) — rarely used, * regex cost outweighs the benefit. * - Balanced-parens inside URLs [link](foo(bar)) — uncommon, complex. * - Image-in-link [](url) — requires GfmMedia plus nested * recursion across modes. * - Title HTML attribute — DokuWiki link handler instructions have no * title-attribute slot, and plumbing one through every renderer just * for this is out of scope. The title parses cleanly but is discarded. */ class GfmLink extends AbstractMode { use LinkDispatch; /** @inheritdoc */ public function getSort() { return 300; } /** @inheritdoc */ public function connectTo($mode) { // Pattern breakdown: // \[(?!\[) — single `[`, not part of DW `[[` // [^\[\]\n]+ — text: no nested brackets, single line // \]\( — `]` immediately followed by `(` (GFM // forbids whitespace between them) // \s* — optional whitespace around the URL // [^\s()\n]+ — URL: no whitespace, no parens, single line // (?:\s+(?:"[^"\n]*" // |'[^'\n]*'))? — optional title in "..." or '...' // \s*\) — optional trailing whitespace, close paren $pattern = '\[(?!\[)[^\[\]\n]+\]\(' . '\s*[^\s()\n]+' . '(?:\s+(?:"[^"\n]*"|\'[^\'\n]*\'))?' . '\s*\)'; $this->Lexer->addSpecialPattern($pattern, $mode, 'gfm_link'); } /** @inheritdoc */ public function handle($match, $state, $pos, Handler $handler) { // The entry pattern has already validated the `[text](url)` // shape, so we can destructure with plain string ops. Split on // `](` to separate text from "url and optional title"; the URL // is the first whitespace-delimited token of the remainder, and // anything after it is the title — discarded, since DokuWiki // link instructions have no title-attribute slot. $sep = strpos($match, ']('); $text = substr($match, 1, $sep - 1); $inside = trim(substr($match, $sep + 2, -1)); $url = substr($inside, 0, strcspn($inside, " \t\n")); $this->dispatchLink($url, $text, $pos, $handler); return true; } }