Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-24 05:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request 'Experimental domain-searching feature' (#67) from master into release

Browse Source 
Reviewed-on: https://git.marginalia.nu/marginalia/marginalia.nu/pulls/67
This commit is contained in:
Viktor Lofgren 2022-07-28 21:10:18 +02:00
commit 29efdf944d
2 changed files with 14 additions and 2 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/EdgeDataStoreDaoImpl.java
View File

@ -277,7 +277,7 @@ public class EdgeDataStoreDaoImpl implements EdgeDataStoreDao {
// this is safe, string concatenation is of integers // this is safe, string concatenation 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 EC_URL_VIEW.ID IN " + inStmt + " ORDER BY RANK ASC"); 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 QUALITY>-10 AND EC_URL_VIEW.ID IN " + inStmt + " ORDER BY (KNOWN_URLS<1000) ASC, (KNOWN_URLS<2000) ASC");
while (rsp.next()) { while (rsp.next()) {
int id = rsp.getInt(1); int id = rsp.getInt(1);
String domain = rsp.getString(2); String domain = rsp.getString(2);

14
marginalia_nu/src/main/java/nu/marginalia/wmsa/edge/search/EdgeSearchOperator.java
View File

@ -135,8 +135,20 @@ public class EdgeSearchOperator {
Set<EdgeId<EdgeUrl>> results = new LinkedHashSet<>(); Set<EdgeId<EdgeUrl>> results = new LinkedHashSet<>();
List<Iterator<EdgeId<EdgeUrl>>> iters = new ArrayList<>();
for (var result : indexClient.queryDomains(ctx, requests)) { for (var result : indexClient.queryDomains(ctx, requests)) {
results.addAll(result.results); iters.add(result.results.iterator());
}
while (!iters.isEmpty()) {
iters.removeIf(iter -> {
if (!iter.hasNext()) return true;
else {
results.add(iter.next());
return false;
}
});
} }
return edgeDataStoreDao.getBrowseResultFromUrlIds(new ArrayList<>(results)); return edgeDataStoreDao.getBrowseResultFromUrlIds(new ArrayList<>(results));