Package org.archive.wayback.resourceindex.cdx.format

Examples of org.archive.wayback.resourceindex.cdx.format.CDXFormat

395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
   * @param args
   */
  public static void main(String[] args) {
//    String cdxSpec = CDXFormatIndex.CDX_HEADER_MAGIC;
    String cdxSpec = " CDX N b a m s k r V g";
    CDXFormat format = null;
    BlockLoader blockLoader = new GenericBlockLoader();
    try {
      format = new CDXFormat(cdxSpec);
    } catch (CDXFormatException e1) {
      e1.printStackTrace();
      System.exit(1);
    }
    ZiplinesSearchResultSource zl = new ZiplinesSearchResultSource(format);
    PrintWriter pw = new PrintWriter(System.out);
    int idx;
    boolean blockDump = false;
    for(idx = 0; idx < args.length; idx++) {
      if(args[idx].equals("-format")) {
        idx++;
        if(idx >= args.length) {
          USAGE();
        }
        try {
          zl.setFormat(new CDXFormat(args[idx]));
        } catch (CDXFormatException e1) {
          e1.printStackTrace();
          System.exit(1);
        }
      } else if(args[idx].equals("-flexFormat")) {
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
 
  protected CloseableIterator<CaptureSearchResult> adaptIterator(Iterator<String> itr)
  throws IOException {
   
    long nowMod = file.lastModified();
    CDXFormat cdx = format;
    if(cdx == null) {
      if(nowMod > lastMod) {
        try {
          // BUGBUG: I don't think java will let us do much better than
          // this... No way to stat() a filehandle, right?
          FileInputStream fis = new FileInputStream(file);
          InputStreamReader isr = new InputStreamReader(fis,ByteOp.UTF8);
          BufferedReader fr = new BufferedReader(isr);
          cdx = new CDXFormat(fr.readLine());
          lastMod = nowMod;
          fr.close();
        } catch (CDXFormatException e) {
          lastMod = -1;
          try {
            cdx = new CDXFormat(CDX_HEADER_MAGIC);
          } catch (CDXFormatException e1) {
            throw new IOException(e1.getMessage());
          }
        }
      }
View Full Code Here
207
208
209
210
211
212
213
214
215
216
217
    worker.canonicalizer = canonicalizer;
    worker.interval = 0;
    worker.init();
    try {
      CloseableIterator<CaptureSearchResult> itr = worker.indexFile(path);
      CDXFormat cdxFormat = new CDXFormat(cdxSpec);
      Iterator<String> lines =
        SearchResultToCDXFormatAdapter.adapt(itr, cdxFormat);
      pw.println(cdxSpec);
      while(lines.hasNext()) {
        pw.println(lines.next());
View Full Code Here
26
27
28
29
30
31
32
33
34
35
36
import junit.framework.TestCase;

public class CDXFormatTest extends TestCase {
  public void testParseSpec() {
    CaptureSearchResult c;
    CDXFormat f = OKFormat(" CDX a V");
    c = OKParse(f,"http://foo.com 12");
    assertEquals("http://foo.com",c.getOriginalUrl());
    assertEquals(c.getOffset(), 12);

   
View Full Code Here
67
68
69
70
71
72
73
74
75
76
77
78
79
      fail(e.getLocalizedMessage());
    }
    return r;
  }
  private CDXFormat OKFormat(String format) {
    CDXFormat f = null;
    try {
      f = new CDXFormat(format);
    } catch (CDXFormatException e) {
      fail("Format '" + format + "' should NOT have thrown exception");
    }
    return f;
  }
View Full Code Here
77
78
79
80
81
82
83
84
85
86
87
    }
    return f;
  }
  private void exceptionFormat(String format) {
    try {
      new CDXFormat(format);
      fail("Format '" + format + "' should have thrown exception");
    } catch (CDXFormatException e) {
    }
   
  }
View Full Code Here
TOP

Related Classes of org.archive.wayback.resourceindex.cdx.format.CDXFormat

  • org.archive.wayback.core.CaptureSearchResult
  • org.archive.wayback.resourceindex.cdx.CDXFormatIndex
  • org.archive.wayback.resourceindex.cdx.format.CDXFormatTest
  • org.archive.wayback.resourceindex.ziplines.ZiplinesSearchResultSource
  • org.archive.wayback.resourcestore.indexer.IndexWorker
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.