friendica-github/include/markdownify/markdownify_extra.php

<?php
/**
 * Class to convert HTML to Markdown with PHP Markdown Extra syntax support.
 *
 * @version 1.0.0 alpha
 * @author Milian Wolff (<mail@milianw.de>, <http://milianw.de>)
 * @license LGPL, see LICENSE_LGPL.txt and the summary below
 * @copyright (C) 2007  Milian Wolff
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
 */

/**
 * standard Markdownify class
 */
require_once dirname(__FILE__).'/markdownify.php';

class Markdownify_Extra extends Markdownify {
  /**
   * table data, including rows with content and the maximum width of each col
   *
   * @var array
   */
  var $table = array();
  /**
   * current col
   *
   * @var int
   */
  var $col = -1;
  /**
   * current row
   *
   * @var int
   */
  var $row = 0;
  /**
   * constructor, see Markdownify::Markdownify() for more information
   */
  function Markdownify_Extra($linksAfterEachParagraph = MDFY_LINKS_EACH_PARAGRAPH, $bodyWidth = MDFY_BODYWIDTH, $keepHTML = MDFY_KEEPHTML) {
    parent::Markdownify($linksAfterEachParagraph, $bodyWidth, $keepHTML);

    ### new markdownable tags & attributes
    # header ids: # foo {bar}
    $this->isMarkdownable['h1']['id'] = 'optional';
    $this->isMarkdownable['h2']['id'] = 'optional';
    $this->isMarkdownable['h3']['id'] = 'optional';
    $this->isMarkdownable['h4']['id'] = 'optional';
    $this->isMarkdownable['h5']['id'] = 'optional';
    $this->isMarkdownable['h6']['id'] = 'optional';
    # tables
    $this->isMarkdownable['table'] = array();
    $this->isMarkdownable['th'] = array(
      'align' => 'optional',
    );
    $this->isMarkdownable['td'] = array(
      'align' => 'optional',
    );
    $this->isMarkdownable['tr'] = array();
    array_push($this->ignore, 'thead');
    array_push($this->ignore, 'tbody');
    array_push($this->ignore, 'tfoot');
    # definition lists
    $this->isMarkdownable['dl'] = array();
    $this->isMarkdownable['dd'] = array();
    $this->isMarkdownable['dt'] = array();
    # footnotes
    $this->isMarkdownable['fnref'] = array(
      'target' => 'required',
    );
    $this->isMarkdownable['footnotes'] = array();
    $this->isMarkdownable['fn'] = array(
      'name' => 'required',
    );
    $this->parser->blockElements['fnref'] = false;
    $this->parser->blockElements['fn'] = true;
    $this->parser->blockElements['footnotes'] = true;
    # abbr
    $this->isMarkdownable['abbr'] = array(
      'title' => 'required',
    );
    # build RegEx lookahead to decide wether table can pe parsed or not
    $inlineTags = array_keys($this->parser->blockElements, false);
    $colContents = '(?:[^<]|<(?:'.implode('|', $inlineTags).'|[^a-z]))+';
    $this->tableLookaheadHeader = '{
    ^\s*(?:<thead\s*>)?\s*                               # open optional thead
      <tr\s*>\s*(?:                                    # start required row with headers
        <th(?:\s+align=("|\')(?:left|center|right)\1)?\s*>   # header with optional align
        \s*'.$colContents.'\s*                       # contents
        </th>\s*                                     # close header
      )+</tr>                                          # close row with headers
    \s*(?:</thead>)?                                     # close optional thead
    }sxi';
    $this->tdSubstitute = '\s*'.$colContents.'\s*        # contents
          </td>\s*';
    $this->tableLookaheadBody = '{
      \s*(?:<tbody\s*>)?\s*                            # open optional tbody
        (?:<tr\s*>\s*                                # start row
          %s                                       # cols to be substituted
        </tr>)+                                      # close row
      \s*(?:</tbody>)?                                 # close optional tbody
    \s*</table>                                          # close table
    }sxi';
  }
  /**
   * handle header tags (<h1> - <h6>)
   *
   * @param int $level 1-6
   * @return void
   */
  function handleHeader($level) {
    static $id = null;
    if ($this->parser->isStartTag) {
      if (isset($this->parser->tagAttributes['id'])) {
        $id = $this->parser->tagAttributes['id'];
      }
    } else {
      if (!is_null($id)) {
        $this->out(' {#'.$id.'}');
        $id = null;
      }
    }
    parent::handleHeader($level);
  }
  /**
   * handle <abbr> tags
   *
   * @param void
   * @return void
   */
  function handleTag_abbr() {
    if ($this->parser->isStartTag) {
      $this->stack();
      $this->buffer();
    } else {
      $tag = $this->unstack();
      $tag['text'] = $this->unbuffer();
      $add = true;
      foreach ($this->stack['abbr'] as $stacked) {
        if ($stacked['text'] == $tag['text']) {
          /** TODO: differing abbr definitions, i.e. different titles for same text **/
          $add = false;
          break;
        }
      }
      $this->out($tag['text']);
      if ($add) {
        array_push($this->stack['abbr'], $tag);
      }
    }
  }
  /**
   * flush stacked abbr tags
   *
   * @param void
   * @return void
   */
  function flushStacked_abbr() {
    $out = array();
    foreach ($this->stack['abbr'] as $k => $tag) {
      if (!isset($tag['unstacked'])) {
        array_push($out, ' *['.$tag['text'].']: '.$tag['title']);
        $tag['unstacked'] = true;
        $this->stack['abbr'][$k] = $tag;
      }
    }
    if (!empty($out)) {
      $this->out("\n\n".implode("\n", $out));
    }
  }
  /**
   * handle <table> tags
   *
   * @param void
   * @return void
   */
  function handleTag_table() {
    if ($this->parser->isStartTag) {
      # check if upcoming table can be converted
      if ($this->keepHTML) {
        if (preg_match($this->tableLookaheadHeader, $this->parser->html, $matches)) {
          # header seems good, now check body
          # get align & number of cols
          preg_match_all('#<th(?:\s+align=("|\')(left|right|center)\1)?\s*>#si', $matches[0], $cols);
          $regEx = '';
          $i = 1;
          $aligns = array();
          foreach ($cols[2] as $align) {
            $align = strtolower($align);
            array_push($aligns, $align);
            if (empty($align)) {
              $align = 'left'; # default value
            }
            $td = '\s+align=("|\')'.$align.'\\'.$i;
            $i++;
            if ($align == 'left') {
              # look for empty align or left
              $td = '(?:'.$td.')?';
            }
            $td = '<td'.$td.'\s*>';
            $regEx .= $td.$this->tdSubstitute;
          }
          $regEx = sprintf($this->tableLookaheadBody, $regEx);
          if (preg_match($regEx, $this->parser->html, $matches, null, strlen($matches[0]))) {
            # this is a markdownable table tag!
            $this->table = array(
              'rows' => array(),
              'col_widths' => array(),
              'aligns' => $aligns,
            );
            $this->row = 0;
          } else {
            # non markdownable table
            $this->handleTagToText();
          }
        } else {
          # non markdownable table
          $this->handleTagToText();
        }
      } else {
        $this->table = array(
          'rows' => array(),
          'col_widths' => array(),
          'aligns' => array(),
        );
        $this->row = 0;
      }
    } else {
      # finally build the table in Markdown Extra syntax
      $separator = array();
      # seperator with correct align identifikators
      foreach($this->table['aligns'] as $col => $align) {
        if (!$this->keepHTML && !isset($this->table['col_widths'][$col])) {
          break;
        }
        $left = ' ';
        $right = ' ';
        switch ($align) {
          case 'left':
            $left = ':';
            break;
          case 'center':
            $right = ':';
            $left = ':';
          case 'right':
            $right = ':';
            break;
        }
        array_push($separator, $left.str_repeat('-', $this->table['col_widths'][$col]).$right);
      }
      $separator = '|'.implode('|', $separator).'|';

      $rows = array();
      # add padding
      array_walk_recursive($this->table['rows'], array(&$this, 'alignTdContent'));
      $header = array_shift($this->table['rows']);
      array_push($rows, '| '.implode(' | ', $header).' |');
      array_push($rows, $separator);
      foreach ($this->table['rows'] as $row) {
        array_push($rows, '| '.implode(' | ', $row).' |');
      }
      $this->out(implode("\n".$this->indent, $rows));
      $this->table = array();
      $this->setLineBreaks(2);
    }
  }
  /**
   * properly pad content so it is aligned as whished
   * should be used with array_walk_recursive on $this->table['rows']
   *
   * @param string &$content
   * @param int $col
   * @return void
   */
  function alignTdContent(&$content, $col) {
    switch ($this->table['aligns'][$col]) {
      default:
      case 'left':
        $content .= str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content));
        break;
      case 'right':
        $content = str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content)).$content;
        break;
      case 'center':
        $paddingNeeded = $this->table['col_widths'][$col] - $this->strlen($content);
        $left = floor($paddingNeeded / 2);
        $right = $paddingNeeded - $left;
        $content = str_repeat(' ', $left).$content.str_repeat(' ', $right);
        break;
    }
  }
  /**
   * handle <tr> tags
   *
   * @param void
   * @return void
   */
  function handleTag_tr() {
    if ($this->parser->isStartTag) {
      $this->col = -1;
    } else {
      $this->row++;
    }
  }
  /**
   * handle <td> tags
   *
   * @param void
   * @return void
   */
  function handleTag_td() {
    if ($this->parser->isStartTag) {
      $this->col++;
      if (!isset($this->table['col_widths'][$this->col])) {
        $this->table['col_widths'][$this->col] = 0;
      }
      $this->buffer();
    } else {
      $buffer = trim($this->unbuffer());
      $this->table['col_widths'][$this->col] = max($this->table['col_widths'][$this->col], $this->strlen($buffer));
      $this->table['rows'][$this->row][$this->col] = $buffer;
    }
  }
  /**
   * handle <th> tags
   *
   * @param void
   * @return void
   */
  function handleTag_th() {
    if (!$this->keepHTML && !isset($this->table['rows'][1]) && !isset($this->table['aligns'][$this->col+1])) {
      if (isset($this->parser->tagAttributes['align'])) {
        $this->table['aligns'][$this->col+1] = $this->parser->tagAttributes['align'];
      } else {
        $this->table['aligns'][$this->col+1] = '';
      }
    }
    $this->handleTag_td();
  }
  /**
   * handle <dl> tags
   *
   * @param void
   * @return void
   */
  function handleTag_dl() {
    if (!$this->parser->isStartTag) {
      $this->setLineBreaks(2);
    }
  }
  /**
   * handle <dt> tags
   *
   * @param void
   * @return void
   **/
  function handleTag_dt() {
    if (!$this->parser->isStartTag) {
      $this->setLineBreaks(1);
    }
  }
  /**
   * handle <dd> tags
   *
   * @param void
   * @return void
   */
  function handleTag_dd() {
    if ($this->parser->isStartTag) {
      if (substr(ltrim($this->parser->html), 0, 3) == '<p>') {
        # next comes a paragraph, so we'll need an extra line
        $this->out("\n".$this->indent);
      } elseif (substr($this->output, -2) == "\n\n") {
        $this->output = substr($this->output, 0, -1);
      }
      $this->out(':   ');
      $this->indent('    ', false);
    } else {
      # lookahead for next dt
      if (substr(ltrim($this->parser->html), 0, 4) == '<dt>') {
        $this->setLineBreaks(2);
      } else {
        $this->setLineBreaks(1);
      }
      $this->indent('    ');
    }
  }
  /**
   * handle <fnref /> tags (custom footnote references, see markdownify_extra::parseString())
   *
   * @param void
   * @return void
   */
  function handleTag_fnref() {
    $this->out('[^'.$this->parser->tagAttributes['target'].']');
  }
  /**
   * handle <fn> tags (custom footnotes, see markdownify_extra::parseString()
   * and markdownify_extra::_makeFootnotes())
   *
   * @param void
   * @return void
   */
  function handleTag_fn() {
    if ($this->parser->isStartTag) {
      $this->out('[^'.$this->parser->tagAttributes['name'].']:');
      $this->setLineBreaks(1);
    } else {
      $this->setLineBreaks(2);
    }
    $this->indent('    ');
  }
  /**
   * handle <footnotes> tag (custom footnotes, see markdownify_extra::parseString()
   *  and markdownify_extra::_makeFootnotes())
   *
   *  @param void
   *  @return void
   */
  function handleTag_footnotes() {
    if (!$this->parser->isStartTag) {
      $this->setLineBreaks(2);
    }
  }
  /**
   * parse a HTML string, clean up footnotes prior
   *
   * @param string $HTML input
   * @return string Markdown formatted output
   */
  function parseString($html) {
    /** TODO: custom markdown-extra options, e.g. titles & classes **/
    # <sup id="fnref:..."><a href"#fn..." rel="footnote">...</a></sup>
    # => <fnref target="..." />
    $html = preg_replace('@<sup id="fnref:([^"]+)">\s*<a href="#fn:\1" rel="footnote">\s*\d+\s*</a>\s*</sup>@Us', '<fnref target="$1" />', $html);
    # <div class="footnotes">
    # <hr />
    # <ol>
    #
    # <li id="fn:...">...</li>
    # ...
    #
    # </ol>
    # </div>
    # =>
    # <footnotes>
    #   <fn name="...">...</fn>
    #   ...
    # </footnotes>
    $html = preg_replace_callback('#<div class="footnotes">\s*<hr />\s*<ol>\s*(.+)\s*</ol>\s*</div>#Us', array(&$this, '_makeFootnotes'), $html);
    return parent::parseString($html);
  }
  /**
   * replace HTML representation of footnotes with something more easily parsable
   *
   * @note this is a callback to be used in parseString()
   *
   * @param array $matches
   * @return string
   */
  function _makeFootnotes($matches) {
    # <li id="fn:1">
    #   ...
    #   <a href="#fnref:block" rev="footnote">&#8617;</a></p>
    # </li>
    # => <fn name="1">...</fn>
    # remove footnote link
    $fns = preg_replace('@\s*(&#160;\s*)?<a href="#fnref:[^"]+" rev="footnote"[^>]*>&#8617;</a>\s*@s', '', $matches[1]);
    # remove empty paragraph
    $fns = preg_replace('@<p>\s*</p>@s', '', $fns);
    # <li id="fn:1">...</li> -> <footnote nr="1">...</footnote>
    $fns = str_replace('<li id="fn:', '<fn name="', $fns);

    $fns = '<footnotes>'.$fns.'</footnotes>';
    return preg_replace('#</li>\s*(?=(?:<fn|</footnotes>))#s', '</fn>$1', $fns);
  }
}