Package-level declarations

Recognizes a block quote.

Recognizes any non-blank line no other recognizer claimed. As the last recognizer of every block lexer, it guarantees the scan always makes progress.

Recognizes a comment, at a line start or in the middle of a line the scan resumed inside. A comment may span lines and absorbs nothing after its closing delimiter, so its text is not line-aligned and its match carries an explicit end index.

Recognizes content fenced by three or more backticks or tildes. The block ends at the first line that holds nothing but a run of the same character, at least as long as the opening one, and trailing whitespace. A longer opening fence therefore wraps shorter fences as content, and a block whose fence is never closed runs to the end of the document.

Recognizes a footnote definition. Its text follows the paragraph continuation rule, so the definition ends where a paragraph does.

Recognizes a heading opened by one to six #. The trailing # run and the whitespace before it are dropped, and the heading absorbs every blank line that follows it.

Recognizes content indented by at least four columns.

Recognizes a referenceable link definition. The URL may start on the line after the colon and the title on the line after the URL, so a definition spans up to three lines. The block absorbs the empty lines that follow it.

Recognizes one item of a list. Like the list itself, it keeps an explicit loop, because its stopping rule chooses between ending before the terminator, ending after it, and continuing.

The two kinds of list, which differ in the bullets that may open and continue them.

Recognizes a list, ordered or not.

Recognizes a run of blank lines. The token's text spans every blank line with its terminator, so consecutive blank lines collapse into a single token.

Recognizes plain text content: a non-empty line, continued by every following line that is neither empty nor an interruption.

Recognizes a heading underlined by = or -. Every line from the first up to the underline is the heading's content; none of them may start with a bullet or be an underline itself.

Recognizes a GFM table. The rows run from the third line up to the first blank line or interruption, and the trailing empty lines belong to the token. A caption or custom ID in the last row is the parser's business.

Recognizes a thematic break. It absorbs the empty lines that follow it, but not a whitespace-only line, which ends it.

The block recognizers of the base Markdown flavor, highest priority first.