From ed42df627454dbeeff96f309aa1f65d290848cb0 Mon Sep 17 00:00:00 2001
From: Masahiro Minami <masahiro@minamirnd.work>
Date: Sun, 13 Mar 2022 23:24:10 +0900
Subject: [PATCH 1/2] Add XML, Draw.IO support

Signed-off-by: Masahiro Minami <masahiro@minamirnd.work>
---
 lib/Service/FilesService.php | 85 +++++++++++++++++++++++++++++++++---
 1 file changed, 80 insertions(+), 5 deletions(-)

diff --git a/lib/Service/FilesService.php b/lib/Service/FilesService.php
index a95a2bd..7e50acb 100644
--- a/lib/Service/FilesService.php
+++ b/lib/Service/FilesService.php
@@ -898,6 +898,18 @@ private function parseMimeTypeText(string $mimeType, string $extension, string &
 			throw new KnownFileMimeTypeException();
 		}
 
+		// 20220219 Parse XML files as TEXT files
+		if (substr($mimeType, 0, 15) === 'application/xml') {
+			$parsed = self::MIMETYPE_TEXT;
+			throw new KnownFileMimeTypeException();
+		}
+
+		// 20220219 Parse .drawio file
+		if ($extension  === 'drawio') {
+			$parsed = self::MIMETYPE_TEXT;
+			throw new KnownFileMimeTypeException();
+		}
+
 		$textMimes = [
 			'application/epub+zip'
 		];
@@ -1039,14 +1051,77 @@ private function extractContentFromFilePDF(FilesDocument $document, File $file)
 			return;
 		}
 
-		try {
-			$document->setContent(
-				base64_encode($file->getContent()), IIndexDocument::ENCODED_BASE64
-			);
-		} catch (NotPermittedException | LockedException $e) {
+		// 20220219 Inflate drawio file
+		if ( $file->getExtension() === 'drawio') {
+			$content = $file->getContent();
+
+			try {
+				$xml = simplexml_load_string($content);
+
+				// Initialize $content
+				$content = '';
+
+				foreach ($xml->diagram as $child) {
+				    $deflated_content = (string)$child;
+				    $base64decoded = base64_decode($deflated_content);
+				    $urlencoded_content = gzinflate($base64decoded);
+				    $urldecoded_content = urldecode($urlencoded_content);
+
+				    // Remove image tag
+				    $diagram_str = preg_replace('/style=\"shape=image[^"]*\"/', '', $urldecoded_content);
+
+				    // Construct XML
+				    $diagram_xml = simplexml_load_string($diagram_str);
+				    $content = $content . ' ' . $this->readDrawioXmlValue($diagram_xml);
+				}
+
+			} catch (\Throwable $t) {
+			}
+
+			try {
+				$document->setContent(
+					// 20220219 Pass content of inflated drawio graph xml
+					base64_encode($content), IIndexDocument::ENCODED_BASE64
+				);
+			} catch (NotPermittedException | LockedException $e) {
+			}
+		} else {
+			try {
+				$document->setContent(
+					base64_encode($file->getContent()), IIndexDocument::ENCODED_BASE64
+				);
+			} catch (NotPermittedException | LockedException $e) {
+			}
 		}
 	}
 
+	// 20220220 Read Draw.io XML elements and return a space separated
+	// strings, stripped of HTML tags, to be indexed.
+	/**
+	 * @param SimpleXMLElement $element
+	 *
+	 * @return string
+	 */
+	private function readDrawioXmlValue(\SimpleXMLElement $element) {
+		$str = '';
+		if( $element['value'] != null && trim(strval($element['value'])) !== '') {
+			$str = $str . " " .  trim(strval($element['value']));
+		}
+		if( $element != null && trim(strval($element)) !== '') {
+			$str = $str . " " . trim(strval($element));
+		}
+
+		try {
+			foreach ($element->children() as $child) {
+				$str = $str . " " . $this->readDrawioXmlValue($child);
+			}
+		} finally {
+		}
+
+		// Strip HTML tags
+		$str_without_tags = preg_replace('/<[^>]*>/', ' ', $str);
+		return $str_without_tags;
+	}
 
 	/**
 	 * @param FilesDocument $document

From a6a465b88667690d1a17db5ef05a2eb69e845c08 Mon Sep 17 00:00:00 2001
From: Masahiro Minami <masahiro@minamirnd.work>
Date: Sat, 14 May 2022 10:30:50 +0900
Subject: [PATCH 2/2] Added test files for xml and drawio support.

Signed-off-by: Masahiro Minami <masahiro@minamirnd.work>
---
 test.drawio | 1 +
 test.xml    | 4 ++++
 2 files changed, 5 insertions(+)
 create mode 100644 test.drawio
 create mode 100644 test.xml

diff --git a/test.drawio b/test.drawio
new file mode 100644
index 0000000..67dda3f
--- /dev/null
+++ b/test.drawio
@@ -0,0 +1 @@
+<mxfile host="embed.diagrams.net" modified="2022-05-14T01:26:13.931Z" agent="5.0 (X11)" etag="mcFH4BbWC4NME3q3iiyV" version="18.0.3" type="embed" pages="2"><diagram id="bHMW1FkPiVIQ63f7flve" name="Page-1">jZLBToQwEIafhqMJ0Mh6FtfdiydMPDftSJsUStphAZ/eItMFJCaemPnm73T6Dwkrm/HieKferAST5KkcE/aS5HlWFHn4zGRayImdFlA7LUm0gkp/AcGUaK8l+J0QrTWouz0Utm1B4I5x5+ywl31as7+14zUcQCW4OdIPLVEt9OkxXfkVdK3izVlKlYZHMQGvuLTDBrFzwkpnLS5RM5ZgZvOiL8u51z+q98EctPifA7SIGzc9ve0dPD5UvRDgPbXzOMWXO9u3EubDacKeB6URqo6LuTqEXQemsDEhy0J4HIYa3sAhjBtEw13ANoBuChKqsoKMmn7lw+p7Fs1UG8+jjtOq63vr1Y0QkCExXY3/qW1+X3b+Bg==</diagram><diagram id="nxyU9POgWWsITp6xB2Gz" name="Page-2">jZJNb4MwDIZ/DcdKQDS6XUdpd1h3odLOEfFIpkBQMAX66xeGU0DVpJ1wHn/y2gFLq+FkeSPPRoAO4lAMATsEcRwlSew+Exlnsmf7GZRWCQpaQK5uQDAk2ikB7SYQjdGomi0sTF1DgRvGrTX9NuzL6G3XhpfwAPKC60f6qQTKmT4/hQt/A1VK3zkKyVNxH0yglVyYfoVYFrDUGoOzVQ0p6Ek8r8ucd/zDex/MQo3/SagP7IKnl/fzR6ay/Ijft+S6oypXrjv64Qu0uMu7ooC2pd21OHo5rOlqAVPFMGCvvVQIecOLydu7A3BMYqXdK3Lm44S+HViEYYVo4hOYCtCOLoS8LCH16HziPb37ZRmRV1iuFuHzOO2/vJdeJHIGqeSfyzZ+faubZtkP</diagram></mxfile>
\ No newline at end of file
diff --git a/test.xml b/test.xml
new file mode 100644
index 0000000..33f9c06
--- /dev/null
+++ b/test.xml
@@ -0,0 +1,4 @@
+<test>
+<test-key>Test-key-success</test-key>
+<test-value>Test-value-success</test-value>
+</test>
\ No newline at end of file