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

(converter) Route sizeHint to SideloadProcessing

Browse Source 
Route the sizeHint from the input parquet file to SideloadProcessing, so that it can set sideloadSizeAdvice appropriately, instead of using a fixed "large" number.

This is necessary to populate the KNOWN_URL column in the domain data table, which is important as it is used in e.g. calculating how far to re-crawl the site in the future.
This commit is contained in:
Viktor Lofgren 2023-12-30 13:05:10 +01:00
parent 0b112cb4d4
commit 7ba296ccdf
2 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
code/processes/converting-process/src/main/java/nu/marginalia/converting/processor/DomainProcessor.java
View File

@ -57,19 +57,20 @@ public class DomainProcessor {
} }
public ConverterBatchWritableIf createWritable(SerializableCrawlDataStream domain) { public ConverterBatchWritableIf createWritable(SerializableCrawlDataStream domain) {
if (domain.sizeHint() > 10_000) { final int sizeHint = domain.sizeHint();
if (sizeHint > 10_000) {
// If the file is too big, we run a processing mode that doesn't // If the file is too big, we run a processing mode that doesn't
// require loading the entire dataset into RAM // require loading the entire dataset into RAM
logger.info("Sideloading {}", domain.path()); return sideloadProcessing(domain, sizeHint);
return sideloadProcessing(domain);
} }
return fullProcessing(domain); return fullProcessing(domain);
} }
public SideloadProcessing sideloadProcessing(SerializableCrawlDataStream dataStream) { public SideloadProcessing sideloadProcessing(SerializableCrawlDataStream dataStream, int sizeHint) {
try { try {
return new SideloadProcessing(dataStream); return new SideloadProcessing(dataStream, sizeHint);
} }
catch (Exception ex) { catch (Exception ex) {
logger.warn("Failed to process domain sideload", ex); logger.warn("Failed to process domain sideload", ex);
@ -86,17 +87,16 @@ public class DomainProcessor {
private final DomainLinks externalDomainLinks; private final DomainLinks externalDomainLinks;
private final LshDocumentDeduplicator deduplicator = new LshDocumentDeduplicator(); private final LshDocumentDeduplicator deduplicator = new LshDocumentDeduplicator();
SideloadProcessing(SerializableCrawlDataStream dataStream) throws IOException { SideloadProcessing(SerializableCrawlDataStream dataStream, int sizeHint) throws IOException {
this.dataStream = dataStream; this.dataStream = dataStream;
if (!dataStream.hasNext() if (!dataStream.hasNext() || !(dataStream.next() instanceof CrawledDomain crawledDomain))
|| !(dataStream.next() instanceof CrawledDomain crawledDomain))
{ {
throw new IllegalStateException("First record must be a domain, was " + dataStream.next().getClass().getSimpleName()); throw new IllegalStateException("First record must be a domain, was " + dataStream.next().getClass().getSimpleName());
} }
domain = new ProcessedDomain(); domain = new ProcessedDomain();
domain.sizeloadSizeAdvice = 10_000; domain.sizeloadSizeAdvice = sizeHint == 0 ? 10_000 : sizeHint;
documentDecorator = new DocumentDecorator(anchorTextKeywords); documentDecorator = new DocumentDecorator(anchorTextKeywords);
processDomain(crawledDomain, domain, documentDecorator); processDomain(crawledDomain, domain, documentDecorator);

2
code/processes/converting-process/src/test/java/nu/marginalia/converting/ConvertingIntegrationTest.java
View File

@ -96,7 +96,7 @@ public class ConvertingIntegrationTest {
@Test @Test
public void testMemexMarginaliaNuSideloadProcessing() throws IOException { public void testMemexMarginaliaNuSideloadProcessing() throws IOException {
var ret = domainProcessor.sideloadProcessing(asSerializableCrawlData(readMarginaliaWorkingSet())); var ret = domainProcessor.sideloadProcessing(asSerializableCrawlData(readMarginaliaWorkingSet()), 100);
assertNotNull(ret); assertNotNull(ret);
assertEquals("memex.marginalia.nu", ret.id()); assertEquals("memex.marginalia.nu", ret.id());