+ * @param Entry $entry Entry to update
+ * @param string $url Url of the content
+ * @param array $content Array with content provided for import with AT LEAST keys title, html, url to skip the fetchContent from the url
+ * @param bool $disableContentUpdate Whether to skip trying to fetch content using Graby
+ */
+ public function updateEntry(Entry $entry, $url, array $content = [], $disableContentUpdate = false)
+ {
+ if (!empty($content['html'])) {
+ $content['html'] = $this->graby->cleanupHtml($content['html'], $url);
+ }
+
+ if ((empty($content) || false === $this->validateContent($content)) && false === $disableContentUpdate) {
+ $fetchedContent = $this->graby->fetchContent($url);
+ $fetchedContent['title'] = $this->sanitizeUTF8Text($fetchedContent['title']);
+
+ // when content is imported, we have information in $content
+ // in case fetching content goes bad, we'll keep the imported information instead of overriding them
+ if (empty($content) || $fetchedContent['html'] !== $this->fetchingErrorMessage) {
+ $content = $fetchedContent;
+ }
+ }
+
+ // be sure to keep the url in case of error
+ // so we'll be able to refetch it in the future
+ $content['url'] = !empty($content['url']) ? $content['url'] : $url;
+
+ $this->stockEntry($entry, $content);
+ }
+
+ /**
+ * Remove invalid UTF-8 characters from the given string in following steps:
+ * - try to interpret the given string as ISO-8859-1, convert it to UTF-8 and return it (if its valid)
+ * - simply remove every invalid UTF-8 character and return the result (https://stackoverflow.com/a/1433665)
+ * @param String $rawText
+ * @return string
+ */
+ private function sanitizeUTF8Text($rawText) {
+ if (mb_check_encoding($rawText, 'utf-8')) {
+ return $rawText; // return because its valid utf-8 text
+ }
+
+ // we assume that $text is encoded in ISO-8859-1 (and not the similar Windows-1252 or other encoding)
+ $convertedText = utf8_encode($rawText);
+ if (mb_check_encoding($convertedText, 'utf-8')) {
+ return $convertedText;
+ }
+
+ // last resort: simply remove invalid UTF-8 character because $rawText can have some every exotic encoding
+ return iconv("UTF-8", "UTF-8//IGNORE", $rawText);
+ }
+
+ /**
+ * Use a Symfony validator to ensure the language is well formatted.
+ *
+ * @param Entry $entry
+ * @param string $value Language to validate and save
+ */
+ public function updateLanguage(Entry $entry, $value)
+ {
+ // some lang are defined as fr-FR, es-ES.
+ // replacing - by _ might increase language support
+ $value = str_replace('-', '_', $value);
+
+ $errors = $this->validator->validate(
+ $value,
+ (new LocaleConstraint())
+ );
+
+ if (0 === \count($errors)) {
+ $entry->setLanguage($value);
+
+ return;
+ }
+
+ $this->logger->warning('Language validation failed. ' . (string) $errors);
+ }
+
+ /**
+ * Use a Symfony validator to ensure the preview picture is a real url.
+ *
+ * @param Entry $entry
+ * @param string $value URL to validate and save
+ */
+ public function updatePreviewPicture(Entry $entry, $value)
+ {
+ $errors = $this->validator->validate(
+ $value,
+ (new UrlConstraint())
+ );
+
+ if (0 === \count($errors)) {
+ $entry->setPreviewPicture($value);
+
+ return;
+ }
+
+ $this->logger->warning('PreviewPicture validation failed. ' . (string) $errors);
+ }
+
+ /**
+ * Update date.
+ *
+ * @param Entry $entry
+ * @param string $value Date to validate and save
+ */
+ public function updatePublishedAt(Entry $entry, $value)
+ {
+ $date = $value;
+
+ // is it a timestamp?
+ if (false !== filter_var($date, FILTER_VALIDATE_INT)) {
+ $date = '@' . $date;
+ }
+
+ try {
+ // is it already a DateTime?
+ // (it's inside the try/catch in case of fail to be parse time string)
+ if (!$date instanceof \DateTime) {
+ $date = new \DateTime($date);
+ }
+
+ $entry->setPublishedAt($date);
+ } catch (\Exception $e) {
+ $this->logger->warning('Error while defining date', ['e' => $e, 'url' => $entry->getUrl(), 'date' => $value]);
+ }
+ }
+
+ /**
+ * Helper to extract and save host from entry url.
+ *
+ * @param Entry $entry
+ */
+ public function setEntryDomainName(Entry $entry)
+ {
+ $domainName = parse_url($entry->getUrl(), PHP_URL_HOST);
+ if (false !== $domainName) {
+ $entry->setDomainName($domainName);
+ }
+ }
+
+ /**
+ * Helper to set a default title using:
+ * - url basename, if applicable
+ * - hostname.
+ *
+ * @param Entry $entry
+ */
+ public function setDefaultEntryTitle(Entry $entry)
+ {
+ $url = parse_url($entry->getUrl());
+ $path = pathinfo($url['path'], PATHINFO_BASENAME);
+
+ if (empty($path)) {
+ $path = $url['host'];
+ }
+
+ $entry->setTitle($path);
+ }
+
+ /**
+ * Stock entry with fetched or imported content.
+ * Will fall back to OpenGraph data if available.