Marown/yacy_search_server
1
0
Fork 0
mirror of https://github.com/yacy/yacy_search_server.git synced 2024-09-19 00:01:41 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity

prevent loading of content from the cache when retrieval with IFFRESH is

Browse Source 
used and cache is stale. Should speed up snippet generation when cache
strategy is IFFRESH.
This commit is contained in:
Michael Peter Christen 2012-07-06 08:29:41 +02:00
parent 91f14ea38e
commit c3db015410
2 changed files with 24 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
source/de/anomic/crawler/Cache.java
View File

@ -297,6 +297,16 @@ public final class Cache {
}
}
public static boolean hasContent(final byte[] hash) {
// load the url as resource from the cache
try {
return fileDB.containsKey(hash);
} catch (final OutOfMemoryError e) {
Log.logException(e);
return false;
}
}
/**
* removed response header and cached content from the database
* @param url

23
source/net/yacy/repository/LoaderDispatcher.java
View File

@ -201,8 +201,7 @@ public final class LoaderDispatcher {
// now see if there is a cache entry
final ResponseHeader cachedResponse = (url.isLocal()) ? null : Cache.getResponseHeader(url.hash());
final byte[] content = (cachedResponse == null) ? null : Cache.getContent(url.hash());
if (cachedResponse != null && content != null) {
if (cachedResponse != null && Cache.hasContent(url.hash())) {
// yes we have the content
// create request header values and a response object because we need that
@ -218,26 +217,32 @@ public final class LoaderDispatcher {
cachedResponse,
crawlProfile,
true,
content);
null);
// check which caching strategy shall be used
if (cacheStrategy == CacheStrategy.IFEXIST || cacheStrategy == CacheStrategy.CACHEONLY) {
// well, just take the cache and don't care about freshness of the content
this.log.logInfo("cache hit/useall for: " + url.toNormalform(true, false));
return response;
final byte[] content = Cache.getContent(url.hash());
if (content != null) {
this.log.logInfo("cache hit/useall for: " + url.toNormalform(true, false));
response.setContent(content);
return response;
}
}
// now the cacheStrategy must be CACHE_STRATEGY_IFFRESH, that means we should do a proxy freshness test
assert cacheStrategy == CacheStrategy.IFFRESH : "cacheStrategy = " + cacheStrategy;
if (response.isFreshForProxy()) {
this.log.logInfo("cache hit/fresh for: " + url.toNormalform(true, false));
return response;
final byte[] content = Cache.getContent(url.hash());
if (content != null) {
this.log.logInfo("cache hit/fresh for: " + url.toNormalform(true, false));
response.setContent(content);
return response;
}
}
this.log.logInfo("cache hit/stale for: " + url.toNormalform(true, false));
} else if (cachedResponse != null) {
this.log.logWarning("HTCACHE contained response header, but not content for url " + url.toNormalform(true, false));
} else if (content != null) {
this.log.logWarning("HTCACHE contained content, but not response header for url " + url.toNormalform(true, false));
}
}