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

(crawler) Make DomainCrawlFrontier a bit less aggressive with throwing away excess links when it's approaching full.

Browse Source 
Also be a bit smarter about pre-allocating queues and sets based on depth rather than the number of provided URLs, which was always zero outside of tests.
This commit is contained in:
Viktor Lofgren 2024-10-15 14:22:40 +02:00
parent 4b16022556
commit 481f999b70
1 changed files with 4 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
code/processes/crawling-process/java/nu/marginalia/crawl/retreival/DomainCrawlFrontier.java
View File

@ -46,9 +46,9 @@ public class DomainCrawlFrontier {
this.urlBlocklist = new UrlBlocklist(); this.urlBlocklist = new UrlBlocklist();
this.depth = depth; this.depth = depth;
queue = new ArrayDeque<>(10 + (int) (urls.size()*1.2)); queue = new ArrayDeque<>(depth);
visited = new LongOpenHashSet(10 + (int)(urls.size() * 1.5)); visited = new LongOpenHashSet(depth);
known = new LongOpenHashSet(10 + urls.size() * 2); known = new LongOpenHashSet(depth);
for (String urlStr : urls) { for (String urlStr : urls) {
EdgeUrl.parse(urlStr).ifPresent(this::addToQueue); EdgeUrl.parse(urlStr).ifPresent(this::addToQueue);
@ -63,7 +63,6 @@ public class DomainCrawlFrontier {
int maxDepthIncreaseAbsolute int maxDepthIncreaseAbsolute
) { ) {
int base = Math.max(visited.size(), depth); int base = Math.max(visited.size(), depth);
int scaledUp = (int)(base * depthIncreaseFactor); int scaledUp = (int)(base * depthIncreaseFactor);
depth = Math.min(base + maxDepthIncreaseAbsolute, scaledUp); depth = Math.min(base + maxDepthIncreaseAbsolute, scaledUp);
@ -142,7 +141,7 @@ public class DomainCrawlFrontier {
return; return;
// reduce memory usage by not growing queue huge when crawling large sites // reduce memory usage by not growing queue huge when crawling large sites
if (queue.size() + visited.size() >= depth + 200) if (queue.size() + visited.size() >= depth + 10_000)
return; return;
if (isVisited(url)) if (isVisited(url))