lee/heritrix3
1
0
Fork 0
mirror of https://github.com/internetarchive/heritrix3.git synced 2026-09-26 15:46:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix bug pointed out in https://webarchive.jira.com/browse/ARI-3176?focusedCommentId=35314 - url-agnostic dedupe not reported in host report (and other reports)

Browse Source
This commit is contained in:
Noah Levitt
2013-04-29 18:11:24 -07:00
parent dafd95eef7
commit d5693563f8
1 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/src/main/java/org/archive/modules/fetcher/FetchStats.java
+7 -1
View File
@@ -72,6 +72,9 @@ public class FetchStats implements Serializable, FetchStatusCodes, Reporter {
protected long lastSuccessTime;
/*
* XXX redundancy with StatisticsTracker.onApplicationEvent() ... CrawledBytesHistotable.accumulate() code path
*/
public synchronized void tally(CrawlURI curi, Stage stage) {
switch(stage) {
case SCHEDULED:
@@ -91,7 +94,10 @@ public class FetchStats implements Serializable, FetchStatusCodes, Reporter {
if (curi.getFetchStatus() == HttpStatus.SC_NOT_MODIFIED) {
notModifiedBytes += curi.getContentSize();
notModifiedUrls++;
} else if (IdenticalDigestDecideRule.hasIdenticalDigest(curi)){
} else if (IdenticalDigestDecideRule.hasIdenticalDigest(curi)) {
dupByHashBytes += curi.getContentSize();
dupByHashUrls++;
} else if (curi.getAnnotations().contains("warcRevisit:uriAgnosticDigest")) {
dupByHashBytes += curi.getContentSize();
dupByHashUrls++;
} else {