diff --git a/src/markup/engine/__tests__/remarkup/ordered-list-with-numbers.txt b/src/markup/engine/__tests__/remarkup/ordered-list-with-numbers.txt index 5e79679..c00cd66 100644 --- a/src/markup/engine/__tests__/remarkup/ordered-list-with-numbers.txt +++ b/src/markup/engine/__tests__/remarkup/ordered-list-with-numbers.txt @@ -1,42 +1,48 @@ # aasdx # asdf 1. asa # asdf -234/ asdf +234) asdf 234) asd 1. asd 234) asd + +1/ This explicitly should not be formatted as a list. ~~~~~~~~~~
234) asd
1/ This explicitly should not be formatted as a list.
~~~~~~~~~~ 1. aasdx 2. asdf 1. asa 1. asdf 2. asdf 234) asd 1. asd 2. asd + +1/ This explicitly should not be formatted as a list. diff --git a/src/markup/engine/remarkup/blockrule/PhutilRemarkupEngineRemarkupListBlockRule.php b/src/markup/engine/remarkup/blockrule/PhutilRemarkupEngineRemarkupListBlockRule.php index 33403b7..b3481d6 100644 --- a/src/markup/engine/remarkup/blockrule/PhutilRemarkupEngineRemarkupListBlockRule.php +++ b/src/markup/engine/remarkup/blockrule/PhutilRemarkupEngineRemarkupListBlockRule.php @@ -1,433 +1,433 @@ $line) { $matches = null; if (preg_match($regex, $line)) { $regex = self::CONT_BLOCK_PATTERN; if (preg_match('/^(\s+)/', $line, $matches)) { $space = strlen($matches[1]); } else { $space = 0; } $min_space = min($min_space, $space); } } $regex = self::START_BLOCK_PATTERN; if ($min_space) { foreach ($lines as $key => $line) { if (preg_match($regex, $line)) { $regex = self::CONT_BLOCK_PATTERN; $lines[$key] = substr($line, $min_space); } } } // The input text may have linewraps in it, like this: // // - derp derp derp derp // derp derp derp derp // - blarp blarp blarp blarp // // Group text lines together into list items, stored in $items. So the // result in the above case will be: // // array( // array( // "- derp derp derp derp", // " derp derp derp derp", // ), // array( // "- blarp blarp blarp blarp", // ), // ); $item = array(); $regex = self::START_BLOCK_PATTERN; foreach ($lines as $line) { if (preg_match($regex, $line)) { $regex = self::CONT_BLOCK_PATTERN; if ($item) { $items[] = $item; $item = array(); } } $item[] = $line; } if ($item) { $items[] = $item; } // Process each item to normalize the text, remove line wrapping, and // determine its depth (indentation level) and style (ordered vs unordered). // // Given the above example, the processed array will look like: // // array( // array( // 'text' => 'derp derp derp derp derp derp derp derp', // 'depth' => 0, // 'style' => '-', // ), // array( // 'text' => 'blarp blarp blarp blarp', // 'depth' => 0, // 'style' => '-', // ), // ); foreach ($items as $key => $item) { $item = preg_replace('/\s*\n\s*/', ' ', implode("\n", $item)); $item = rtrim($item); if (!strlen($item)) { unset($items[$key]); continue; } $matches = null; if (preg_match('/^\s*([-*#]{2,})/', $item, $matches)) { // Alternate-style indents; use number of list item symbols. $depth = strlen($matches[1]) - 1; } else if (preg_match('/^(\s+)/', $item, $matches)) { // Markdown-style indents; use indent depth. $depth = strlen($matches[1]); } else { $depth = 0; } if (preg_match('/^\s*(?:#|[0-9])/', $item)) { $style = '#'; } else { $style = '-'; } // If we don't match the block pattern (for example, because the user // has typed only " " or " -"), treat the line as containing no text. // This prevents newly added items from rendering with a bullet and // the text "-", e.g. $text = preg_replace(self::CONT_BLOCK_PATTERN, '', $item); if ($text == $item) { $text = ''; } $items[$key] = array( 'text' => $text, 'depth' => $depth, 'style' => $style, ); } $items = array_values($items); // Users can create a sub-list by indenting any deeper amount than the // previous list, so these are both valid: // // - a // - b // // - a // - b // // In the former case, we'll have depths (0, 2). In the latter case, depths // (0, 4). We don't actually care about how many spaces there are, only // how many list indentation levels (that is, we want to map both of // those cases to (0, 1), indicating "outermost list" and "first sublist"). // // This is made more complicated because lists at two different indentation // levels might be at the same list level: // // - a // - b // - c // - d // // Here, 'b' and 'd' are at the same list level (2) but different indent // levels (2, 4). // // Users can also create "staircases" like this: // // - a // - b // # c // // While this is silly, we'd like to render it as faithfully as possible. // // In order to do this, we convert the list of nodes into a tree, // normalizing indentation levels and inserting dummy nodes as necessary to // make the tree well-formed. See additional notes at buildTree(). // // In the case above, the result is a tree like this: // // -