Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-23 13:09:00 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix build errors after merge

Browse Source
This commit is contained in:
Viktor Lofgren 2024-09-08 10:22:32 +02:00
parent 8f367d96f8
commit a3b0189934
5 changed files with 3 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
code/processes/crawling-process/java/nu/marginalia/crawl/retreival/CrawlerRetreiver.java
View File

@ -9,9 +9,6 @@ import nu.marginalia.crawl.retreival.fetcher.warc.WarcRecorder;
import nu.marginalia.crawl.retreival.revisit.CrawlerRevisitor;
import nu.marginalia.crawl.retreival.revisit.DocumentWithReference;
import nu.marginalia.crawl.retreival.sitemap.SitemapFetcher;
import nu.marginalia.crawling.body.HttpFetchResult;
import nu.marginalia.crawling.model.CrawledDomain;
import nu.marginalia.crawling.model.CrawlerDomainStatus;
import nu.marginalia.ip_blocklist.UrlBlocklist;
import nu.marginalia.link_parser.LinkParser;
import nu.marginalia.model.EdgeDomain;

1
code/processes/crawling-process/java/nu/marginalia/crawl/retreival/fetcher/HttpFetcher.java
View File

@ -4,7 +4,6 @@ import com.google.inject.ImplementedBy;
import crawlercommons.robots.SimpleRobotRules;
import nu.marginalia.crawl.retreival.RateLimitException;
import nu.marginalia.crawl.retreival.fetcher.warc.WarcRecorder;
import nu.marginalia.crawling.body.HttpFetchResult;
import nu.marginalia.model.EdgeDomain;
import nu.marginalia.model.EdgeUrl;
import nu.marginalia.model.body.HttpFetchResult;

3
code/processes/crawling-process/java/nu/marginalia/crawl/retreival/fetcher/HttpFetcherImpl.java
View File

@ -12,9 +12,6 @@ import nu.marginalia.crawl.retreival.fetcher.socket.FastTerminatingSocketFactory
import nu.marginalia.crawl.retreival.fetcher.socket.IpInterceptingNetworkInterceptor;
import nu.marginalia.crawl.retreival.fetcher.socket.NoSecuritySSL;
import nu.marginalia.crawl.retreival.fetcher.warc.WarcRecorder;
import nu.marginalia.crawling.body.ContentTypeLogic;
import nu.marginalia.crawling.body.DocumentBodyExtractor;
import nu.marginalia.crawling.body.HttpFetchResult;
import nu.marginalia.model.EdgeDomain;
import nu.marginalia.model.EdgeUrl;
import nu.marginalia.model.body.ContentTypeLogic;

1
code/processes/crawling-process/model/java/nu/marginalia/io/crawldata/CrawledDomainReader.java
View File

@ -1,7 +1,6 @@
package nu.marginalia.io.crawldata;
import nu.marginalia.io.crawldata.format.ParquetSerializableCrawlDataStream;
import nu.marginalia.crawling.io.format.ParquetSerializableCrawlDataStream;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

6
code/processes/crawling-process/test/nu/marginalia/crawling/HttpFetcherTest.java
View File

@ -6,10 +6,10 @@ import nu.marginalia.crawl.retreival.fetcher.ContentTags;
import nu.marginalia.crawl.retreival.fetcher.HttpFetcher;
import nu.marginalia.crawl.retreival.fetcher.HttpFetcherImpl;
import nu.marginalia.crawl.retreival.fetcher.warc.WarcRecorder;
import nu.marginalia.crawling.body.ContentTypeLogic;
import nu.marginalia.crawling.body.DocumentBodyExtractor;
import nu.marginalia.crawling.body.DocumentBodyResult;
import nu.marginalia.model.EdgeUrl;
import nu.marginalia.model.body.ContentTypeLogic;
import nu.marginalia.model.body.DocumentBodyExtractor;
import nu.marginalia.model.body.DocumentBodyResult;
import org.junit.jupiter.api.Assertions;
import org.junit.jupiter.api.Test;