Improve ODText Content Reader

Fix PHPOffice#2493. There is much that the ODT Reader ignores. This change adds support for the `text:section`, `text:span`, `text:s`, and `text:tab` tags, thereby handling multiple sections, text runs, tab characters, and multiple spaces. There will still be many omissions (e.g. styles and tables), but you will now often be able to access the text content of valid ODT documents. The issue suggests variations in a simple file created on its own by LibreOffice, and a similar file created by PhpWord. Both are unit-tested.

A `getText` method is added to TextRun to facilitate testing (and can be useful on its own). It will return the concatenated texts of all elements of the text run.

phpstan-baseline.neon

@@ -165,11 +165,6 @@ parameters:
 			count: 1
 			path: src/PhpWord/Reader/HTML.php
 
-		-
-			message: "#^Call to an undefined method DOMNode\\:\\:getAttribute\\(\\)\\.$#"
-			count: 2
-			path: src/PhpWord/Reader/ODText/Content.php
-
 		-
 			message: "#^Offset 'textNodes' on array\\{changed\\: PhpOffice\\\\PhpWord\\\\Element\\\\TrackChange, textNodes\\: DOMNodeList\\<DOMElement\\>\\} in isset\\(\\) always exists and is not nullable\\.$#"
 			count: 1

src/PhpWord/Element/TextRun.php

@@ -78,4 +78,16 @@ public function setParagraphStyle($style = null)
 
         return $this->paragraphStyle;
     }
+
+    public function getText(): string
+    {
+        $outstr = '';
+        foreach ($this->getElements() as $element) {
+            if ($element instanceof Text) {
+                $outstr .= $element->getText();
+            }
+        }
+
+        return $outstr;
+    }
 }

src/PhpWord/Reader/ODText/Content.php

@@ -18,6 +18,9 @@
 namespace PhpOffice\PhpWord\Reader\ODText;
 
 use DateTime;
+use DOMElement;
+use DOMNodeList;
+use PhpOffice\PhpWord\Element\Section;
 use PhpOffice\PhpWord\Element\TrackChange;
 use PhpOffice\PhpWord\PhpWord;
 use PhpOffice\PhpWord\Shared\XMLReader;
@@ -29,6 +32,9 @@
  */
 class Content extends AbstractPart
 {
+    /** @var ?Section */
+    private $section;
+
     /**
      * Read content.xml.
      */
@@ -40,18 +46,31 @@ public function read(PhpWord $phpWord): void
         $trackedChanges = [];
 
         $nodes = $xmlReader->getElements('office:body/office:text/*');
+        $this->section = null;
+        $this->processNodes($nodes, $xmlReader, $phpWord);
+        $this->section = null;
+    }
+
+    /** @param DOMNodeList<DOMElement> $nodes */
+    public function processNodes(DOMNodeList $nodes, XMLReader $xmlReader, PhpWord $phpWord): void
+    {
         if ($nodes->length > 0) {
-            $section = $phpWord->addSection();
             foreach ($nodes as $node) {
                 // $styleName = $xmlReader->getAttribute('text:style-name', $node);
                 switch ($node->nodeName) {
                     case 'text:h': // Heading
                         $depth = $xmlReader->getAttribute('text:outline-level', $node);
-                        $section->addTitle($node->nodeValue, $depth);
+                        $this->getSection($phpWord)->addTitle($node->nodeValue, $depth);
 
                         break;
                     case 'text:p': // Paragraph
+                        $styleName = $xmlReader->getAttribute('text:style-name', $node);
+                        if (substr($styleName, 0, 2) === 'SB') {
+                            break;
+                        }
                         $children = $node->childNodes;
+                        $spans = false;
+                        /** @var DOMElement $child */
                         foreach ($children as $child) {
                             switch ($child->nodeName) {
                                 case 'text:change-start':
@@ -71,16 +90,50 @@ public function read(PhpWord $phpWord): void
                                         $changed = $trackedChanges[$changeId];
                                     }
 
+                                    break;
+                                case 'text:span':
+                                    $spans = true;
+
                                     break;
                             }
                         }
 
-                        $element = $section->addText($node->nodeValue);
+                        if ($spans) {
+                            $element = $this->getSection($phpWord)->addTextRun();
+                            foreach ($children as $child) {
+                                switch ($child->nodeName) {
+                                    case 'text:span':
+                                        /** @var DOMElement $child2 */
+                                        foreach ($child->childNodes as $child2) {
+                                            switch ($child2->nodeName) {
+                                                case '#text':
+                                                    $element->addText($child2->nodeValue);
+
+                                                    break;
+                                                case 'text:tab':
+                                                    $element->addText("\t");
+
+                                                    break;
+                                                case 'text:s':
+                                                    $spaces = (int) $child2->getAttribute('text:c') ?: 1;
+                                                    $element->addText(str_repeat(' ', $spaces));
+
+                                                    break;
+                                            }
+                                        }
+
+                                        break;
+
+                                }
+                            }
+                        } else {
+                            $element = $this->getSection($phpWord)->addText($node->nodeValue);
+                        }
                         if (isset($changed) && is_array($changed)) {
                             $element->setTrackChange($changed['changed']);
                             if (isset($changed['textNodes'])) {
                                 foreach ($changed['textNodes'] as $changedNode) {
-                                    $element = $section->addText($changedNode->nodeValue);
+                                    $element = $this->getSection($phpWord)->addText($changedNode->nodeValue);
                                     $element->setTrackChange($changed['changed']);
                                 }
                             }
@@ -91,7 +144,7 @@ public function read(PhpWord $phpWord): void
                         $listItems = $xmlReader->getElements('text:list-item/text:p', $node);
                         foreach ($listItems as $listItem) {
                             // $listStyleName = $xmlReader->getAttribute('text:style-name', $listItem);
-                            $section->addListItem($listItem->nodeValue, 0);
+                            $this->getSection($phpWord)->addListItem($listItem->nodeValue, 0);
                         }
 
                         break;
@@ -110,9 +163,26 @@ public function read(PhpWord $phpWord): void
                             $trackedChanges[$changedRegion->getAttribute('text:id')] = ['changed' => $changed, 'textNodes' => $textNodes];
                         }
 
+                        break;
+                    case 'text:section': // Section
+                        // $sectionStyleName = $xmlReader->getAttribute('text:style-name', $listItem);
+                        $this->section = $phpWord->addSection();
+                        $children = $node->childNodes;
+                        $this->processNodes($children, $xmlReader, $phpWord);
+
                         break;
                 }
             }
         }
     }
+
+    private function getSection(PhpWord $phpWord): Section
+    {
+        $section = $this->section;
+        if ($section === null) {
+            $section = $this->section = $phpWord->addSection();
+        }
+
+        return $section;
+    }
 }

tests/PhpWordTests/Reader/ODText/ODTextSectionTest.php

@@ -0,0 +1,83 @@
+<?php
+/**
+ * This file is part of PHPWord - A pure PHP library for reading and writing
+ * word processing documents.
+ *
+ * PHPWord is free software distributed under the terms of the GNU Lesser
+ * General Public License version 3 as published by the Free Software Foundation.
+ *
+ * For the full copyright and license information, please read the LICENSE
+ * file that was distributed with this source code. For the full list of
+ * contributors, visit https://github.com/PHPOffice/PHPWord/contributors.
+ *
+ * @see         https://github.com/PHPOffice/PHPWord
+ *
+ * @license     http://www.gnu.org/licenses/lgpl.txt LGPL version 3
+ */
+
+namespace PhpOffice\PhpWordTests\Reader\ODText;
+
+use PhpOffice\PhpWord\IOFactory;
+use PhpOffice\PhpWord\PhpWord;
+use PhpOffice\PhpWord\Settings;
+
+class ODTextSectionTest extends \PHPUnit\Framework\TestCase
+{
+    /** @var string */
+    private $filename = '';
+
+    protected function tearDown(): void
+    {
+        if ($this->filename !== '') {
+            unlink($this->filename);
+            $this->filename = '';
+        }
+    }
+
+    public function testWriteThenReadSection(): void
+    {
+        $dir = 'tests/PhpWordTests/_files';
+        Settings::setOutputEscapingEnabled(true);
+        $phpWord = new PhpWord();
+        $section = $phpWord->addSection();
+        $inputText = ['days', 'monday', 'tuesday'];
+        $inputText[] = "Tab\tthen two spaces  then done.";
+        foreach ($inputText as $text) {
+            $section->addText($text);
+        }
+        $writer = IOFactory::createWriter($phpWord, 'ODText');
+        $this->filename = "$dir/sectiontest.odt";
+        $writer->save($this->filename);
+
+        $reader = IOFactory::createReader('ODText');
+        $phpWord2 = $reader->load($this->filename);
+        $outputText = [];
+        foreach ($phpWord2->getSections() as $section) {
+            foreach ($section->getElements() as $element) {
+                if (is_object($element) && method_exists($element, 'getText')) {
+                    $outputText[] = $element->getText();
+                }
+            }
+        }
+        self::assertSame($inputText, $outputText);
+    }
+
+    public function testReadNoSections(): void
+    {
+        $dir = 'tests/PhpWordTests/_files/documents';
+        $inputText = ['days', 'monday', 'tuesday'];
+
+        $reader = IOFactory::createReader('ODText');
+        $filename = "$dir/word.2493.nosection.odt";
+        $phpWord2 = $reader->load($filename);
+        $outputText = [];
+        foreach ($phpWord2->getSections() as $section) {
+            foreach ($section->getElements() as $element) {
+                if (is_object($element) && method_exists($element, 'getText')) {
+                    $outputText[] = $element->getText();
+                }
+            }
+        }
+        self::assertSame($inputText, $outputText);
+    }
+}