imageone paragraph with special characters: äöüľščťžýáíéäňú©«®™»
+one paragraph with a
break line
one paragraph with html entities: &<>
+and here we have some link break line
+one paragraph with an image
hello world this is simple html
elements
+ * @link http://docs.oasis-open.org/xliff/v1.2/xliff-profile-html/xliff-profile-html-1.2-cd02.html#Elem_preformatted
*/
class TMGMTFileformatXLIFF extends XMLWriter implements TMGMTFileFormatInterface {
@@ -12,6 +52,9 @@ class TMGMTFileformatXLIFF extends XMLWriter implements TMGMTFileFormatInterface
*/
protected $job;
+ protected $importedXML;
+ protected $importedTransUnits;
+
/**
* Adds a job item to the xml export.
*
@@ -44,17 +87,21 @@ class TMGMTFileformatXLIFF extends XMLWriter implements TMGMTFileFormatInterface
*/
protected function addTransUnit($key, $element) {
+ $key_array = tmgmt_ensure_keys_array($key);
+
$this->startElement('trans-unit');
$this->writeAttribute('id', $key);
$this->writeAttribute('resname', $key);
$this->startElement('source');
$this->writeAttribute('xml:lang', $this->job->getTranslator()->mapToRemoteLanguage($this->job->source_language));
- $this->text($element['#text']);
+ $this->writeRaw($this->processForExport($element['#text'], $key_array[0]));
$this->endElement();
$this->startElement('target');
$this->writeAttribute('xml:lang', $this->job->getTranslator()->mapToRemoteLanguage($this->job->target_language));
- $this->text($element['#text']);
+ // @todo - temporary solution, it will be changed in the [#2006786]
+ $target = (!empty($element['#translation']['#text']) ? $element['#translation']['#text'] : $element['#text']);
+ $this->writeRaw($this->processForExport($target, $key_array[0]));
$this->endElement();
if (isset($element['#label'])) {
$this->writeElement('note', $element['#label']);
@@ -124,39 +171,20 @@ class TMGMTFileformatXLIFF extends XMLWriter implements TMGMTFileFormatInterface
* {@inheritdoc}
*/
public function import($imported_file) {
- // It is not possible to load the file directly with simplexml as it gets
- // url encoded due to the temporary://. This is a PHP bug, see
- // https://bugs.php.net/bug.php?id=61469
- $xml_string = file_get_contents($imported_file);
- $xml = simplexml_load_string($xml_string);
-
- // Register the xliff namespace, required for xpath.
- $xml->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:1.2');
-
- $data = array();
- foreach ($xml->xpath('//xliff:trans-unit') as $unit) {
- $data[(string) $unit['id']]['#text'] = (string) $unit->target;
- }
- return tmgmt_unflatten_data($data);
+ $this->getImportedXML($imported_file);
+ return tmgmt_unflatten_data($this->getImportedTargets());
}
/**
- * {@inheritdoc}
+ * Validates imported XLIFF file.
+ *
+ * Checks:
+ * - Job ID.
+ * - Target ans source languages
+ * - Content integrity by
*/
public function validateImport($imported_file) {
- // It is not possible to load the file directly with simplexml as it gets
- // url encoded due to the temporary://. This is a PHP bug, see
- // https://bugs.php.net/bug.php?id=61469
- $xml_string = file_get_contents($imported_file);
- $xml = simplexml_load_string($xml_string);
-
- if (!$xml) {
- return FALSE;
- }
-
- // Register the xliff namespace, required for xpath.
- $xml->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:1.2');
-
+ $xml = $this->getImportedXML($imported_file);
// Check if our phase information is there.
$phase = $xml->xpath("//xliff:phase[@phase-name='extraction']");
if ($phase) {
@@ -181,8 +209,331 @@ class TMGMTFileformatXLIFF extends XMLWriter implements TMGMTFileFormatInterface
return FALSE;
}
+ $targets = $this->getImportedTargets();
+
+ if (empty($targets)) {
+ return FALSE;
+ }
+
+ $reader = new XMLReader();
+ $elements_counts = $job->getSetting('elements_counts');
+
+ foreach ($targets as $id => $target) {
+ $count = 0;
+ $reader->XML('' . $target['#text'] . ' ');
+ while ($reader->read()) {
+ if (in_array($reader->name, array('translation', '#text'))) {
+ continue;
+ }
+ $count++;
+ }
+
+ $keys_array = tmgmt_ensure_keys_array($id);
+ $tjiid = array_shift($keys_array);
+ if (!isset($elements_counts[$tjiid]) || $elements_counts[$tjiid] != $count) {
+ return FALSE;
+ }
+ }
+
// Validation successful.
return $job;
}
+ protected function getImportedXML($imported_file) {
+ if (empty($this->importedXML)) {
+ // It is not possible to load the file directly with simplexml as it gets
+ // url encoded due to the temporary://. This is a PHP bug, see
+ // https://bugs.php.net/bug.php?id=61469
+ $xml_string = file_get_contents($imported_file);
+ $this->importedXML = simplexml_load_string($xml_string);
+ // Register the XLIFF namespace, required for xpath.
+ $this->importedXML->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:1.2');
+ }
+
+ return $this->importedXML;
+ }
+
+ protected function getImportedTargets() {
+ if (empty($this->importedXML)) {
+ return FALSE;
+ }
+
+ if (empty($this->importedTransUnits)) {
+ $reader = new XMLReader();
+ foreach ($this->importedXML->xpath('//xliff:trans-unit') as $unit) {
+ $reader->XML($unit->target->asXML());
+ $reader->read();
+ $this->importedTransUnits[(string) $unit['id']]['#text'] = $this->processForImport($reader->readInnerXML());
+ }
+ }
+
+ return $this->importedTransUnits;
+ }
+
+ /**
+ * Processes trans-unit/target to rebuild back the HTML.
+ *
+ * @param string $translation
+ * Job data array.
+ *
+ * @return string
+ */
+ protected function processForImport($translation) {
+ $reader = new XMLReader();
+ $reader->XML('' . $translation . ' ');
+ $text = '';
+
+ while ($reader->read()) {
+ // If the current element is text append it to the result text.
+ if ($reader->name == '#text' || $reader->name == '#cdata-section') {
+ $text .= $reader->value;
+ }
+ elseif ($reader->name == 'x') {
+ if ($reader->getAttribute('ctype') == 'lb') {
+ $text .= '
';
+ }
+ }
+ elseif ($reader->name == 'ph') {
+ if ($reader->getAttribute('ctype') == 'image') {
+ $text .= 'moveToNextAttribute()) {
+ // @todo - we have to use x-html: prefixes for attributes.
+ if ($reader->name != 'ctype' && $reader->name != 'id') {
+ $text .= " {$reader->name}=\"{$reader->value}\"";
+ }
+ }
+ $text .= ' />';
+ }
+ }
+ }
+ return $text;
+ }
+
+ /**
+ * Helper function to process the source text.
+ *
+ * @param string $source
+ * Job data array.
+ * @param int $tjiid
+ * Job item ID.
+ *
+ * @return string
+ */
+ protected function processForExport($source, $tjiid) {
+ // The reason why we use DOMDocument object here and not just XMLReader
+ // is the DOMDocument's ability to deal with broken HTML.
+ $dom = new DOMDocument();
+ // We need to append the head with encoding so that special characters
+ // are read correctly.
+ $dom->loadHTML("" . $source . '');
+
+ $iterator = new RecursiveIteratorIterator(
+ new RecursiveDOMIterator($dom),
+ RecursiveIteratorIterator::SELF_FIRST);
+
+ $writer = new XMLWriter();
+ $writer->openMemory();
+ $writer->startDocument('1.0', 'UTF-8');
+ $writer->startElement('wrapper');
+
+ $tray = array();
+ $non_pair_tags = array('br', 'img');
+ $elements_count = 0;
+
+ /** @var DOMElement $node */
+ foreach ($iterator as $node) {
+
+ if (in_array($node->nodeName, array('html', 'body', 'head', 'meta'))) {
+ continue;
+ }
+
+ if ($node->nodeType === XML_ELEMENT_NODE) {
+ // Increment the elements count and compose element id.
+ $elements_count++;
+ $id = 'tjiid' . $tjiid . '-' . $elements_count;
+
+ $is_pair_tag = !in_array($node->nodeName, $non_pair_tags);
+
+ if ($is_pair_tag) {
+ $this->writeBPT($writer, $node, $id);
+ }
+ elseif ($node->nodeName == 'img') {
+ $this->writeIMG($writer, $node, $id);
+ }
+ elseif ($node->nodeName == 'br') {
+ $this->writeBR($writer, $node, $id);
+ }
+
+ // Add to tray new element info.
+ $tray[$id] = array(
+ 'name' => $node->nodeName,
+ 'id' => $id,
+ 'value' => $node->nodeValue,
+ 'built_text' => '',
+ 'is_pair_tag' => $is_pair_tag,
+ );
+
+ }
+ // The current node is a text.
+ elseif ($node->nodeName == '#text') {
+ // Add the node value to the text output.
+ $writer->writeCdata($this->toEntities($node->nodeValue));
+ foreach ($tray as &$info) {
+ $info['built_text'] .= $node->nodeValue;
+ }
+ }
+
+ // Reverse so that pair tags are closed in the expected order.
+ $reversed_tray = array_reverse($tray);
+ foreach ($reversed_tray as $_info) {
+ // If the build_text equals to the node value and it is not a pair tag
+ // add the end pair tag markup.
+ if ($_info['value'] == $_info['built_text'] && $_info['is_pair_tag']) {
+ // Count also for the closing elements.
+ $elements_count++;
+ $this->writeEPT($writer, $_info['name'], $_info['id']);
+ // When the end pair tag has been written unset the element info
+ // from the tray.
+ unset($tray[$_info['id']]);
+ }
+ }
+ }
+
+ // Set the elements count for the current item into job settings.
+ $this->job->settings['elements_counts'][$tjiid] = $elements_count;
+ $this->job->save();
+
+ $writer->endElement();
+ // Load the output with XMLReader so that we can easily get the inner xml.
+ $reader = new XMLReader();
+ $reader->XML($writer->outputMemory());
+ $reader->read();
+ return $reader->readInnerXML();
+ }
+
+ /**
+ * Writes br tag.
+ *
+ * @param XMLWriter $writer
+ * Writer that writes the output.
+ * @param DOMElement $node
+ * Current node.
+ * @param $id
+ * Current node id.
+ */
+ protected function writeBR(XMLWriter $writer, DOMElement $node, $id) {
+ $writer->startElement('x');
+ $writer->writeAttribute('id', $id);
+ $writer->writeAttribute('ctype', 'lb');
+ $writer->endElement();
+ }
+
+ /**
+ * Writes beginning pair tag.
+ *
+ * @param XMLWriter $writer
+ * Writer that writes the output.
+ * @param DOMElement $node
+ * Current node.
+ * @param $id
+ * Current node id.
+ */
+ protected function writeBPT(XMLWriter $writer, DOMElement $node, $id) {
+ $beginning_tag = '<' . $node->nodeName;
+ if ($node->hasAttributes()) {
+ $attributes = array();
+ /** @var DOMAttr $attribute */
+ foreach ($node->attributes as $attribute) {
+ $attributes[] = $attribute->name . '="' . $attribute->value . '"';
+ }
+
+ $beginning_tag .= ' '. implode(' ', $attributes);
+ }
+ $beginning_tag .= '>';
+ $writer->startElement('bpt');
+ $writer->writeAttribute('id', $id);
+ $writer->text($beginning_tag);
+ $writer->endElement();
+ }
+
+ /**
+ * Writes ending pair tag.
+ *
+ * @param XMLWriter $writer
+ * Writer that writes the output.
+ * @param string $name
+ * Ending tag name.
+ * @param $id
+ * Current node id.
+ */
+ protected function writeEPT(XMLWriter $writer, $name, $id) {
+ $writer->startElement('ept');
+ $writer->writeAttribute('id', $id);
+ $writer->text('');
+ $writer->endElement();
+ }
+
+ /**
+ * Writes img tag.
+ *
+ * Note that alt and title attributes are not written as sub elements as
+ * Trados studio is not able to deal with two sub elements at one level.
+ *
+ * @param XMLWriter $writer
+ * Writer that writes the output.
+ * @param DOMElement $node
+ * Current node.
+ * @param $id
+ * Current node id.
+ */
+ protected function writeIMG(XMLWriter $writer, DOMElement $node, $id) {
+ $writer->startElement('ph');
+ $writer->writeAttribute('id', $id);
+ $writer->writeAttribute('ctype', 'image');
+ foreach ($node->attributes as $attribute) {
+ // @todo - uncomment when issue with Trados/sub elements fixed.
+ /*
+ if (in_array($attribute->name, array('title', 'alt'))) {
+ continue;
+ }
+ */
+ $writer->writeAttribute($attribute->name, $attribute->value);
+ }
+ /*
+ if ($alt_attribute = $node->getAttribute('alt')) {
+ $writer->startElement('sub');
+ $writer->writeAttribute('id', $id . '-img-alt');
+ $writer->writeAttribute('ctype', 'x-img-alt');
+ $writer->text($alt_attribute);
+ $writer->endElement();
+ $this->elementsCount++;
+ }
+ if ($title_attribute = $node->getAttribute('title')) {
+ $writer->startElement('sub');
+ $writer->writeAttribute('id', $id . '-img-title');
+ $writer->writeAttribute('ctype', 'x-img-title');
+ $writer->text($title_attribute);
+ $writer->endElement();
+ $this->elementsCount++;
+ }
+ */
+ $writer->endElement();
+ }
+
+ /**
+ * Convert critical characters to HTML entities.
+ *
+ * DOMDocument will convert HTML entities to its actual characters. This can
+ * lead into situation when not allowed characters will appear in the content.
+ *
+ * @param string $string
+ * String to escape.
+ *
+ * @return string
+ * Escaped string.
+ */
+ protected function toEntities($string) {
+ return str_replace(array('&', '>', '<'), array('&', '>', '<'), $string);
+ }
+
}
diff --git a/translators/file/tmgmt_file.info b/translators/file/tmgmt_file.info
index 6331f00..786d890 100644
--- a/translators/file/tmgmt_file.info
+++ b/translators/file/tmgmt_file.info
@@ -9,4 +9,5 @@ files[] = tmgmt_file.ui.inc
files[] = tmgmt_file.format.interface.inc
files[] = tmgmt_file.format.xliff.inc
files[] = tmgmt_file.format.html.inc
+files[] = tmgmt_file.recursive_iterator.inc
files[] = tmgmt_file.test
diff --git a/translators/file/tmgmt_file.module b/translators/file/tmgmt_file.module
index a3b6a29..fdadb5c 100644
--- a/translators/file/tmgmt_file.module
+++ b/translators/file/tmgmt_file.module
@@ -56,7 +56,7 @@ function tmgmt_file_import_form_submit($form, &$form_state) {
try {
// Validation successful, start import.
$job->addTranslatedData($controller->import($file->uri));
- $job->addMessage('Sucessfully imported file.');
+ $job->addMessage('Successfully imported file.');
} catch (Exception $e) {
$job->addMessage('File import failed with the following message: @message', array('@message' => $e->getMessage()), 'error');
}
@@ -80,7 +80,7 @@ function tmgmt_file_import_form_submit($form, &$form_state) {
* @param $plugin
* (Optional) Name of a plugin/extension.
*
- * @return
+ * @return array
* If a plugin name is provided, information about that plugin, an array of
* plugin information otherwise. The information of each plugin consists of
* the label and plugin controller class, keyed by the plugin name which is
@@ -100,6 +100,9 @@ function tmgmt_file_format_plugin_labels() {
/**
* Returns the file format plugin controller.
*
+ * @param $plugin
+ * (Optional) Name of a plugin/extension.
+ *
* @return TMGMTFileFormatInterface
* Either a specific file format plugin controller instance or an array of
* available controllers.
diff --git a/translators/file/tmgmt_file.recursive_iterator.inc b/translators/file/tmgmt_file.recursive_iterator.inc
new file mode 100644
index 0000000..36f2f83
--- /dev/null
+++ b/translators/file/tmgmt_file.recursive_iterator.inc
@@ -0,0 +1,99 @@
+position = 0;
+ $this->nodeList = $domNode->childNodes;
+ }
+
+ /**
+ * Returns the current DOMNode.
+ *
+ * @return DOMNode
+ * Current DOMNode object.
+ */
+ public function current() {
+ return $this->nodeList->item($this->position);
+ }
+
+ /**
+ * Returns an iterator for the current iterator entry.
+ *
+ * @return RecursiveDOMIterator
+ * Iterator with children elements.
+ */
+ public function getChildren() {
+ return new self($this->current());
+ }
+
+ /**
+ * Checks if current element has children.
+ *
+ * @return bool
+ * Has children.
+ */
+ public function hasChildren() {
+ return $this->current()->hasChildNodes();
+ }
+
+ /**
+ * Returns the current position.
+ *
+ * @return int
+ * Current position
+ */
+ public function key() {
+ return $this->position;
+ }
+
+ /**
+ * Moves the current position to the next element.
+ */
+ public function next() {
+ $this->position++;
+ }
+
+ /**
+ * Rewind the Iterator to the first element.
+ */
+ public function rewind() {
+ $this->position = 0;
+ }
+
+ /**
+ * Checks if current position is valid.
+ *
+ * @return bool
+ * Is valid.
+ */
+ public function valid() {
+ return $this->position < $this->nodeList->length;
+ }
+}
diff --git a/translators/file/tmgmt_file.test b/translators/file/tmgmt_file.test
index 15a6f47..814693d 100644
--- a/translators/file/tmgmt_file.test
+++ b/translators/file/tmgmt_file.test
@@ -25,6 +25,143 @@ class TMGMTFileTestCase extends TMGMTBaseTestCase {
}
/**
+ * Test the content processing for XLIFF export and import.
+ */
+ function testXLIFFTextProcessing() {
+ $translator = $this->createTranslator();
+ $translator->plugin = 'file';
+ $translator->settings = array(
+ 'export_format' => 'xlf',
+ );
+ $translator->save();
+
+ // Create the reader instance, it will be used through the tests.
+ $reader = new XMLReader();
+ $xliff_elements = array('bpt', 'ept', 'ph', 'x', '#text', '#cdata-section', 'content');
+
+ // ==== First test the whole cycle ==== //
+ $job = $this->createJob();
+ $job->translator = $translator->name;
+ $job->addItem('test_html_source', 'test', '1');
+
+ // Requesting translation will mask the html.
+ $job->requestTranslation();
+ $content = $this->getTransUnitsContent($job);
+ // Test that the exported trans unit contains only xliff elements.
+ $reader->XML('' . $content[0]['source'] . ' ');
+ while ($reader->read()) {
+ if (!in_array($reader->name, $xliff_elements)) {
+ $this->fail(t('The source contains unexpected element %element', array('%element' => $reader->name)));
+ }
+ }
+ $reader->XML('' . $content[0]['target'] . ' ');
+ while ($reader->read()) {
+ if (!in_array($reader->name, $xliff_elements)) {
+ $this->fail(t('The target contains unexpected element %element', array('%element' => $reader->name)));
+ }
+ }
+
+ // Import the file, make sure all the html has been revealed and no xliff
+ // elements are present in the job translation.
+ $messages = $job->getMessages();
+ $message = reset($messages);
+ $download_url = $message->variables['!link'];
+ $xml = simplexml_load_file($download_url);
+ $translated_file = 'public://tmgmt_file/translated.xlf';
+ $xml->asXML($translated_file);
+ $uri = $job->uri();
+ $edit = array(
+ 'files[file]' => $translated_file,
+ );
+ $this->drupalPost($uri['path'] . '/manage', $edit, t('Import'));
+ // Reset caches and reload job.
+ entity_get_controller('tmgmt_job')->resetCache();
+ entity_get_controller('tmgmt_job_item')->resetCache();
+ $job = tmgmt_job_load($job->tjid);
+
+ // Do the comparison of the translation text and the source. I must be the
+ // same as there was change done to the translation.
+ $item_data = $job->getData(array(1, 'dummy', 'deep_nesting'));
+ $source_text = trim(file_get_contents(drupal_get_path('module', 'tmgmt') . '/tests/testing_html/sample.html'));
+ $this->assertEqual(trim($item_data[1]['#text']), trim($source_text));
+ $job_items = $job->getItems();
+ /** @var TMGMTJobItem $job_item */
+ $job_item = array_shift($job_items);
+ // Job item must be in review.
+ $this->assertTrue($job_item->isNeedsReview());
+
+ // ==== Test integrity check ==== //
+ $job = $this->createJob();
+ $job->translator = $translator->name;
+ $job->addItem('test_html_source', 'test', '1');
+ $job->requestTranslation();
+
+ $messages = $job->getMessages();
+ $message = reset($messages);
+ // Get the xml content and remove the element representing
. This will
+ // result in different element counts in the source and target and should
+ // trigger an error and not import the translation.
+ $content = str_replace('