Changeset 33911

Timestamp:

2020-02-12T19:12:42+13:00 (4 years ago)

Author:

ak19

Message:

Correct commit message for previous and current commit: 1. After refactoring MongoDBAccess class into additional subclass MongoDBQueryer, I split the import statements accordingly too. 2. Renamed WebPageURLsListing.java class to SummaryTool.java

Location:

other-projects/maori-lang-detection/src/org/greenstone/atea

Files:

• MongoDBAccess.java (modified) (2 diffs)
• SummaryTool.java (moved) (moved from other-projects/maori-lang-detection/src/org/greenstone/atea/WebPageURLsListing.java ) (4 diffs)

other-projects/maori-lang-detection/src/org/greenstone/atea/MongoDBAccess.java

@@ -10 +10 @@
 import com.mongodb.MongoClientOptions;
 
-import com.mongodb.Block;
-
-import org.bson.BsonArray;
-import org.bson.BsonString;
-import org.bson.BsonValue;
 import org.bson.Document;
-import org.bson.conversions.Bson;
-import org.bson.json.JsonMode;
-import org.bson.json.JsonWriterSettings;
-    
-import com.mongodb.util.JSON;
-//import com.mongodb.DBObject;
-
-
-import com.google.gson.*; // for pretty printing
-
-import java.io.BufferedReader;
-import java.io.BufferedWriter;
-import java.io.File;
-import java.io.FileReader;
-import java.io.FileWriter;
-import java.io.IOException;
-import java.io.UncheckedIOException;
-import java.io.Writer;
-
-import java.util.Arrays;
-import java.util.ArrayList;
-import java.util.List;
+
 import java.util.Properties;
-import java.util.TreeSet;
-import java.util.regex.Pattern;
 
 import org.apache.log4j.Logger;
@@ -47 +19 @@
 import dev.morphia.*;
 
-import org.apache.commons.csv.*;
 
 /** 

other-projects/maori-lang-detection/src/org/greenstone/atea/SummaryTool.java

@@ -16 +16 @@
  * TO COMPILE:
  *    maori-lang-detection/src$
- *       javac -cp ".:../conf:../lib/*" org/greenstone/atea/WebPageURLsListing.java
+ *       javac -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool.java
  *
  * TO RUN:
  *    maori-lang-detection/src$
- *       java -cp ".:../conf:../lib/*" org/greenstone/atea/WebPageURLsListing
+ *       java -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool
  * OR:
- *       java -cp ".:../conf:../lib/*" org/greenstone/atea/WebPageURLsListing ../mongodb-data/domainsNZ_IsMRI.txt
+ *       java -cp ".:../conf:../lib/*" org/greenstone/atea/SummaryTool ../mongodb-data/domainsNZ_IsMRI.txt
  *
 */
-public class WebPageURLsListing {
-    static Logger logger = Logger.getLogger(org.greenstone.atea.WebPageURLsListing.class.getName());
+public class SummaryTool {
+    static Logger logger = Logger.getLogger(org.greenstone.atea.SummaryTool.class.getName());
     static private final long FIXED_SEED = 1000;
     
@@ -49 +49 @@
     
     
-    public WebPageURLsListing(MongoDBQueryer mongodbQueryer, File outFolder)
+    public SummaryTool(MongoDBQueryer mongodbQueryer, File outFolder)
     {
     this.mongodbQueryer = mongodbQueryer;
@@ -312 +312 @@
     
     public static void printUsage() {
-    System.err.println("Usage: WebPageURLsListing [domains.txt]");
+    System.err.println("Usage: SummaryTool [domains.txt]");
     }
 
@@ -340 +340 @@
         // output files will be stored in mongodb-data-auto
         File outFolder = new File("../mongodb-data-auto/").getAbsoluteFile();
-        WebPageURLsListing listing = new WebPageURLsListing(mongodb, outFolder);
+        SummaryTool listing = new SummaryTool(mongodb, outFolder);
 
         System.out.println("*************************************");
         
         
-        if(args.length >= 1) {
+        if(args.length >= 1) { // web page urls listing mode
         File domainsFile = new File(args[0]);
         if(!domainsFile.exists()) {