X-Git-Url: http://wamblee.org/gitweb/?a=blobdiff_plain;f=crawler%2Fkissweb%2Fsrc%2Forg%2Fwamblee%2Fcrawler%2Fkiss%2Fscheduling%2FCrawlerSchedule.java;h=5121f92058df76a437ca159e595c1b0e1f089866;hb=3c68ffbb4f2b908015301accc3f4473acc4619e2;hp=e341169a82e76e5cad7da3f6cacda4b7cc84097b;hpb=26ae3728e2da2d86fe42994b9223fe44690dbfd0;p=utils diff --git a/crawler/kissweb/src/org/wamblee/crawler/kiss/scheduling/CrawlerSchedule.java b/crawler/kissweb/src/org/wamblee/crawler/kiss/scheduling/CrawlerSchedule.java index e341169a..5121f920 100644 --- a/crawler/kissweb/src/org/wamblee/crawler/kiss/scheduling/CrawlerSchedule.java +++ b/crawler/kissweb/src/org/wamblee/crawler/kiss/scheduling/CrawlerSchedule.java @@ -43,14 +43,20 @@ public class CrawlerSchedule implements Serializable { /** * Constructs the scheduler. - * @param aCrawler The interface by which the crawler is executed. + * The crawler will run if it is triggered in the range between the minimum (included) + * and maximum (included) hour of the day if either + *
aHourMin
* @param aHourMax The crawler may only run if hour <= aHourMax
*/
public CrawlerSchedule(CrawlerExecutor aCrawler, int aHourMin, int aHourMax) {
_crawler = aCrawler;
- _lastExecuted = null;
- _lastResult = false;
+ _lastExecuted = new Date();
+ _lastResult = true; // the crawler will automatically run the next day.
_lastException = null;
_hourMin = aHourMin;
_hourMax = aHourMax;
@@ -62,6 +68,7 @@ public class CrawlerSchedule implements Serializable {
* @param aDate Time at which we are executing now.
*/
public void execute(Date aDate) {
+
if (mustExecute(aDate)) {
LOG.info("Executing crawler at " + aDate);
try {
@@ -85,6 +92,10 @@ public class CrawlerSchedule implements Serializable {
return _lastExecuted;
}
+ public void setLastExecuted(Date aDate) {
+ _lastExecuted = aDate;
+ }
+
/**
* Gets the result of the last execution.
* @return True iff last execution was a success.
@@ -108,11 +119,14 @@ public class CrawlerSchedule implements Serializable {
* @return True iff the crawler must be run.
*/
private boolean mustExecute(Date aDate) {
+ if ( _lastExecuted == null ) {
+ return false; // crawler must be started manually at least once after deployment.
+ }
Calendar calendar = Calendar.getInstance();
calendar.setTime(aDate);
int hour = calendar.get(Calendar.HOUR_OF_DAY);
if ( hour < _hourMin ) {
- return false;
+ return false;
}
if (hour > _hourMax ) {
return false;
@@ -128,17 +142,24 @@ public class CrawlerSchedule implements Serializable {
return false; // already run successfully today.
}
+ /**
+ * Determines if the last execution was on the same day.
+ * @param aDate Current time.
+ * @return True iff last execution was on the same day.
+ */
private boolean lastExecutionWasOnSameDay(Date aDate) {
if ( _lastExecuted == null ) {
return false;
}
int curDay = getDayOfYear(aDate);
int lastDay = getDayOfYear(_lastExecuted);
- return curDay == lastDay;
+ return curDay == lastDay; // check can be invalid only if scheduling interval is one year,
+ // which is ridiculous.
}
/**
- * @param aDate
+ * Gets the day of the year
+ * @param aDate Date to compute day for.
*/
private int getDayOfYear(Date aDate) {
Calendar calendar = Calendar.getInstance();