Changeset 33757 for main/trunk/gli/src/org/greenstone/gatherer/metadata/DocXMLFile.java

Timestamp:

2019-12-07T01:40:13+13:00 (4 years ago)

Author:

ak19

Message:

1. Windows bugfix for getting exMeta to be loaded into GLI where there are subdirs involved in the Gather pane, or there are non-ASCII filenames, or the file rename method is set to base64. 2. Bugfix for Linux and Windows: Using Base64 to rename files was still a problem despite the previous commit (which was supposed to have fixed all GLI exMeta loading issues on Linux) in the special case where a subfolder was pure ASCII. The perl code wouldn't base64 encode such subdirs. However, GLI won't know which part of a relative file path to decode based on the file rename method used and which parts are not to be decoded. So GLI uniformly decoded them, and ASCII named subfolders that were not base64 encoded (but contained files that were to be renamed with base64) got base64 decoded into garbage, so that exMeta still did not get attached. 3. This commit contains debug stmts.

File:

• main/trunk/gli/src/org/greenstone/gatherer/metadata/DocXMLFile.java (modified) (7 diffs)

main/trunk/gli/src/org/greenstone/gatherer/metadata/DocXMLFile.java

@@ -34 +34 @@
 import org.greenstone.gatherer.util.Utility;
 
-import org.apache.commons.codec.binary.Base64;
-
-//import org.greenstone.gatherer.feedback.Base64;
+//import org.greenstone.gatherer.feedback.Base64; // decode() from Base64 didn't work
+import org.apache.commons.codec.binary.Base64; // decoding from Base64 works
 
 /** This class represents one doc.xml file */
@@ -58 +57 @@
     }
 
+    /**
+     * Checks if various versions of the file object's filename, denoted relatively by file_relative_path,
+     * occur in the source_file_name_to_description_elements_mapping map
+    */
+    private ArrayList findSourceFileMapKeyMatch(File file, String file_relative_path) {
+        ArrayList description_elements_list = null;
+                
+        System.err.println("Looking for key " + file_relative_path);
+        description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
+        if(description_elements_list != null) {
+            System.err.println("   Found key matching REGULAR filepath: " + file_relative_path);
+            return description_elements_list;           
+        }
+        else if(!Utility.isWindows()) { // couldn't find a matching key, we're done
+            System.err.println("Unable to find meta for regular file path form " + file_relative_path);
+            return null;
+        }
+        
+        // Now we can try windows short filename as map key
+        
+        String win_short_file_relative_path = "";
+        try{
+            win_short_file_relative_path = Utility.getWindowsShortFileName(file.getAbsolutePath());             
+            //System.err.println("@@@ Searching for short file name: " + win_short_file_relative_path);
+        } catch(Exception e) { // we're done trying to find a matching key
+            System.err.println("Failed to convert to windows short file name: " + win_short_file_relative_path);            
+            return null;
+        }
+        
+        // Got a windows short file name, lop off import folder again
+        int import_index = win_short_file_relative_path.indexOf("import");
+        if (import_index != -1) {
+            win_short_file_relative_path = win_short_file_relative_path.substring(import_index + "import".length() + 1);
+        }
+            
+        System.err.println("### Looking for Windows short file name |" + win_short_file_relative_path +  "| in map of sourcefilenames to doc.xml's ex meta.");
+        description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(win_short_file_relative_path);
+        if (description_elements_list != null) {
+            System.err.println("   Found key matching FULL win shortfile path: " + win_short_file_relative_path);
+            return description_elements_list; // found
+        }
+        
+        // else, check whether a map key is matched by any REMAINING combination of windows shortfile path and regular path:
+        // - windows shortfilename's rel-dir-path with regular tailname
+        // - and regular rel-dir-path with windows shortfilename's tailname
+                
+        String shortFileTailName = win_short_file_relative_path;
+        String shortFileRelDirPath = "";
+        int lastSep = win_short_file_relative_path.lastIndexOf(File.separator);
+        if(lastSep != -1) {         
+            shortFileTailName = win_short_file_relative_path.substring(lastSep+1);
+            shortFileRelDirPath = win_short_file_relative_path.substring(0, lastSep+1); // include the slash
+        }
+        
+        String fileTailName = file_relative_path;
+        String fileRelDirPath = "";
+        lastSep = file_relative_path.lastIndexOf(File.separator);
+        if(lastSep != -1) {         
+            fileTailName = file_relative_path.substring(lastSep+1);
+            fileRelDirPath = file_relative_path.substring(0, lastSep+1); // include the slash
+        }
+        
+        String path = shortFileRelDirPath + fileTailName;
+        System.err.println("### Looking for Windows short file name |" + path +  "| in map of sourcefilenames to doc.xml's ex meta.");
+        description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(path);
+        
+        if(description_elements_list != null) {
+            System.err.println("   Found key matching MIX of win shortfile path and regular path: " + path);
+            return description_elements_list; // found
+        }
+
+        // try the other combination
+        path = fileRelDirPath + shortFileTailName;
+        System.err.println("### Looking for Windows short file name |" + path +  "| in map of sourcefilenames to doc.xml's ex meta.");
+        description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(path);
+        
+        if(description_elements_list != null) {
+            System.err.println("   Found key matching MIX of regular path and win shortfile path: " + path);
+            return description_elements_list; // found
+        }       
+        
+        return description_elements_list;
+    }
+    
 
     public ArrayList getMetadataExtractedFromFile(File file)
@@ -70 +153 @@
     }
 
-    ///for (Object relFilename : source_file_name_to_description_elements_mapping.keySet()) {
-    ///    System.err.println("@@@ relFilename: " + relFilename);
-    ///}
+    for (Object relFilename : source_file_name_to_description_elements_mapping.keySet()) {
+        System.err.println("\n@@@ relFilename: " + relFilename);
+    }
     
     // Check whether this file (i.e. doc.xml or docmets.xml on inheritance) file contains extracted metadata for the specified file
-    ArrayList description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
+    //ArrayList description_elements_list = (ArrayList) source_file_name_to_description_elements_mapping.get(file_relative_path);
+    ArrayList description_elements_list = findSourceFileMapKeyMatch(file, file_relative_path);
     if (description_elements_list == null) {
-        // ...it doesn't
-        return metadata_values;
+            // ...it doesn't
+            System.err.println("Unable to find meta for (regular file path form) " + file_relative_path);
+            if(Utility.isWindows()) {
+                System.err.println("    Or for windows shortFile path form, or for combinations with regular file path form");
+            }
+            return metadata_values; // we're done
     }
 
@@ -275 +363 @@
             }
             
-            ///System.err.println("@@@@ Found gsdlsourcefilename: " + gsdlsourcefilename_value);
+            System.err.println("@@@@ Found gsdlsourcefilename: " + gsdlsourcefilename_value);
             // Remember this for quick access later
             if (source_file_name_to_description_elements_mapping.get(gsdlsourcefilename_value) == null) {
@@ -326 +414 @@
         gsdlsourcefilename_value = decodeSourceFilename(gsdlsourcefilename_value, fileRenameMethod, is_unix_path);
         source_file_name_to_description_elements_mapping.put(gsdlsourcefilename_value, value_list);
-        }
+        }       
     }
     catch (FileNotFoundException exception) {
@@ -379 +467 @@
         // General info: https://stackoverflow.com/questions/43089541/difference-between-basic-and-url-base64-encoding-in-java-8
         byte[] bytes = Base64.decodeBase64(importFilePathParts[i].getBytes());
-        ///System.err.println("Got base64 string: " + importFilePathParts[i]);
-        ///System.err.println("Decoded from base64 to bytes: " + bytes);
+        System.err.println("Got base64 string: " + importFilePathParts[i]);
+        System.err.println("Decoded from base64 to bytes: " + new String(bytes, System.getProperty("file.encoding")));
         // Using system file.encoding to interpret the resulting bytestring as a String,
         // just as we always did with URL decoding method
@@ -397 +485 @@
     decoded_gsdlsourcefilename += file_ext;
     
-    ///System.err.println("@@@@ decoded_gsdlsourcefilename: " + Utility.debugUnicodeString(decoded_gsdlsourcefilename));
+    System.err.println("@@@@ decoded_gsdlsourcefilename: " + Utility.debugUnicodeString(decoded_gsdlsourcefilename));
 
     return decoded_gsdlsourcefilename;