Attachment 'deduplicator-0.3.0-20061218-patch-local-dateformat.patch'
Download 1 diff -Naur deduplicator-0.3.0-20061218/src/main/java/is/hi/bok/deduplicator/CrawlDataItem.java deduplicator-0.3.0-20061218-patch-local-dateformat/src/main/java/is/hi/bok/deduplicator/CrawlDataItem.java
2 --- deduplicator-0.3.0-20061218/src/main/java/is/hi/bok/deduplicator/CrawlDataItem.java 2006-12-18 08:38:04.000000000 +0100
3 +++ deduplicator-0.3.0-20061218-patch-local-dateformat/src/main/java/is/hi/bok/deduplicator/CrawlDataItem.java 2008-05-22 17:05:02.000000000 +0200
4 @@ -31,13 +31,6 @@
5 * @author Kristinn Sigurðsson
6 */
7 public class CrawlDataItem {
8 -
9 - /**
10 - * The proper formating of {@link #setURL(String)} and {@link #getURL()}
11 - */
12 - public static final SimpleDateFormat dateFormat =
13 - new SimpleDateFormat("yyyyMMddHHmmssSSS");
14 -
15 protected String URL;
16 protected String contentDigest;
17 protected String timestamp;
18 diff -Naur deduplicator-0.3.0-20061218/src/main/java/is/hi/bok/deduplicator/CrawlLogIterator.java deduplicator-0.3.0-20061218-patch-local-dateformat/src/main/java/is/hi/bok/deduplicator/CrawlLogIterator.java
19 --- deduplicator-0.3.0-20061218/src/main/java/is/hi/bok/deduplicator/CrawlLogIterator.java 2006-12-18 08:36:02.000000000 +0100
20 +++ deduplicator-0.3.0-20061218-patch-local-dateformat/src/main/java/is/hi/bok/deduplicator/CrawlLogIterator.java 2008-05-22 17:04:56.000000000 +0200
21 @@ -40,8 +40,10 @@
22 */
23 public class CrawlLogIterator extends CrawlDataIterator {
24
25 - protected static final SimpleDateFormat crawlDateFormat =
26 + protected final SimpleDateFormat crawlDateFormat =
27 new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ss.SSS'Z'");
28 + protected final SimpleDateFormat crawlItemDateFormat =
29 + new SimpleDateFormat("yyyyMMddHHmmssSSS");
30
31 /**
32 * A reader for the crawl.log file being processed
33 @@ -53,8 +55,7 @@
34 * has not been prepared or there are no more elements
35 */
36 protected CrawlDataItem next;
37 -
38 -
39 +
40 /**
41 * Create a new CrawlLogIterator that reads items from a Heritrix crawl.log
42 *
43 @@ -146,7 +147,7 @@
44 try {
45 // Convert from crawl.log format to the format specified by
46 // CrawlDataItem
47 - timestamp = CrawlDataItem.dateFormat.format(
48 + timestamp = crawlItemDateFormat.format(
49 crawlDateFormat.parse(lineParts[0]));
50 } catch (ParseException e) {
51 System.err.println("Error parsing date for: " + line);
Attached Files
To refer to attachments on a page, use attachment:filename, as shown below in the list of files. Do NOT use the URL of the [get] link, since this is subject to change and can break easily.You are not allowed to attach a file to this page.