Skip to content

[Translation] XLIFF 2.0 read support #12853

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 5 commits into from
Closed

[Translation] XLIFF 2.0 read support #12853

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
3e45e57
add xliff-code-2.0 from oasis
xphere Nov 29, 2014
f9de3f3
detect xliff version
xphere Nov 29, 2014
0c5dfd2
extract XliffVersion 1.2
xphere Dec 3, 2014
971b032
extract AbstractXliffVersion
xphere Dec 3, 2014
0971c65
add XliffVersion 2.0
xphere Dec 4, 2014
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
167 changes: 77 additions & 90 deletions src/Symfony/Component/Translation/Loader/XliffFileLoader.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,119 +41,43 @@ public function load($resource, $locale, $domain = 'messages')
throw new NotFoundResourceException(sprintf('File "%s" not found.', $resource));
}

list($xml, $encoding) = $this->parseFile($resource);
$xml->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:1.2');

$catalogue = new MessageCatalogue($locale);
foreach ($xml->xpath('//xliff:trans-unit') as $translation) {
$attributes = $translation->attributes();

if (!(isset($attributes['resname']) || isset($translation->source)) || !isset($translation->target)) {
continue;
}

$source = isset($attributes['resname']) && $attributes['resname'] ? $attributes['resname'] : $translation->source;
// If the xlf file has another encoding specified, try to convert it because
// simple_xml will always return utf-8 encoded values
$target = $this->utf8ToCharset((string) $translation->target, $encoding);

$catalogue->set((string) $source, $target, $domain);
try {
$dom = XmlUtils::loadFile($resource);
$version = $this->getVersion($dom);
} catch (\InvalidArgumentException $e) {
$message = sprintf('Unable to load "%s": %s', $resource, $e->getMessage());

if (isset($translation->note)) {
$notes = array();
foreach ($translation->note as $xmlNote) {
$noteAttributes = $xmlNote->attributes();
$note = array('content' => $this->utf8ToCharset((string) $xmlNote, $encoding));
if (isset($noteAttributes['priority'])) {
$note['priority'] = (int) $noteAttributes['priority'];
}
throw new InvalidResourceException($message, $e->getCode(), $e);
}

if (isset($noteAttributes['from'])) {
$note['from'] = (string) $noteAttributes['from'];
}
$this->validateSchema($dom, $version->getSchema());

$notes[] = $note;
}
$catalogue = new MessageCatalogue($locale);
$version->extract($dom, $catalogue, $domain);

$catalogue->setMetadata((string) $source, array('notes' => $notes), $domain);
}
}
$catalogue->addResource(new FileResource($resource));

return $catalogue;
}

/**
* Convert a UTF8 string to the specified encoding.
*
* @param string $content String to decode
* @param string $encoding Target encoding
*
* @return string
*/
private function utf8ToCharset($content, $encoding = null)
{
if ('UTF-8' !== $encoding && !empty($encoding)) {
if (function_exists('mb_convert_encoding')) {
return mb_convert_encoding($content, $encoding, 'UTF-8');
}

if (function_exists('iconv')) {
return iconv('UTF-8', $encoding, $content);
}

throw new \RuntimeException('No suitable convert encoding function (use UTF-8 as your encoding or install the iconv or mbstring extension).');
}

return $content;
}

/**
* Validates and parses the given file into a SimpleXMLElement.
*
* @param string $file
*
* @throws \RuntimeException
*
* @return \SimpleXMLElement
* @param \DOMDocument $dom
* @param string $schema source of the schema
*
* @throws InvalidResourceException
*/
private function parseFile($file)
private function validateSchema(\DOMDocument $dom, $schema)
{
try {
$dom = XmlUtils::loadFile($file);
} catch (\InvalidArgumentException $e) {
throw new InvalidResourceException(sprintf('Unable to load "%s": %s', $file, $e->getMessage()), $e->getCode(), $e);
}

$internalErrors = libxml_use_internal_errors(true);

$location = str_replace('\\', '/', __DIR__).'/schema/dic/xliff-core/xml.xsd';
$parts = explode('/', $location);
if (0 === stripos($location, 'phar://')) {
$tmpfile = tempnam(sys_get_temp_dir(), 'sf2');
if ($tmpfile) {
copy($location, $tmpfile);
$parts = explode('/', str_replace('\\', '/', $tmpfile));
}
}
$drive = '\\' === DIRECTORY_SEPARATOR ? array_shift($parts).'/' : '';
$location = 'file:///'.$drive.implode('/', array_map('rawurlencode', $parts));

$source = file_get_contents(__DIR__.'/schema/dic/xliff-core/xliff-core-1.2-strict.xsd');
$source = str_replace('http://www.w3.org/2001/xml.xsd', $location, $source);

if (!@$dom->schemaValidateSource($source)) {
if (!@$dom->schemaValidateSource($schema)) {
throw new InvalidResourceException(implode("\n", $this->getXmlErrors($internalErrors)));
}

$dom->normalizeDocument();

libxml_clear_errors();
libxml_use_internal_errors($internalErrors);

return array(simplexml_import_dom($dom), strtoupper($dom->encoding));
}

/**
Expand Down Expand Up @@ -182,4 +106,67 @@ private function getXmlErrors($internalErrors)

return $errors;
}

/**
* Detects xliff version from file.
*
* @param \DOMDocument $dom
*
* @throws \InvalidArgumentException
*
* @return XliffVersion\AbstractXliffVersion
*/
private function getVersion(\DOMDocument $dom)
{
$versionNumber = $this->getVersionNumber($dom);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You don't handle an UnexpectedValueException that might be thrown in getVersionNumber() neither here or in the load() method.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sure, my bad!
This should throw a InvalidArgumentException so we can catch on load method for better message readability.
Easy fix. Thank you!


switch ($versionNumber) {
case '1.2':
return new XliffVersion\XliffVersion12();

case '2.0':
return new XliffVersion\XliffVersion20();
}

throw new \InvalidArgumentException(sprintf(
'No support implemented for loading XLIFF version "%s".',
$versionNumber
));
}

/**
* Gets xliff file version based on the root "version" attribute.
* Defaults to 1.2 for backwards compatibility
*
* @param \DOMDocument $dom
*
* @throws \InvalidArgumentException
*
* @return string
*/
private function getVersionNumber(\DOMDocument $dom)
{
/** @var \DOMNode $xliff */
foreach ($dom->getElementsByTagName('xliff') as $xliff) {
$version = $xliff->attributes->getNamedItem('version');
if ($version) {
return $version->nodeValue;
}

$namespace = $xliff->attributes->getNamedItem('xmlns');
if ($namespace) {
if (substr_compare('urn:oasis:names:tc:xliff:document:', $namespace->nodeValue, 0, 34) !== 0) {
throw new \InvalidArgumentException(sprintf(
'Not a valid XLIFF namespace "%s"',
$namespace
));
}

return substr($namespace, 34);
}
}

// Falls back to v1.2
return '1.2';
}
}
91 changes: 91 additions & 0 deletions src/Symfony/Component/Translation/Loader/XliffVersion/AbstractXliffVersion.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,91 @@
<?php

/*
* This file is part of the Symfony package.
*
* (c) Fabien Potencier <fabien@symfony.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/

namespace Symfony\Component\Translation\Loader\XliffVersion;

use Symfony\Component\Translation\MessageCatalogue;

/**
* Base Xliff version loader class.
*
* @author Berny Cantos <be@rny.cc>
*/
abstract class AbstractXliffVersion
{
/**
* Get validation schema source for this version
*
* @return string
*/
abstract public function getSchema();

/**
* Extract messages and metadata from DOMDocument into a MessageCatalogue
*
* @param \DOMDocument $dom Source to extract messages and metadata
* @param MessageCatalogue $catalogue Catalogue where we'll collect messages and metadata
* @param string $domain The domain
*/
abstract public function extract(\DOMDocument $dom, MessageCatalogue $catalogue, $domain);

/**
* Internally changes the URI of a dependent xsd to be loaded locally
*
* @param string $schemaSource Current content of schema file
* @param string $xmlUri External URI of XML to convert to local
*
* @return string
*/
protected function fixXmlLocation($schemaSource, $xmlUri)
{
$newPath = str_replace('\\', '/', __DIR__).'/../schema/dic/xliff-core/xml.xsd';
$parts = explode('/', $newPath);
if (0 === stripos($newPath, 'phar://')) {
$tmpfile = tempnam(sys_get_temp_dir(), 'sf2');
if ($tmpfile) {
copy($newPath, $tmpfile);
$parts = explode('/', str_replace('\\', '/', $tmpfile));
}
}
$drive = '\\' === DIRECTORY_SEPARATOR ? array_shift($parts).'/' : '';
$newPath = 'file:///'.$drive.implode('/', array_map('rawurlencode', $parts));

return str_replace($xmlUri, $newPath, $schemaSource);
}

/**
* Convert a UTF8 string to the specified encoding
*
* @param string $content String to decode
* @param string $encoding Target encoding
*
* @throws \RuntimeException
* @return string
*/
protected function utf8ToCharset($content, $encoding = null)
{
if (empty($encoding) || 'UTF-8' === $encoding) {
return $content;
}

if (function_exists('mb_convert_encoding')) {
return mb_convert_encoding($content, $encoding, 'UTF-8');
}

if (function_exists('iconv')) {
return iconv('UTF-8', $encoding, $content);
}

throw new \RuntimeException(
'No suitable convert encoding function (use UTF-8 as your encoding or install the iconv or mbstring extension).'
);
}
}
83 changes: 83 additions & 0 deletions src/Symfony/Component/Translation/Loader/XliffVersion/XliffVersion12.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
<?php

/*
* This file is part of the Symfony package.
*
* (c) Fabien Potencier <fabien@symfony.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/

namespace Symfony\Component\Translation\Loader\XliffVersion;

use Symfony\Component\Translation\MessageCatalogue;

/**
* XliffVersion12 loads XLIFF files identified with version 1.2
*
* @author Berny Cantos <be@rny.cc>
*/
class XliffVersion12 extends AbstractXliffVersion
{
/**
* Get validation schema source for this version
*
* @return string
*/
public function getSchema()
{
$source = file_get_contents(__DIR__.'/../schema/dic/xliff-core/xliff-core-1.2-strict.xsd');

return $this->fixXmlLocation($source, 'http://www.w3.org/2001/xml.xsd');
}

/**
* Extract messages and metadata from DOMDocument into a MessageCatalogue
*
* @param \DOMDocument $dom Source to extract messages and metadata
* @param MessageCatalogue $catalogue Catalogue where we'll collect messages and metadata
* @param string $domain The domain
*/
public function extract(\DOMDocument $dom, MessageCatalogue $catalogue, $domain)
{
$xml = simplexml_import_dom($dom);
$encoding = strtoupper($dom->encoding);

$xml->registerXPathNamespace('xliff', 'urn:oasis:names:tc:xliff:document:1.2');

foreach ($xml->xpath('//xliff:trans-unit') as $translation) {
$attributes = $translation->attributes();

if (!(isset($attributes['resname']) || isset($translation->source)) || !isset($translation->target)) {
continue;
}

$source = isset($attributes['resname']) && $attributes['resname'] ? $attributes['resname'] : $translation->source;
// If the xlf file has another encoding specified, try to convert it because
// simple_xml will always return utf-8 encoded values
$target = $this->utf8ToCharset((string) $translation->target, $encoding);

$catalogue->set((string) $source, $target, $domain);

if (isset($translation->note)) {
$notes = array();
foreach ($translation->note as $xmlNote) {
$noteAttributes = $xmlNote->attributes();
$note = array('content' => $this->utf8ToCharset((string) $xmlNote, $encoding));
if (isset($noteAttributes['priority'])) {
$note['priority'] = (int) $noteAttributes['priority'];
}

if (isset($noteAttributes['from'])) {
$note['from'] = (string) $noteAttributes['from'];
}

$notes[] = $note;
}

$catalogue->setMetadata((string) $source, array('notes' => $notes), $domain);
}
}
}
}
Loading