Add: HtmlRenderer.NO_P_TAGS_USE_BR, default false. When enabled instead of wrapping loose paragraphs in <p> tag adds 2 x <br /> after paragraph.
  • Fix: extension auto links would not be processed if paragraph had inline parser processed auto links before.

0.34.49

  • Change: revert docx4j to 3.3.6 in docx4j-3 branch

0.34.48

  • Add: Macros extension handling to docx converter. Macros allow nesting of block elements inside tables.

0.34.46

  • Add: flexmark-ext-macros extension to allow macro definition and inclusion in inline contexts. Extensions: Macros
  • Fix: Link.setTextChars(BasedSequence) was trimming the string before setting Link.text. Now the full untrimmed text is used.

0.34.44

0.34.42

  • Add: url parsing to AutoLink to set other link parts of the node like: pageref, anchor marker and anchor

0.34.40

  • Fix: add HtmlRenderer.HEADER_ID_GENERATOR_NON_ASCII_TO_LOWERCASE, default true. When set to false changes the default header id generator to not convert non-ascii alphabetic characters to lowercase. Needed for GitHub id compatibility.
  • Fix: add HtmlRenderer.HEADER_ID_GENERATOR_NON_ASCII_TO_LOWERCASE set to false to GitHub and GitHubDoc parser profiles.

0.34.38

  • Fix: docx conversion of hyperlinks did not take relationship part into account when optimizing rel references causing invalid rel ids to be used if the same link was included in multiple contexts.

0.34.36

  • Fix: docx conversion of footnote refs in tables to footnotes which contained links generated invalid docx.

0.34.34

  • Change: docx4j library version to 6.0.1

0.34.32

  • Fix: change anonymous classes to static classes in CoreNodeDocxRenderer
  • Add: test case for Parser.SPACE_IN_LINK_URLS to validate allowing spaces in URLs
  • Fix: Docx Converter to use style names instead of style ids because Word localizes ids and leaves the names common to all locales. Completely opposite to intuition but that is Word.
  • Fix: Docx Converter remove hardcoded "CENTER" alignment for table headings overriding the Table Heading style provided alignment.

0.34.30

  • Fix: #268, Pipe characters are not escaped in Table (FlexmarkHtmlParser)
    • Fix: escape pipe characters in text (to avoid accidental use as table or other markup) when not inline code nor fenced code
    • Fix: escape back ticks when inside code
    • Fix: disable escaping of [] when inside code
    • Fix: disable escaping of \ when inside code
    • Fix: replace non-break space with space when inside code
  • Fix: FlexmarkHtmlParser.BR_AS_EXTRA_BLANK_LINES now adds <br /> followed by blank line
  • Add: GitLab extension handling to docx converter. For now math and mermaid blocks are not converted.
  • Add: GitLab extension formatter to properly handle block quote formatting
  • Add: ParagraphContainer interface to allow container nodes to control how their paragraphs are output for formatting/rendering with respect to blank line before/after control.
  • Fix: GitLab block quote parser bug for terminating sequence.

0.34.28

  • Add: html parser options to specify how to generate some markdown formatting elements, default for all is ExtensionConversion.MARKDOWN:

    • EXT_INLINE_STRONG - strong
    • EXT_INLINE_EMPHASIS - emphasis
    • EXT_INLINE_CODE - code
    • EXT_INLINE_DEL - del
    • EXT_INLINE_INS - ins
    • EXT_INLINE_SUB - sub
    • EXT_INLINE_SUP - sup

    Available settings:

    • ExtensionConversion.MARKDOWN - convert to markdown
    • ExtensionConversion.TEXT - convert to inner text
    • ExtensionConversion.HTML - leave HTML as is

    Corresponding SKIP_ options have been deprecated since their function is duplicated by new options.

0.34.26

  • Fix: Add DoNotLinkDecorate to Code for preventing AutoLink extension from decorating inline code with autolinks.
  • Fix: BasedSequenceImpl.lastIndexOfAny(CharSequence) and BasedSequenceImpl.lastIndexOfAnyNot(CharSequence) were using indexOf... instead of lastIndexOf...

0.34.24

  • Fix: BasedSequenceImpl.countCharsReversed(char, int, int) and BasedSequenceImpl.countNotCharsReversed(char, int, int) to not return -1, should return 0
  • Fix: BasedSequenceImpl.spliceAtEnd() not to fail if this or other is empty
  • Add: GitLabExtension, Documented in Gitlab Flavoured Markdown Extensions

0.34.22

  • Add: OSGi module support thanks to @klcodanr

0.34.20

  • Add: renderer type equivalences for allowing new renderer types which override existing ones. Now "YOUTRACK" is defined as an override of "JIRA" by YouTrackConverterExtension
    • use HtmlRenderer.Builder.isRendererType(String) to test for recognizing renderer type instead of rendererType.equals("...") make sure renderer type checking is done with most specific one, otherwise the wrong match will occur. For example, test for "YOUTRACK" before "JIRA" because "JIRA" matches "YOUTRACK".
    • HtmlRenderer.isCompatibleRendererType(MutableDataHolder, String)
    • HtmlRenderer.isCompatibleRendererType(MutableDataHolder, String, String)
    • HtmlRenderer.addRenderTypeEquivalence(MutableDataHolder, String, String) add new to old renderer type mapping.

0.34.18

  • Add: html parser options to suppress generating some markdown formatting elements, default for all is false:
    • SKIP_INLINE_STRONG - strong
    • SKIP_INLINE_EMPHASIS - emphasis
    • SKIP_INLINE_CODE - code
    • SKIP_INLINE_DEL - del
    • SKIP_INLINE_INS - ins
    • SKIP_INLINE_SUB - sub
    • SKIP_INLINE_SUP - sup
    • SKIP_HEADING_1 - heading 1
    • SKIP_HEADING_2 - heading 2
    • SKIP_HEADING_3 - heading 3
    • SKIP_HEADING_4 - heading 4
    • SKIP_HEADING_5 - heading 5
    • SKIP_HEADING_6 - heading 6
    • SKIP_ATTRIBUTES - attribute extension formatting
  • Add: html parser option ADD_TRAILING_EOL, default false. Will add trailing EOL to generated markdown text.
  • Fix: html parser did not add a blank line before first list item of the first list

0.34.16

0.34.14

0.34.12

0.34.10

  • Fix: #247, Admonition Expression may lack a part of the text.
  • Add: TocExtension.TOC_CONTENT, TocExtension.TOC_LIST, SimTocExtension.TOC_CONTENT, SimTocExtension.TOC_LIST attributable parts for div and ul/ol HTML tags. Documented in Table-of-Contents-Extension
    • Fix: TOC element to use TITLE option if one is defined. Default for TOC is "", for sim TOC it is "Table of Contents"
  • Fix: #248, Request to add CSS class of TOC
    • Add: TocExtension.DIV_CLASS default "", class attribute to use on table of content div wrapper, duplicated as SimTocExtension.DIV_CLASS
    • Add: TocExtension.LIST_CLASS default "", class attribute to use on table of content ul or ol element, duplicated as SimTocExtension.LIST_CLASS

0.34.8

0.34.6

0.34.4

0.34.2

  • Fix: update nexus staging plugin to 1.6.7
  • Fix: move nexus-staging plugin to deploy profile so it is not needed for ci test build
  • Fix: gfm tables extension to only require a single separator dash
  • Fix: for matched parenthesis destination link parsing, only treat backslash as escape if followed by characters which can be escaped.

0.34.0

  • Fix: table formatter for translations to insert a space if the translated text is empty.
  • Add: changed regex for parsing unquoted HTML attribute value to exclude {}.
  • Add: translation formatting awareness to: abbreviation, admonition, attributes, emoji, enumerated reference footnotes and tables extensions
  • Add: translation API sample: TranslationSample.java
  • Fix: AttributesExtension incorrectly parsed consecutive attributes as a single element
  • Add: TranslationHandlerFactory to allow format options to be passed for custom handler creation.
  • Add: RenderPurpose arg to Formatter.translationRender eliminating the need for separate render purpose to be set in translation handler instance.
  • Add: Translation helper API to Formatter module. A bit of a kludge but has the potential to work. Preserves non-translatable entities, uses a single translation for all occurrences of text snippets such as ref text. Leaves inline emphasis in, Yandex has no issues with it. Replaces all element text which needs to be preserved with # where is unique integer. Will replace anchor refs to headings in file if the heading text is translated.
  • Add: to html parser recognition that <img alt="emoji category:shortcut" src=""> is an emoji shortcut generated by flexmark-java emoji extension and convert it to such.

0.32.24

  • Fix: #232, List continuation, add Formatter.LIST_REMOVE_EMPTY_ITEMS, default false. If true then empty list items or list items which only contain a BlankLine node are removed from the output.
  • Fix: formatter output of empty list items
  • Fix: formatter output of empty loose task list items

0.32.22

  • Fix: #229, gfm-tasklist extension - add state class to list item

    • Add: TaskListExtension.ITEM_DONE_CLASS and TaskListExtension.ITEM_NOT_DONE_CLASS li tag classes to use for done and not done items.
    • Add: TaskListExtension.TIGHT_ITEM_CLASS to complement the TaskListExtension.LOOSE_ITEM_CLASS to replace generically named TaskListExtension.ITEM_CLASS
    • Add: depreciation of TaskListExtension.ITEM_CLASS in favour of TaskListExtension.TIGHT_ITEM_CLASS

  • Fix: merge #231, PR: Fix two small bugs in ext-toc @BlueBoxWare

  • Add: TaskListAttributeProviderSample to flexmark-java-samples module.

  • Add: GitHubParsingSample to flexmark-java-samples module to show GitHub compatible parser setup.

  • Add: ParserEmulationProfile.GITHUB to reflect current GitHub profile, effectively it is ParserEmulationProfile.COMMONMARK_0_28 with GitHub compatible id generator settings.

  • Fix: #221, XSS: Javascript execution through links, add HtmlRenderer.SUPPRESSED_LINKS default "javascript:.*", a regular expression to suppress any links that match. The test occurs before the link is resolved using a link resolver. Therefore any link matching will be suppressed before it is resolved. Likewise, a link resolver returning a suppressed link will not be suppressed since this is up to the custom link resolver to handle.

    Suppressed links will render only the child nodes, effectively [Something New](javascript:alert(1)) will render as if it was Something New.

    Link suppression based on URL prefixes does not apply to HTML inline or block elements. Use HTML suppression options for this.

  • Add: AutolinkExtension.IGNORE_LINKS default "", a regex expression to match link text which should not be auto-linked. This can include full urls like www.google.com or parts by including wildcard match patterns. Any recognized auto-link which matches the expression will be rendered as text.

0.32.20

0.32.18

0.32.16

  • Fix: change all javadoc/overview.html to be generated from javadoc/overview.md files.
  • Fix: admonition.css remove non-existent horiz-align property.
  • Add: HtmlToMarkdownSample.java
  • Add: HtmlRenderer.HEADER_ID_GENERATOR_NON_DASH_CHARS, default "". Any characters in the string will be preserved and passed through to the heading ID.
  • Fix: ParserEmulationProfile.GITHUB_DOC now sets HtmlRenderer.HEADER_ID_GENERATOR_NO_DUPED_DASHES to " -" and HtmlRenderer.HEADER_ID_GENERATOR_NON_DASH_CHARS to "_" to match current GitHub heading id generation rules.

0.32.14

  • Change: AdmonitionExtension.WITH_LEAD_SPACES_INTERRUPTS_ITEM_PARAGRAPH to AdmonitionExtension.WITH_SPACES_INTERRUPTS_ITEM_PARAGRAPH
  • Fix: Formatter for Tables now treats header column alignment as centered if no alignment marker is given for the column, body cells remain left aligned. This is default browser rendered table appearance.
  • Fix: #213, Attributes element does not allow spaces before closing }
  • Fix: update to open html to pdf 0.0.1-RC12

0.32.12

  • Add: DocxRenderer.TABLE_PREFERRED_WIDTH_PCT default 0, if 0 then table width is auto, otherwise specifies percent of page width.
  • Add: DocxRenderer.TABLE_LEFT_INDENT default 120, in twips
  • Add: DocxRenderer.TABLE_STYLE default "", if not empty then table then table style will be set to this value and default cell borders will not be generated.

0.32.10

  • Add: AdmonitionBlock getters/setters for titleOpeningMarker, title, titleClosingMarker
  • Fix: AdmonitionNodeRenderer remove erroneous, extra </div>
  • Add: AdmonitionNodeRenderer attributable parts:
    • ADMONITION_SVG_OBJECT_PART - svg object at top of body, part is on Document node
    • ADMONITION_HEADING_PART - heading div
    • ADMONITION_ICON_PART - heading icon svg
    • ADMONITION_TITLE_PART - heading title span
    • ADMONITION_BODY_PART - admonition body div

0.32.8

0.32.6

  • Add: DocxRenderer.PREFIX_WWW_LINKS, default true, when true adds http:// to link addresses which start with www.

  • Fix: DocxRenderer some images would not be embedded and generate a CRC Error when loaded via URL vs file.

0.32.4

0.32.2

  • Add: EmojiExtension.ATTR_IMAGE_CLASS, default "", if not empty will set the image tag class attribute to this value.

  • Fix: when image type was set to unicode then unicode char would be used for a shortcut even when that shortcut was not defined for requested shortcut type: github or ecs.

  • Fix: update to docx4j version 3.3.6

0.32.0

  • API Change: removed EmojiCheatSheet class to replace with EmojiShortcuts which has better referencing for GitHub shortcuts and unicode chars for all emojis from emoji-cross-reference

    • Removed: EmojiExtension.USE_IMAGE_URL
    • Add: EmojiExtension.USE_SHORTCUT_TYPE, default EmojiShortcutType.EMOJI_CHEAT_SHEET, select type of shortcuts:
      • EmojiShortcutType.EMOJI_CHEAT_SHEET
      • EmojiShortcutType.GITHUB
      • EmojiShortcutType.ANY_EMOJI_CHEAT_SHEET_PREFERRED use any shortcut from any source. If image type options is not UNICODE_ONLY, will generate links to Emoji Cheat Sheet files or GitHub URL, with preference given to Emoji Cheat Sheet files.
      • EmojiShortcutType.ANY_GITHUB_PREFERRED - use any shortcut from any source. If image type options is not UNICODE_ONLY, will generate links to Emoji Cheat Sheet files or GitHub URL, with preference given to GitHub URL.
    • Add: EmojiExtension.USE_IMAGE_TYPE, default EmojiImageType.IMAGE_ONLY, to select what type of images are allowed.
      • EmojiImageType.IMAGE_ONLY, only use image link
      • EmojiImageType.UNICODE_ONLY convert to unicode and if there is no unicode treat as invalid emoji shortcut
      • EmojiImageType.UNICODE_FALLBACK_TO_IMAGE convert to unicode and if no unicode use image.

  • API Change: removed HtmlRenderer.WRAP_TIGHT_ITEM_PARAGRAPH_IN_SPAN, it was only needed to have somewhere to set attributes for tight item text for Attributes which now implement intelligent span wrapping of text.

  • Change: AttributesExtension.ASSIGN_TEXT_ATTRIBUTES, default true. Set to false for backward compatibility.

    When true attribute assignment rules to nodes are changed to allow text elements to get attributes. If an attribute element is spaced from the previous plain text element some text {attributes} then attributes are for the parent of the text. If they are attached to the text element some text{attributes} then they are assigned to the immediately preceding plain text span. Within an inline element same rules apply: **bold text {attr}** are for the strong emphasis span, while **bold text{attr}** will wrap the text between the strong emphasis tags, in a span with given attributes.

    If a plain text span is interrupted by an HTML comment then it is considered as two separate plain text spans. ie. some <!---->text{attr} will result in some <!----><span attr>text</span> rendering.

    Full spec: ext_attributes_ast_spec

  • Fix: Attributes, when they are first child then delete leading spaces of following node first rendered child will have leading spaces.

  • Add: Attribute syntax option allows Attributes on reference definition so that all elements referencing it, inherit the attributes.

    To assign attributes to a reference definition you need to leave a blank line between the reference definition and the attributes element, placed by itself and followed by a blank line.

    ![reference 1][ref]

![reference 2][ref]

[ref]: http://example.com/image.png

{width=20 height=20}

  • Fix: HTML Parser:

    • image alt text should escape markdown special characters which can affect parsing of image link text like: []
    • convert <pre><tt> to indented code
    • add attribute extraction
      • id : if FlexmarkHtmlParser.OUTPUT_ATTRIBUTES_ID is true, default true
      • FlexmarkHtmlParser.OUTPUT_ATTRIBUTES_NAMES_REGEX default "", if not empty then will output attributes extension syntax for attribute names that are matched by the regex.
    • did not output a line end at start of <div>, this tended to splice text in div to previous element.

  • Add: Emoji Extension support to DocxRenderer.

    • Add: Emoji Cheat Sheet images to resources in DocxRender jar, default configuration will resolve emoji image files to the files in the jar.
    • Add: DocxRenderer.DOC_EMOJI_IMAGE_VERT_OFFSET, default -0.1, vertical offset of emoji image as a factor of line height at point of insertion. The final value is rounded to nearest pt so jumps of 1 pt for small changes of this value can occur.
    • Add: DocxRenderer.DOC_EMOJI_IMAGE_VERT_SIZE, default 1.05, size of emoji image as a factor of line height at point of insertion.

0.30.0

  • Fix: #198, StringIndexOutOfBoundsException in AbbreviationExtension if abbreviation definition had an empty abbreviation.

  • API Change: Refactoring of Interfaces to allow extensions only providing link resolver, attribute provider and html id generator to be re-used by the DocxRenderer and HtmlRenderer without modifications other than changing the implemented extension from HtmlRenderer.HtmlRendererExtension to RendererExtension

    • AttributesProviderFactory pass only LinkResolverContext instead of NodeRenderingContext to allow for attribute provider extensions to be re-used with DocxRender

    • HeadIdGenerator pass only LinkResolverContext instead of NodeRenderingContext to allow for header id generator provider extensions to be re-used with DocxRender

    • new RendererExtension with only ability to register html id generator, link resolver and attribute provider. Such an extension can be used as is with HtmlRenderer and DocxRenderer

      RendererExtension.extend(RendererBuilder, String) method of these gets passed RendererBuilder instead of HtmlRenderer.Builder

      extensions that implement both RendererExtension and HtmlRendererExtension will have only have the html renderer extension extend method called.

  • Add: flexmark-ext-aside handling to DocxConverter

  • Fix: DocxConverter formatting around footnote reference would be applied to footnote text

  • Add: DocxRenderer.LOCAL_HYPERLINK_MISSING_HIGHLIGHT, default "red" to highlight in document hyperlinks with missing targets. Set to "" to disable this. NOTE: must be one of the 8 named colors used by Word or it might complain.

  • Add: DocxRenderer.LOCAL_HYPERLINK_MISSING_FORMAT, default "Missing target id: #%s" to change the tooltip text of the missing hyperlink. %s will be replaced with the reference id of the link.

  • Fix: DocxConverter self referencing ref anchors should be converted to bookmark references in the docx

  • Add: DocxConverter now supports AttributesExtension and EnumeratedReferenceExtension by converting id attributes to bookmarks and enumerated reference links to hyperlinks to bookmarks.

  • Change: Enumerated references which are missing the definition for their type now default to type # where # is the reference ordinal instead of just #.

  • Add: Enumerated links now have title set to the text value of the reference.

  • Add: DocxConverter options to control heading id generation to resolve anchor refs to document anchors

    • DocxRenderer.HTML_ID_GENERATOR, default HtmlIdGenerator instance, the id generator to use for generating heading ids.

    • DocxRenderer.LOCAL_HYPERLINK_SUFFIX, default "", used to add a suffix to hyperlink bookmark names, except ones linking to first heading in a document. Only needed if you have some post processor on the docx adding suffixes to bookmark names.

For convenience these HtmlRenderer keys are aliased through DocxRenderer, keep in mind that setting either will affect both keys. For information on these keys see HtmlRenderer options

  • DocxRenderer.HEADER_ID_GENERATOR_RESOLVE_DUPES, default true,

  • DocxRenderer.HEADER_ID_GENERATOR_TO_DASH_CHARS, default " -_"

  • DocxRenderer.HEADER_ID_GENERATOR_NO_DUPED_DASHES, default false

  • Add: DocxConverter options to re-map standard style names to user defined ones.

    • DocxRenderer.DEFAULT_STYLE, default "Normal"
    • DocxRenderer.LOOSE_PARAGRAPH_STYLE, default "ParagraphTextBody"
    • DocxRenderer.TIGHT_PARAGRAPH_STYLE, default "BodyText"
    • DocxRenderer.PREFORMATTED_TEXT_STYLE, default "PreformattedText"
    • DocxRenderer.BLOCK_QUOTE_STYLE, default "Quotations"
    • DocxRenderer.ASIDE_BLOCK_STYLE, default "AsideBlock"
    • DocxRenderer.HORIZONTAL_LINE_STYLE, default "HorizontalLine"
    • DocxRenderer.TABLE_CAPTION, default "TableCaption"
    • DocxRenderer.TABLE_CONTENTS, default "TableContents"
    • DocxRenderer.TABLE_HEADING, default "TableHeading"
    • DocxRenderer.FOOTNOTE_STYLE, default "Footnote"
  • DocxRenderer.BOLD_STYLE, default "StrongEmphasis"
  • DocxRenderer.ITALIC_STYLE, default "Emphasis"
  • DocxRenderer.STRIKE_THROUGH_STYLE, default "Strikethrough"
  • DocxRenderer.SUBSCRIPT_STYLE, default "Subscript"
  • DocxRenderer.SUPERSCRIPT_STYLE, default "Superscript"
  • DocxRenderer.INS_STYLE, default "Underlined"
  • DocxRenderer.INLINE_CODE_STYLE, default "SourceText"
  • DocxRenderer.HYPERLINK_STYLE, default "Hyperlink"