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

(crawler) Add crawl delays around probe call and deal with 429:s properly during this phase

Browse Source
This commit is contained in:
Viktor Lofgren 2024-07-16 15:33:24 +02:00
parent 7eb955cc42
commit 6665e447aa
1 changed files with 20 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
code/processes/crawling-process/java/nu/marginalia/crawl/retreival/CrawlerRetreiver.java
View File

@ -120,7 +120,9 @@ public class CrawlerRetreiver implements AutoCloseable {
final SimpleRobotRules robotsRules = fetcher.fetchRobotRules(rootUrl.domain, warcRecorder); final SimpleRobotRules robotsRules = fetcher.fetchRobotRules(rootUrl.domain, warcRecorder);
final CrawlDelayTimer delayTimer = new CrawlDelayTimer(robotsRules.getCrawlDelay()); final CrawlDelayTimer delayTimer = new CrawlDelayTimer(robotsRules.getCrawlDelay());
sniffRootDocument(rootUrl); delayTimer.waitFetchDelay(0); // initial delay after robots.txt
sniffRootDocument(rootUrl, delayTimer);
delayTimer.waitFetchDelay(0); // delay after sniffing
// Play back the old crawl data (if present) and fetch the documents comparing etags and last-modified // Play back the old crawl data (if present) and fetch the documents comparing etags and last-modified
int recrawled = crawlerRevisitor.recrawl(oldCrawlData, robotsRules, delayTimer); int recrawled = crawlerRevisitor.recrawl(oldCrawlData, robotsRules, delayTimer);
@ -193,13 +195,28 @@ public class CrawlerRetreiver implements AutoCloseable {
return fetchedCount; return fetchedCount;
} }
private void sniffRootDocument(EdgeUrl rootUrl) { private void sniffRootDocument(EdgeUrl rootUrl, CrawlDelayTimer timer) {
try { try {
logger.debug("Configuring link filter"); logger.debug("Configuring link filter");
var url = rootUrl.withPathAndParam("/", null); var url = rootUrl.withPathAndParam("/", null);
var result = fetcher.fetchContent(url, warcRecorder, ContentTags.empty()); HttpFetchResult result = null;
for (int i = 0; i <= HTTP_429_RETRY_LIMIT; i++) {
try {
result = fetcher.fetchContent(url, warcRecorder, ContentTags.empty());
break;
}
catch (RateLimitException ex) {
timer.waitRetryDelay(ex);
}
catch (Exception ex) {
logger.warn("Failed to fetch {}", url, ex);
result = new HttpFetchResult.ResultException(ex);
}
}
if (!(result instanceof HttpFetchResult.ResultOk ok)) if (!(result instanceof HttpFetchResult.ResultOk ok))
return; return;