Author: dogacan Date: Mon Jul 16 23:19:06 2007 New Revision: 556824 URL: http://svn.apache.org/viewvc?view=rev&rev=556824 Log: NUTCH-515 - Next fetch time is set incorrectly.
Modified: lucene/nutch/trunk/CHANGES.txt lucene/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDatum.java lucene/nutch/trunk/src/java/org/apache/nutch/crawl/Injector.java lucene/nutch/trunk/src/java/org/apache/nutch/parse/ParseOutputFormat.java Modified: lucene/nutch/trunk/CHANGES.txt URL: http://svn.apache.org/viewvc/lucene/nutch/trunk/CHANGES.txt?view=diff&rev=556824&r1=556823&r2=556824 ============================================================================== --- lucene/nutch/trunk/CHANGES.txt (original) +++ lucene/nutch/trunk/CHANGES.txt Mon Jul 16 23:19:06 2007 @@ -87,6 +87,8 @@ 29. NUTCH-513 - suffix-urlfilter.txt does not have a template. (dogacan) +30. NUTCH-515 - Next fetch time is set incorrectly. (dogacan) + Release 0.9 - 2007-04-02 1. Changed log4j confiquration to log to stdout on commandline Modified: lucene/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDatum.java URL: http://svn.apache.org/viewvc/lucene/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDatum.java?view=diff&rev=556824&r1=556823&r2=556824 ============================================================================== --- lucene/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDatum.java (original) +++ lucene/nutch/trunk/src/java/org/apache/nutch/crawl/CrawlDatum.java Mon Jul 16 23:19:06 2007 @@ -90,6 +90,7 @@ statNames.put(STATUS_DB_GONE, "db_gone"); statNames.put(STATUS_DB_REDIR_TEMP, "db_redir_temp"); statNames.put(STATUS_DB_REDIR_PERM, "db_redir_perm"); + statNames.put(STATUS_DB_NOTMODIFIED, "db_notmodified"); statNames.put(STATUS_SIGNATURE, "signature"); statNames.put(STATUS_INJECTED, "injected"); statNames.put(STATUS_LINKED, "linked"); Modified: lucene/nutch/trunk/src/java/org/apache/nutch/crawl/Injector.java URL: http://svn.apache.org/viewvc/lucene/nutch/trunk/src/java/org/apache/nutch/crawl/Injector.java?view=diff&rev=556824&r1=556823&r2=556824 ============================================================================== --- lucene/nutch/trunk/src/java/org/apache/nutch/crawl/Injector.java (original) +++ lucene/nutch/trunk/src/java/org/apache/nutch/crawl/Injector.java Mon Jul 16 23:19:06 2007 @@ -56,7 +56,7 @@ public void configure(JobConf job) { this.jobConf = job; urlNormalizers = new URLNormalizers(job, URLNormalizers.SCOPE_INJECT); - interval = jobConf.getFloat("db.default.fetch.interval", 30f); + interval = jobConf.getFloat("db.fetch.interval.default", 2592000.0f); filters = new URLFilters(jobConf); scfilters = new ScoringFilters(jobConf); scoreInjected = jobConf.getFloat("db.score.injected", 1.0f); Modified: lucene/nutch/trunk/src/java/org/apache/nutch/parse/ParseOutputFormat.java URL: http://svn.apache.org/viewvc/lucene/nutch/trunk/src/java/org/apache/nutch/parse/ParseOutputFormat.java?view=diff&rev=556824&r1=556823&r2=556824 ============================================================================== --- lucene/nutch/trunk/src/java/org/apache/nutch/parse/ParseOutputFormat.java (original) +++ lucene/nutch/trunk/src/java/org/apache/nutch/parse/ParseOutputFormat.java Mon Jul 16 23:19:06 2007 @@ -82,7 +82,7 @@ this.filters = new URLFilters(job); this.scfilters = new ScoringFilters(job); final UrlValidator validator = UrlValidator.get(); - final float interval = job.getFloat("db.default.fetch.interval", 30f); + final float interval = job.getFloat("db.fetch.interval.default", 2592000.0f); final boolean ignoreExternalLinks = job.getBoolean("db.ignore.external.links", false); final int maxOutlinks = job.getInt("db.max.outlinks.per.page", 100); ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Nutch-cvs mailing list Nutch-cvs@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nutch-cvs