lsmyrnaios
/
UrlsController
1
0
Fork 1
Code Issues Pull Requests Releases 10 Wiki Activity

Prioritize the publications, which are specified inside the "publication_boost" table, according to their "boost-level".

This commit is contained in:
Lampros Smyrnaios 2023-06-29 12:32:06 +03:00
parent 4c3e2e6b6e
commit a89abe3f2f
2 changed files with 9 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
src/main/java/eu/openaire/urls_controller/configuration/ImpalaConnector.java
View File

@ -67,9 +67,8 @@ public class ImpalaConnector {
jdbcTemplate.execute("CREATE TABLE IF NOT EXISTS " + testDatabaseName + ".publication_urls stored as parquet as select * from " + initialDatabaseName + ".publication_urls");
jdbcTemplate.execute("COMPUTE STATS " + testDatabaseName + ".publication_urls");
// TODO - Should we add the "publication_oids" table? Will it be used in the "getAssignments" query?
jdbcTemplate.execute("CREATE TABLE IF NOT EXISTS " + testDatabaseName + ".publication_boost stored as parquet as select * from " + initialDatabaseName + ".publication_boost");
jdbcTemplate.execute("COMPUTE STATS " + testDatabaseName + ".publication_boost");
jdbcTemplate.execute("CREATE TABLE IF NOT EXISTS " + testDatabaseName + ".datasource stored as parquet as select * from " + initialDatabaseName + ".datasource");
jdbcTemplate.execute("COMPUTE STATS " + testDatabaseName + ".datasource");

12
src/main/java/eu/openaire/urls_controller/services/UrlsServiceImpl.java
View File

@ -111,31 +111,33 @@ public class UrlsServiceImpl implements UrlsService {
String findAssignmentsQuery =
"select pubid, url, datasourceid, datasourcename\n" + // The datsourceName is currently not used. It may be used by the Worker, in the future to apply a datasource-specific aggregation plugin to take the full-texts quickly, instead of using the general crawling one.
"from (select distinct pubid, url, datasourceid, datasourcename, attempt_count, pub_year\n" +
" from (select p.id as pubid, p.year as pub_year, pu.url as url, d.id as datasourceid, d.name as datasourcename, attempts.counts as attempt_count\n" +
" from (select p.id as pubid, p.year as pub_year, pu.url as url, pb.level as level, d.id as datasourceid, d.name as datasourcename, attempts.counts as attempt_count\n" +
" from " + ImpalaConnector.databaseName + ".publication p\n" +
" join " + ImpalaConnector.databaseName + ".publication_urls pu on pu.id=p.id\n" +
" join " + ImpalaConnector.databaseName + ".datasource d on d.id=p.datasourceid\n" +
" left outer join " + ImpalaConnector.databaseName + ".publication_boost pb\n" +
" on p.id=pb.id\n" +
" left outer join (select count(a.id) as counts, a.id from " + ImpalaConnector.databaseName + ".attempt a group by a.id) as attempts\n" +
" on attempts.id=p.id\n" +
" on attempts.id=p.id\n" +
" left outer join (\n" +
" select a.id, a.original_url from " + ImpalaConnector.databaseName + ".assignment a\n" +
" union all\n" +
" select pl.id, pl.original_url from " + ImpalaConnector.databaseName + ".payload pl\n" + // Here we access the payload-VIEW which includes the three payload-tables.
" ) as existing\n" +
" on existing.id=p.id and existing.original_url=pu.url\n" +
" on existing.id=p.id and existing.original_url=pu.url\n" +
" where d.allow_harvest=true and existing.id is null\n" + // For records not found on existing, the "existing.id" will be null.
((excludedDatasourceIDsStringList != null) ? // If we have an exclusion-list, use it below.
(" and d.id not in " + excludedDatasourceIDsStringList + "\n") : "") +
" and coalesce(attempts.counts, 0) <= " + maxAttemptsPerRecordAtomic.get() + "\n" +
" and not exists (select 1 from " + ImpalaConnector.databaseName + ".attempt a where a.id=p.id and a.error_class = 'noRetry' limit 1)\n" +
" and pu.url != '' and pu.url is not null\n" + // Some IDs have empty-string urls, there are no "null" urls, but keep the relevant check for future-proofing.
" order by coalesce(level, -1000) desc\n" +
" limit " + (assignmentsLimit * 10) + ")\n" +
" as non_distinct_results\n" +
" order by coalesce(attempt_count, 0), coalesce(pub_year, 0) desc, reverse(pubid), url\n" +
" order by coalesce(attempt_count, 0), coalesce(pub_year, 0) desc, reverse(pubid), url\n" + // We also order by "id" and "url", in order to get the exactly same records for consecutive runs, all things being equal.
" limit " + assignmentsLimit + ")\n" +
"as findAssignmentsQuery";
// The "order by" in the end makes sure the older attempted records will be re-attempted after a long time.
//logger.trace("findAssignmentsQuery:\n" + findAssignmentsQuery); // DEBUG!