Mercurial > hg > fulltextIndexer

comparison src/de/mpiwg/dwinter/fulltextIndexer/harvester/CLI/OCRHarvesterCLIMD.java @ 0:dc7622afcfea default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
initial
author dwinter
date Wed, 03 Nov 2010 12:33:16 +0100
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:dc7622afcfea
1 /* */ package de.mpiwg.dwinter.fulltextIndexer.harvester.CLI;
2 /* */
3 /* */ import de.mpiwg.dwinter.fulltextIndexer.harvester.OCRHarvesterThreaded;
4 /* */ import java.io.File;
5 /* */ import java.io.IOException;
6 /* */ import java.io.PrintStream;
7 /* */ import org.apache.lucene.index.CorruptIndexException;
8 /* */ import org.apache.lucene.store.LockObtainFailedException;
9 /* */
10 /* */ public class OCRHarvesterCLIMD
11 /* */ {
12 /* */ public static void main(String[] args)
13 /* */ throws CorruptIndexException, LockObtainFailedException, IOException, InterruptedException
14 /* */ {
15 /* 20 */ String usage = "java de.mpiwg.itgroup.fulltext.harvester.OCRHarvesterCLIMD <root_directory> <index_dir> <mdproviderUrl> --lang=<language>\njava de.mpiwg.itgroup.fulltext.harvester.OCRHarvesterCLIMD <root_directory> <index_dir> <mdproviderUrl>--langfile=<languageFile>\njava de.mpiwg.itgroup.fulltext.harvester.OCRHarvesterCLIMD <root_directory> <index_dir> <mdproviderUrl>";
16 /* */
17 /* 23 */ if ((args.length != 4) && (args.length != 3)) {
18 /* 24 */ System.err.println("Usage: " + usage);
19 /* 25 */ System.exit(1);
20 /* */ }
21 /* */
22 /* 33 */ File docDir = new File(args[0]);
23 /* 34 */ if ((!docDir.exists()) || (!docDir.canRead())) {
24 /* 35 */ System.out.println("Document directory '" + docDir.getAbsolutePath() + "' does not exist or is not readable, please check the path");
25 /* 36 */ System.exit(1);
26 /* */ }
27 /* */
28 /* 39 */ File index_dir = new File(args[1]);
29 /* */
30 /* 41 */ if (args.length == 3) {
31 /* 42 */ doTheHarvest(docDir, index_dir, args[2], null);
32 /* */ }
33 /* */ else {
34 /* 45 */ String[] options = args[3].split("=");
35 /* 46 */ if (options.length != 3) {
36 /* 47 */ System.err.println("wrong options:" + args[3]);
37 /* 48 */ System.exit(1);
38 /* */ }
39 /* */
40 /* 51 */ if (options[0].equals("--lang")) {
41 /* 52 */ doTheHarvestLanguage(docDir, index_dir, args[2], options[1]);
42 /* 53 */ } else if (options[0].equals("--langfile")) {
43 /* 54 */ doTheHarvest(docDir, index_dir, args[2], options[1]);
44 /* */ } else {
45 /* 56 */ System.err.println("wrong options:" + options[0]);
46 /* 57 */ System.exit(1);
47 /* */ }
48 /* */ }
49 /* */ }
50 /* */
51 /* */ protected static void doTheHarvestLanguage(File docDir, File index_dir, String mdProviderUrl, String lang)
52 /* */ throws CorruptIndexException, LockObtainFailedException, IOException, InterruptedException
53 /* */ {
54 /* 65 */ OCRHarvesterThreaded harvester = new OCRHarvesterThreaded(docDir, index_dir, null, mdProviderUrl, lang);
55 /* 66 */ harvester.harvestFolder();
56 /* */ }
57 /* */
58 /* */ protected static void doTheHarvest(File docDir, File index_dir, String mdProviderUrl, String languageFile)
59 /* */ throws CorruptIndexException, LockObtainFailedException, IOException, InterruptedException
60 /* */ {
61 /* 72 */ OCRHarvesterThreaded harvester = new OCRHarvesterThreaded(docDir, index_dir, languageFile, mdProviderUrl, null);
62 /* 73 */ harvester.harvestFolder();
63 /* */ }
64 /* */ }
65
66 /* Location: /private/tmp/fulltextIndexer.jar
67 * Qualified Name: de.mpiwg.dwinter.fulltextIndexer.harvester.CLI.OCRHarvesterCLIMD
68 * JD-Core Version: 0.5.4
69 */