Ticket #290: omindex.diff

xapian-applications/omega/omindex.cc

@@ -178 +178 @@
     } catch (ReadError) {
         // It's probably best to index the document even if pdfinfo fails.
     }
+
+    } else if (startswith(mimetype, "application/vnd.openxmlformats-officedocument.wordprocessingml."))
+    {
+    // Inspired by http://mjr.towers.org.uk/comp/sxw2text
+    string safefile = shell_protect(file);
+    string cmd = "unzip -p " + safefile + " word/document.xml";
+    try {
+        XmlParser xmlparser;
+        xmlparser.parse_html(stdout_to_string(cmd));
+        dump = xmlparser.dump;
+    } catch (ReadError) {
+        cout << "\"" << cmd << "\" failed - skipping\n";
+        return;
+    }
+
+    } else if (startswith(mimetype, "application/vnd.openxmlformats-officedocument.spreadsheetml."))
+    {
+    // Inspired by http://mjr.towers.org.uk/comp/sxw2text
+    string safefile = shell_protect(file);
+    string cmd = "unzip -p " + safefile + " xl/sharedStrings.xml";
+    try {
+        XmlParser xmlparser;
+        xmlparser.parse_html(stdout_to_string(cmd));
+        dump = xmlparser.dump;
+    } catch (ReadError) {
+        cout << "\"" << cmd << "\" failed - skipping\n";
+        return;
+    }
+
 }
 
 static void
@@ -718 +747 @@
     mime_map["sxg"] = "application/vnd.sun.xml.writer.global";
     mime_map["stw"] = "application/vnd.sun.xml.writer.template";
     // Some other word processor formats:
+    mime_map["docx"] = "application/vnd.openxmlformats-officedocument.wordprocessingml.document"; //Word 2007
     mime_map["doc"] = "application/msword";
     mime_map["dot"] = "application/msword"; // Word template
     mime_map["wpd"] = "application/vnd.wordperfect";
@@ -727 +757 @@
     mime_map["zabw"] = "application/x-abiword-compressed"; // AbiWord compressed
     mime_map["rtf"] = "text/rtf";
     // Other MS formats:
+    mime_map["xlsx"] = "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"; //Excel 2007
     mime_map["xls"] = "application/vnd.ms-excel";
     mime_map["xlb"] = "application/vnd.ms-excel";
     mime_map["xlt"] = "application/vnd.ms-excel"; // Excel template