Various html parsing fixes

CHANGELOG.md

@@ -8,10 +8,14 @@ v0.15.0 (?? ??? 2018)
 ### Added
 - Parsing of "align" HTML attribute - @troosan #1231
 - Parse formatting inside HTML lists - @troosan @samimussbach #1239 #945 #1215 #508
+- Parsing of CSS `direction` instruction, HTML `lang` attribute, formatting inside table cell - @troosan #1273 #1252 #1254
 - Add support for Track changes @Cip @troosan #354 #1262
 
 ### Fixed
 - fix reading of docx default style - @troosan #1238
+- fix the size unit of when parsing html images - @troosan #1254
+- fixed HTML parsing of nested lists - @troosan #1265
+- Save PNG alpha information when using remote images. @samsullivan #779
 
 
 

samples/Sample_26_Html.php

@@ -7,11 +7,13 @@
 
 $section = $phpWord->addSection();
 $html = '<h1>Adding element via HTML</h1>';
-$html .= '<p>Some well formed HTML snippet needs to be used</p>';
+$html .= '<p>Some well-formed HTML snippet needs to be used</p>';
 $html .= '<p>With for example <strong>some<sup>1</sup> <em>inline</em> formatting</strong><sub>1</sub></p>';
 
 $html .= '<p>A link to <a href="http://phpword.readthedocs.io/">Read the docs</a></p>';
 
+$html .= '<p lang="he-IL" style="text-align: right; direction: rtl">היי, זה פסקה מימין לשמאל</p>';
+
 $html .= '<p style="margin-top: 240pt;">Unordered (bulleted) list:</p>';
 $html .= '<ul><li>Item 1</li><li>Item 2</li><ul><li>Item 2.1</li><li>Item 2.1</li></ul></ul>';
 
@@ -29,10 +31,12 @@
             <ol>
                 <li>List 2 item 1</li>
                 <li>List 2 item 2</li>
-                <ol>
-                    <li>sub list 1</li>
-                    <li>sub list 2</li>
-                </ol>
+                <li>
+                    <ol>
+                        <li>sub list 1</li>
+                        <li>sub list 2</li>
+                    </ol>
+                </li>
                 <li>List 2 item 3</li>
                 <ol>
                     <li>sub list 1, restarts with a</li>
@@ -65,10 +69,20 @@
                 </thead>
                 <tbody>
                     <tr><td style="border-style: dotted;">1</td><td colspan="2">2</td></tr>
-                    <tr><td>4</td><td>5</td><td>6</td></tr>
+                    <tr><td>This is <b>bold</b> text</td><td></td><td>6</td></tr>
                 </tbody>
             </table>';
 
+$html .= '<p style="margin-top: 240pt;">Table inside another table:</p>';
+$html .= '<table align="center" style="width: 80%; border: 6px #0000FF double;">
+    <tr><td>
+        <table style="width: 100%; border: 4px #FF0000 dotted;">
+            <tr><td>column 1</td><td>column 2</td></tr>
+        </table>
+    </td></tr>
+    <tr><td style="text-align: center;">Cell in parent table</td></tr>
+</table>';
+
 \PhpOffice\PhpWord\Shared\Html::addHtml($section, $html, false, false);
 
 // Save file

src/PhpWord/Element/AbstractElement.php

@@ -93,6 +93,13 @@ abstract class AbstractElement
      */
     private $nestedLevel = 0;
 
+    /**
+     * A reference to the parent
+     *
+     * @var \PhpOffice\PhpWord\Element\AbstractElement
+     */
+    private $parent;
+
     /**
      * changed element info
      *
@@ -328,6 +335,11 @@ public function setCommentRangeEnd(Comment $value)
         $this->commentRangeEnd->setEndElement($this);
     }
 
+    public function getParent()
+    {
+        return $this->parent;
+    }
+
     /**
      * Set parent container
      *
@@ -338,6 +350,7 @@ public function setCommentRangeEnd(Comment $value)
     public function setParentContainer(AbstractElement $container)
     {
         $this->parentContainer = substr(get_class($container), strrpos(get_class($container), '\\') + 1);
+        $this->parent = $container;
 
         // Set nested level
         $this->nestedLevel = $container->getNestedLevel();

src/PhpWord/Shared/Html.php

@@ -31,6 +31,7 @@
 class Html
 {
     private static $listIndex = 0;
+    private static $xpath;
 
     /**
      * Add HTML parts.
@@ -65,6 +66,7 @@ public static function addHtml($element, $html, $fullHTML = false, $preserveWhit
         $dom = new \DOMDocument();
         $dom->preserveWhiteSpace = $preserveWhiteSpace;
         $dom->loadXML($html);
+        self::$xpath = new \DOMXpath($dom);
         $node = $dom->getElementsByTagName('body');
 
         self::parseNode($node->item(0), $element);
@@ -89,6 +91,10 @@ protected static function parseInlineStyle($node, $styles = array())
                         break;
                     case 'align':
                         $styles['alignment'] = self::mapAlign($attribute->value);
+                        break;
+                    case 'lang':
+                        $styles['lang'] = $attribute->value;
+                        break;
                 }
             }
         }
@@ -333,7 +339,7 @@ private static function parseRow($node, $element, &$styles)
      * @param \DOMNode $node
      * @param \PhpOffice\PhpWord\Element\Table $element
      * @param array &$styles
-     * @return \PhpOffice\PhpWord\Element\Cell $element
+     * @return \PhpOffice\PhpWord\Element\Cell|\PhpOffice\PhpWord\Element\TextRun $element
      */
     private static function parseCell($node, $element, &$styles)
     {
@@ -343,8 +349,29 @@ private static function parseCell($node, $element, &$styles)
         if (!empty($colspan)) {
             $cellStyles['gridSpan'] = $colspan - 0;
         }
+        $cell = $element->addCell(null, $cellStyles);
+
+        if (self::shouldAddTextRun($node)) {
+            return $cell->addTextRun(self::parseInlineStyle($node, $styles['paragraph']));
+        }
 
-        return $element->addCell(null, $cellStyles);
+        return $cell;
+    }
+
+    /**
+     * Checks if $node contains an HTML element that cannot be added to TextRun
+     *
+     * @param \DOMNode $node
+     * @return bool Returns true if the node contains an HTML element that cannot be added to TextRun
+     */
+    private static function shouldAddTextRun(\DOMNode $node)
+    {
+        $containsBlockElement = self::$xpath->query('.//table|./p|./ul|./ol', $node)->length > 0;
+        if ($containsBlockElement) {
+            return false;
+        }
+
+        return true;
     }
 
     /**
@@ -375,14 +402,17 @@ private static function recursiveParseStylesInHierarchy(\DOMNode $node, array $s
      */
     private static function parseList($node, $element, &$styles, &$data)
     {
-        $isOrderedList = $node->nodeName == 'ol';
+        $isOrderedList = $node->nodeName === 'ol';
         if (isset($data['listdepth'])) {
             $data['listdepth']++;
         } else {
             $data['listdepth'] = 0;
             $styles['list'] = 'listStyle_' . self::$listIndex++;
             $element->getPhpWord()->addNumberingStyle($styles['list'], self::getListStyle($isOrderedList));
         }
+        if ($node->parentNode->nodeName === 'li') {
+            return $element->getParent();
+        }
     }
 
     private static function getListStyle($isOrderedList)
@@ -469,6 +499,9 @@ private static function parseStyle($attribute, $styles)
                 case 'text-align':
                     $styles['alignment'] = self::mapAlign($cValue);
                     break;
+                case 'direction':
+                    $styles['rtl'] = $cValue === 'rtl';
+                    break;
                 case 'font-size':
                     $styles['size'] = Converter::cssToPoint($cValue);
                     break;
@@ -556,10 +589,12 @@ private static function parseImage($node, $element)
                 case 'width':
                     $width = $attribute->value;
                     $style['width'] = $width;
+                    $style['unit'] = \PhpOffice\PhpWord\Style\Image::UNIT_PX;
                     break;
                 case 'height':
                     $height = $attribute->value;
                     $style['height'] = $height;
+                    $style['unit'] = \PhpOffice\PhpWord\Style\Image::UNIT_PX;
                     break;
                 case 'style':
                     $styleattr = explode(';', $attribute->value);

src/PhpWord/Writer/Word2007/Style/Font.php

@@ -90,6 +90,10 @@ private function writeStyle()
             $xmlWriter->writeAttributeIf($language->getLatin() !== null, 'w:val', $language->getLatin());
             $xmlWriter->writeAttributeIf($language->getEastAsia() !== null, 'w:eastAsia', $language->getEastAsia());
             $xmlWriter->writeAttributeIf($language->getBidirectional() !== null, 'w:bidi', $language->getBidirectional());
+            //if bidi is not set but we are writing RTL, write the latin language in the bidi tag
+            if ($style->isRTL() && $language->getBidirectional() === null && $language->getLatin() !== null) {
+                $xmlWriter->writeAttribute('w:bidi', $language->getLatin());
+            }
             $xmlWriter->endElement();
         }
 

tests/PhpWord/Shared/HtmlTest.php

@@ -150,6 +150,33 @@ public function testParseFontSize()
         $this->assertEquals('15', $doc->getElementAttribute('/w:document/w:body/w:p[2]/w:r/w:rPr/w:sz', 'w:val'));
     }
 
+    /**
+     * Test direction style
+     */
+    public function testParseTextDirection()
+    {
+        $phpWord = new \PhpOffice\PhpWord\PhpWord();
+        $section = $phpWord->addSection();
+        Html::addHtml($section, '<span style="direction: rtl">test</span>');
+
+        $doc = TestHelperDOCX::getDocument($phpWord, 'Word2007');
+        $this->assertTrue($doc->elementExists('/w:document/w:body/w:p/w:r/w:rPr/w:rtl'));
+    }
+
+    /**
+     * Test html lang
+     */
+    public function testParseLang()
+    {
+        $phpWord = new \PhpOffice\PhpWord\PhpWord();
+        $section = $phpWord->addSection();
+        Html::addHtml($section, '<span lang="fr-BE">test</span>');
+
+        $doc = TestHelperDOCX::getDocument($phpWord, 'Word2007');
+        $this->assertTrue($doc->elementExists('/w:document/w:body/w:p/w:r/w:rPr/w:lang'));
+        $this->assertEquals('fr-BE', $doc->getElementAttribute('/w:document/w:body/w:p/w:r/w:rPr/w:lang', 'w:val'));
+    }
+
     /**
      * Test font-family style
      */
@@ -199,7 +226,7 @@ public function testParseTable()
                 </thead>
                 <tbody>
                     <tr><td style="border-style: dotted;">1</td><td colspan="2">2</td></tr>
-                    <tr><td>4</td><td>5</td><td>6</td></tr>
+                    <tr><td>This is <b>bold</b> text</td><td>5</td><td><p>6</p></td></tr>
                 </tbody>
             </table>';
         Html::addHtml($section, $html);
@@ -272,6 +299,43 @@ public function testOrderedListNumbering()
         $this->assertNotEquals($firstListnumId, $secondListnumId);
     }
 
+    /**
+     * Tests parsing of nested ul/li
+     */
+    public function testOrderedNestedListNumbering()
+    {
+        $phpWord = new \PhpOffice\PhpWord\PhpWord();
+        $section = $phpWord->addSection();
+        $html = '<ol>
+                <li>List 1 item 1</li>
+                <li>List 1 item 2</li>
+            </ol>
+            <p>Some Text</p>
+            <ol>
+                <li>List 2 item 1</li>
+                <li>
+                    <ol>
+                        <li>sub list 1</li>
+                        <li>sub list 2</li>
+                    </ol>
+                </li>
+            </ol>';
+        Html::addHtml($section, $html, false, false);
+
+        $doc = TestHelperDOCX::getDocument($phpWord, 'Word2007');
+
+        $this->assertTrue($doc->elementExists('/w:document/w:body/w:p/w:pPr/w:numPr/w:numId'));
+        $this->assertTrue($doc->elementExists('/w:document/w:body/w:p/w:r/w:t'));
+
+        $this->assertEquals('List 1 item 1', $doc->getElement('/w:document/w:body/w:p[1]/w:r/w:t')->nodeValue);
+        $this->assertEquals('List 2 item 1', $doc->getElement('/w:document/w:body/w:p[4]/w:r/w:t')->nodeValue);
+
+        $firstListnumId = $doc->getElementAttribute('/w:document/w:body/w:p[1]/w:pPr/w:numPr/w:numId', 'w:val');
+        $secondListnumId = $doc->getElementAttribute('/w:document/w:body/w:p[4]/w:pPr/w:numPr/w:numId', 'w:val');
+
+        $this->assertNotEquals($firstListnumId, $secondListnumId);
+    }
+
     /**
      * Tests parsing of ul/li
      */
@@ -336,8 +400,8 @@ public function testParseImage()
 
         $baseXpath = '/w:document/w:body/w:p/w:r';
         $this->assertTrue($doc->elementExists($baseXpath . '/w:pict/v:shape'));
-        $this->assertStringMatchesFormat('%Swidth:150pt%S', $doc->getElementAttribute($baseXpath . '[1]/w:pict/v:shape', 'style'));
-        $this->assertStringMatchesFormat('%Sheight:200pt%S', $doc->getElementAttribute($baseXpath . '[1]/w:pict/v:shape', 'style'));
+        $this->assertStringMatchesFormat('%Swidth:150px%S', $doc->getElementAttribute($baseXpath . '[1]/w:pict/v:shape', 'style'));
+        $this->assertStringMatchesFormat('%Sheight:200px%S', $doc->getElementAttribute($baseXpath . '[1]/w:pict/v:shape', 'style'));
         $this->assertStringMatchesFormat('%Smso-position-horizontal:right%S', $doc->getElementAttribute($baseXpath . '[1]/w:pict/v:shape', 'style'));
         $this->assertStringMatchesFormat('%Smso-position-horizontal:left%S', $doc->getElementAttribute($baseXpath . '[2]/w:pict/v:shape', 'style'));
     }