workingDir and outputDir

This commit is contained in:
Michele Artini 2020-12-10 14:47:51 +01:00
parent 10f3f7eca7
commit 467aa77279
7 changed files with 45 additions and 28 deletions

View File

@ -30,8 +30,9 @@ public class CheckDuplictedIdsJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(CheckDuplictedIdsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/check_duplicates.json")));
.toString(
CheckDuplictedIdsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/check_duplicates.json")));
parser.parseArgument(args);
final SparkConf conf = new SparkConf();
@ -59,7 +60,8 @@ public class CheckDuplictedIdsJob {
.write()
.mode(SaveMode.Overwrite)
.option("compression", "gzip")
.json(countPath);;
.json(countPath);
;
}

View File

@ -33,8 +33,9 @@ public class GenerateEventsJob {
public static void main(final String[] args) throws Exception {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(GenerateEventsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/generate_events.json")));
.toString(
GenerateEventsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/generate_events.json")));
parser.parseArgument(args);
final Boolean isSparkSessionManaged = Optional
@ -72,8 +73,10 @@ public class GenerateEventsJob {
.readPath(spark, workingDir + "/duplicates", ResultGroup.class);
final Dataset<Event> dataset = groups
.map(g -> EventFinder
.generateEvents(g, dsIdWhitelist, dsIdBlacklist, dsTypeWhitelist, accumulators), Encoders
.map(
g -> EventFinder
.generateEvents(g, dsIdWhitelist, dsIdBlacklist, dsTypeWhitelist, accumulators),
Encoders
.bean(EventGroup.class))
.flatMap(g -> g.getData().iterator(), Encoders.bean(Event.class));

View File

@ -33,8 +33,9 @@ public class GenerateStatsJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(GenerateStatsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/stats_params.json")));
.toString(
GenerateStatsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/stats_params.json")));
parser.parseArgument(args);
final Boolean isSparkSessionManaged = Optional

View File

@ -39,8 +39,9 @@ public class IndexEventSubsetJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(IndexEventSubsetJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_event_subset.json")));
.toString(
IndexEventSubsetJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_event_subset.json")));
parser.parseArgument(args);
final SparkConf conf = new SparkConf();

View File

@ -47,8 +47,9 @@ public class IndexNotificationsJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(IndexNotificationsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_notifications.json")));
.toString(
IndexNotificationsJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_notifications.json")));
parser.parseArgument(args);
final SparkConf conf = new SparkConf();
@ -116,7 +117,8 @@ public class IndexNotificationsJob {
final long date) {
final List<Notification> list = subscriptions
.stream()
.filter(s -> StringUtils.isBlank(s.getTopic()) || s.getTopic().equals("*") || s.getTopic().equals(e.getTopic()))
.filter(
s -> StringUtils.isBlank(s.getTopic()) || s.getTopic().equals("*") || s.getTopic().equals(e.getTopic()))
.filter(s -> verifyConditions(e.getMap(), s.conditionsAsMap()))
.map(s -> generateNotification(s, e, date))
.collect(Collectors.toList());
@ -147,15 +149,18 @@ public class IndexNotificationsJob {
if (conditions.containsKey("trust")
&& !SubscriptionUtils
.verifyFloatRange(map.getTrust(), conditions.get("trust").get(0).getValue(), conditions.get("trust").get(0).getOtherValue())) {
.verifyFloatRange(
map.getTrust(), conditions.get("trust").get(0).getValue(),
conditions.get("trust").get(0).getOtherValue())) {
return false;
}
if (conditions.containsKey("targetDateofacceptance") && !conditions
.get("targetDateofacceptance")
.stream()
.anyMatch(c -> SubscriptionUtils
.verifyDateRange(map.getTargetDateofacceptance(), c.getValue(), c.getOtherValue()))) {
.anyMatch(
c -> SubscriptionUtils
.verifyDateRange(map.getTargetDateofacceptance(), c.getValue(), c.getOtherValue()))) {
return false;
}

View File

@ -29,8 +29,9 @@ public class IndexOnESJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(IndexOnESJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_es.json")));
.toString(
IndexOnESJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/index_es.json")));
parser.parseArgument(args);
final SparkConf conf = new SparkConf();

View File

@ -42,8 +42,9 @@ public class PartitionEventsByDsIdJob {
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(PartitionEventsByDsIdJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/od_partitions_params.json")));
.toString(
PartitionEventsByDsIdJob.class
.getResourceAsStream("/eu/dnetlib/dhp/broker/oa/od_partitions_params.json")));
parser.parseArgument(args);
final Boolean isSparkSessionManaged = Optional
@ -66,12 +67,13 @@ public class PartitionEventsByDsIdJob {
final Set<String> validOpendoarIds = new HashSet<>();
if (!opendoarIds.trim().equals("-")) {
validOpendoarIds
.addAll(Arrays
.stream(opendoarIds.split(","))
.map(String::trim)
.filter(StringUtils::isNotBlank)
.map(s -> OPENDOAR_NSPREFIX + DigestUtils.md5Hex(s))
.collect(Collectors.toSet()));
.addAll(
Arrays
.stream(opendoarIds.split(","))
.map(String::trim)
.filter(StringUtils::isNotBlank)
.map(s -> OPENDOAR_NSPREFIX + DigestUtils.md5Hex(s))
.collect(Collectors.toSet()));
}
log.info("validOpendoarIds: {}", validOpendoarIds);
@ -82,7 +84,9 @@ public class PartitionEventsByDsIdJob {
.filter((FilterFunction<Event>) e -> StringUtils.isNotBlank(e.getMap().getTargetDatasourceId()))
.filter((FilterFunction<Event>) e -> e.getMap().getTargetDatasourceId().startsWith(OPENDOAR_NSPREFIX))
.filter((FilterFunction<Event>) e -> validOpendoarIds.contains(e.getMap().getTargetDatasourceId()))
.map((MapFunction<Event, ShortEventMessageWithGroupId>) e -> messageFromNotification(e), Encoders.bean(ShortEventMessageWithGroupId.class))
.map(
(MapFunction<Event, ShortEventMessageWithGroupId>) e -> messageFromNotification(e),
Encoders.bean(ShortEventMessageWithGroupId.class))
.coalesce(1)
.write()
.partitionBy("group")