Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-23 21:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Experimental domain-searching feature

Browse Source
This commit is contained in:
vlofgren 2022-07-28 18:46:03 +02:00
parent b94dcb301f
commit b49ebda5dd
3 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
marginalia_nu/src/main/java/nu/marginalia/wmsa/edge/data/dao/EdgeDataStoreDao.java
View File

@ -18,7 +18,7 @@ public interface EdgeDataStoreDao {
List<BrowseResult> getRandomDomains(int count, EdgeDomainBlacklist backlist); List<BrowseResult> getRandomDomains(int count, EdgeDomainBlacklist backlist);
List<BrowseResult> getBrowseResultFromUrlIds(List<EdgeId<EdgeUrl>> urlId); List<BrowseResult> getBrowseResultFromUrlIds(List<EdgeId<EdgeUrl>> urlId, int count);
List<EdgeUrlDetails> getUrlDetailsMulti(List<EdgeId<EdgeUrl>> ids); List<EdgeUrlDetails> getUrlDetailsMulti(List<EdgeId<EdgeUrl>> ids);

6
marginalia_nu/src/main/java/nu/marginalia/wmsa/edge/data/dao/EdgeDataStoreDaoImpl.java
View File

@ -266,7 +266,7 @@ public class EdgeDataStoreDaoImpl implements EdgeDataStoreDao {
} }
@Override @Override
public List<BrowseResult> getBrowseResultFromUrlIds(List<EdgeId<EdgeUrl>> urlId) { public List<BrowseResult> getBrowseResultFromUrlIds(List<EdgeId<EdgeUrl>> urlId, int count) {
if (urlId.isEmpty()) if (urlId.isEmpty())
return Collections.emptyList(); return Collections.emptyList();
@ -277,8 +277,8 @@ public class EdgeDataStoreDaoImpl implements EdgeDataStoreDao {
// this is safe, string cocatenation is of integers // this is safe, string cocatenation is of integers
String inStmt = urlId.stream().map(id -> Integer.toString(id.id())).collect(Collectors.joining(", ", "(", ")")); String inStmt = urlId.stream().map(id -> Integer.toString(id.id())).collect(Collectors.joining(", ", "(", ")"));
var rsp = stmt.executeQuery("SELECT DOMAIN_ID, DOMAIN_NAME FROM EC_URL_VIEW WHERE QUALITY>-10 AND ID IN " + inStmt); var rsp = stmt.executeQuery("SELECT DOMAIN_ID, DOMAIN_NAME FROM EC_URL_VIEW INNER JOIN DOMAIN_METADATA ON EC_URL_VIEW.DOMAIN_ID=DOMAIN_METADATA.ID WHERE VISITED_URLS<500 AND QUALITY>-10 AND ID IN " + inStmt);
while (rsp.next()) { while (rsp.next() && ret.size() < count) {
int id = rsp.getInt(1); int id = rsp.getInt(1);
String domain = rsp.getString(2); String domain = rsp.getString(2);

4
marginalia_nu/src/main/java/nu/marginalia/wmsa/edge/search/EdgeSearchOperator.java
View File

@ -116,10 +116,10 @@ public class EdgeSearchOperator {
.findFirst().map(sq -> sq.searchTermsInclude).orElseGet(Collections::emptyList); .findFirst().map(sq -> sq.searchTermsInclude).orElseGet(Collections::emptyList);
if (keywords.size() == 1) { if (keywords.size() == 1) {
var request = new EdgeDomainSearchSpecification(specs.buckets.get(0), IndexBlock.Topic, keywords.get(0), 1_000_000, 5, 10); var request = new EdgeDomainSearchSpecification(specs.buckets.get(0), IndexBlock.TitleKeywords, keywords.get(0), 1_000_000, 10, 20);
var response = indexClient.queryDomains(ctx, request); var response = indexClient.queryDomains(ctx, request);
return edgeDataStoreDao.getBrowseResultFromUrlIds(response.results); return edgeDataStoreDao.getBrowseResultFromUrlIds(response.results, 5);
} }
return Collections.emptyList(); return Collections.emptyList();