diff --git a/defaults/yacy.init b/defaults/yacy.init index d6b99f872..16127bda2 100644 --- a/defaults/yacy.init +++ b/defaults/yacy.init @@ -694,6 +694,7 @@ crawlPause.remotesearch=10 crawler.clientTimeout=9000 # http crawler specific settings; size in bytes +crawler.http.accept=text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 crawler.http.acceptEncoding=gzip crawler.http.acceptLanguage=en-us,en;q=0.5 crawler.http.acceptCharset=ISO-8859-1,utf-8;q=0.7,*;q=0.7 diff --git a/defaults/yacy.network.allip.unit b/defaults/yacy.network.allip.unit index c2bd5008a..eab34bf69 100644 --- a/defaults/yacy.network.allip.unit +++ b/defaults/yacy.network.allip.unit @@ -23,4 +23,7 @@ network.unit.protocol.control = uncontrolled # white/blacklists network.unit.access.whitelist = 10\..*,127\..*,172\.(1[6-9]|2[0-9]|3[0-1])\..*,169\.254\..*,192\.168\..*,localhost -network.unit.access.blacklist = \ No newline at end of file +network.unit.access.blacklist = + +# optional user agent to keep activities from out of a intranet environment secret +#network.unit.tenant.agent=Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1) diff --git a/source/de/anomic/crawler/RobotsTxt.java b/source/de/anomic/crawler/RobotsTxt.java index 6a439b8e7..f56ea0e4e 100644 --- a/source/de/anomic/crawler/RobotsTxt.java +++ b/source/de/anomic/crawler/RobotsTxt.java @@ -49,6 +49,7 @@ import net.yacy.kelondro.logging.Log; import org.apache.log4j.Logger; +import de.anomic.crawler.retrieval.HTTPLoader; import de.anomic.data.WorkTables; public class RobotsTxt { @@ -308,7 +309,7 @@ public class RobotsTxt { // adding referer reqHeaders.put(RequestHeader.REFERER, (MultiProtocolURI.newURL(robotsURL,"/")).toNormalform(true, true)); - + reqHeaders.put(RequestHeader.ACCEPT, HTTPLoader.DEFAULT_ACCEPT); if (entry != null) { oldEtag = entry.getETag(); reqHeaders = new RequestHeader(); diff --git a/source/de/anomic/crawler/retrieval/HTTPLoader.java b/source/de/anomic/crawler/retrieval/HTTPLoader.java index 7c38c90f8..1b98dd41d 100644 --- a/source/de/anomic/crawler/retrieval/HTTPLoader.java +++ b/source/de/anomic/crawler/retrieval/HTTPLoader.java @@ -1,4 +1,4 @@ -// HTTPLoader.java +// HTTPLoader.java // --------------- // (C) by Michael Peter Christen; mc@yacy.net // first published on http://yacy.net @@ -39,7 +39,6 @@ import net.yacy.kelondro.logging.Log; import net.yacy.repository.Blacklist; import net.yacy.search.Switchboard; import net.yacy.search.index.Segments; - import de.anomic.crawler.CrawlProfile; import de.anomic.crawler.Latency; import de.anomic.crawler.ZURL.FailCategory; @@ -51,80 +50,82 @@ public final class HTTPLoader { private static final String DEFAULT_ENCODING = "gzip,deflate"; private static final String DEFAULT_LANGUAGE = "en-us,en;q=0.5"; private static final String DEFAULT_CHARSET = "ISO-8859-1,utf-8;q=0.7,*;q=0.7"; + public static final String DEFAULT_ACCEPT = "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8"; public static final int DEFAULT_MAXFILESIZE = 1024 * 1024 * 10; public static final int DEFAULT_CRAWLING_RETRY_COUNT = 5; - + /** * The socket timeout that should be used */ private final int socketTimeout; private final Switchboard sb; private final Log log; - + public HTTPLoader(final Switchboard sb, final Log theLog) { this.sb = sb; this.log = theLog; - + // refreshing timeout value this.socketTimeout = (int) sb.getConfigLong("crawler.clientTimeout", 10000); - } - + } + public Response load(final Request entry, final int maxFileSize, final boolean checkBlacklist) throws IOException { - long start = System.currentTimeMillis(); - Response doc = load(entry, DEFAULT_CRAWLING_RETRY_COUNT, maxFileSize, checkBlacklist); + final long start = System.currentTimeMillis(); + final Response doc = load(entry, DEFAULT_CRAWLING_RETRY_COUNT, maxFileSize, checkBlacklist); Latency.update(entry.url(), System.currentTimeMillis() - start); return doc; } - + private Response load(final Request request, final int retryCount, final int maxFileSize, final boolean checkBlacklist) throws IOException { if (retryCount < 0) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection counter exceeded", -1); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection counter exceeded", -1); throw new IOException("Redirection counter exceeded for URL " + request.url().toString() + ". Processing aborted."); } - + DigestURI url = request.url(); - + final String host = url.getHost(); if (host == null || host.length() < 2) throw new IOException("host is not well-formed: '" + host + "'"); final String path = url.getFile(); int port = url.getPort(); final boolean ssl = url.getProtocol().equals("https"); if (port < 0) port = (ssl) ? 443 : 80; - + // check if url is in blacklist final String hostlow = host.toLowerCase(); if (checkBlacklist && Switchboard.urlBlacklist.isListed(Blacklist.BLACKLIST_CRAWLER, hostlow, path)) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_LOAD_CONTEXT, "url in blacklist", -1); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_LOAD_CONTEXT, "url in blacklist", -1); throw new IOException("CRAWLER Rejecting URL '" + request.url().toString() + "'. URL is in blacklist."); } - + // resolve yacy and yacyh domains - AlternativeDomainNames yacyResolver = HTTPDemon.getAlternativeResolver(); + final AlternativeDomainNames yacyResolver = HTTPDemon.getAlternativeResolver(); if(yacyResolver != null) { - String yAddress = yacyResolver.resolve(host); + final String yAddress = yacyResolver.resolve(host); if(yAddress != null) { url = new DigestURI(url.getProtocol() + "://" + yAddress + path); } } - + // take a file from the net Response response = null; - + // create a request header final RequestHeader requestHeader = new RequestHeader(); requestHeader.put(HeaderFramework.USER_AGENT, ClientIdentification.getUserAgent()); DigestURI refererURL = null; - if (request.referrerhash() != null) refererURL = sb.getURL(Segments.Process.LOCALCRAWLING, request.referrerhash()); + if (request.referrerhash() != null) refererURL = this.sb.getURL(Segments.Process.LOCALCRAWLING, request.referrerhash()); if (refererURL != null) requestHeader.put(RequestHeader.REFERER, refererURL.toNormalform(true, true)); - requestHeader.put(HeaderFramework.ACCEPT_LANGUAGE, sb.getConfig("crawler.http.acceptLanguage", DEFAULT_LANGUAGE)); - requestHeader.put(HeaderFramework.ACCEPT_CHARSET, sb.getConfig("crawler.http.acceptCharset", DEFAULT_CHARSET)); - requestHeader.put(HeaderFramework.ACCEPT_ENCODING, sb.getConfig("crawler.http.acceptEncoding", DEFAULT_ENCODING)); + requestHeader.put(HeaderFramework.ACCEPT, this.sb.getConfig("crawler.http.accept", DEFAULT_ACCEPT)); + requestHeader.put(HeaderFramework.ACCEPT_LANGUAGE, this.sb.getConfig("crawler.http.acceptLanguage", DEFAULT_LANGUAGE)); + requestHeader.put(HeaderFramework.ACCEPT_CHARSET, this.sb.getConfig("crawler.http.acceptCharset", DEFAULT_CHARSET)); + requestHeader.put(HeaderFramework.ACCEPT_ENCODING, this.sb.getConfig("crawler.http.acceptEncoding", DEFAULT_ENCODING)); // HTTP-Client final HTTPClient client = new HTTPClient(); client.setRedirecting(false); // we want to handle redirection ourselves, so we don't index pages twice - client.setTimout(socketTimeout); + client.setTimout(this.socketTimeout); client.setHeader(requestHeader.entrySet()); // send request final byte[] responseBody = client.GETbytes(url, maxFileSize); @@ -139,10 +140,10 @@ public final class HTTPLoader { redirectionUrlString = redirectionUrlString.trim(); if (redirectionUrlString.length() == 0) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection header empy", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection header empy", code); throw new IOException("CRAWLER Redirection of URL=" + request.url().toString() + " aborted. Location header is empty."); } - + // normalizing URL final DigestURI redirectionUrl = new DigestURI(MultiProtocolURI.newURL(request.url(), redirectionUrlString)); @@ -152,48 +153,48 @@ public final class HTTPLoader { // if we are already doing a shutdown we don't need to retry crawling if (Thread.currentThread().isInterrupted()) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_LOAD_CONTEXT, "server shutdown", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_LOAD_CONTEXT, "server shutdown", code); throw new IOException("CRAWLER Retry of URL=" + request.url().toString() + " aborted because of server shutdown."); } - + // check if the url was already indexed - final String dbname = sb.urlExists(Segments.Process.LOCALCRAWLING, redirectionUrl.hash()); + final String dbname = this.sb.urlExists(Segments.Process.LOCALCRAWLING, redirectionUrl.hash()); if (dbname != null) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection to double content", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "redirection to double content", code); throw new IOException("CRAWLER Redirection of URL=" + request.url().toString() + " ignored. The url appears already in db " + dbname); } - + // retry crawling with new url request.redirectURL(redirectionUrl); return load(request, retryCount - 1, maxFileSize, checkBlacklist); } else { // no redirection url provided - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "no redirection url provided", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "no redirection url provided", code); throw new IOException("REJECTED EMTPY REDIRECTION '" + client.getHttpResponse().getStatusLine() + "' for URL " + request.url().toString()); } } else if (responseBody == null) { // no response, reject file - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "no response body", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "no response body", code); throw new IOException("REJECTED EMPTY RESPONSE BODY '" + client.getHttpResponse().getStatusLine() + "' for URL " + request.url().toString()); } else if (code == 200 || code == 203) { // the transfer is ok - + // we write the new cache entry to file system directly - long contentLength = responseBody.length; + final long contentLength = responseBody.length; ByteCount.addAccountCount(ByteCount.CRAWLER, contentLength); // check length again in case it was not possible to get the length before loading if (maxFileSize > 0 && contentLength > maxFileSize) { - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_PROCESS_CONTEXT, "file size limit exceeded", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.FINAL_PROCESS_CONTEXT, "file size limit exceeded", code); throw new IOException("REJECTED URL " + request.url() + " because file size '" + contentLength + "' exceeds max filesize limit of " + maxFileSize + " bytes. (GET)"); } // create a new cache entry - final CrawlProfile profile = sb.crawler.getActive(request.profileHandle().getBytes()); + final CrawlProfile profile = this.sb.crawler.getActive(request.profileHandle().getBytes()); response = new Response( request, requestHeader, - header, + header, Integer.toString(code), profile, responseBody @@ -202,37 +203,37 @@ public final class HTTPLoader { return response; } else { // if the response has not the right response type then reject file - sb.crawlQueues.errorURL.push(request, sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "wrong http status code", code); + this.sb.crawlQueues.errorURL.push(request, this.sb.peers.mySeed().hash.getBytes(), new Date(), 1, FailCategory.TEMPORARY_NETWORK_FAILURE, "wrong http status code", code); throw new IOException("REJECTED WRONG STATUS TYPE '" + client.getHttpResponse().getStatusLine() + "' for URL " + request.url().toString()); } } - + public static Response load(final Request request) throws IOException { return load(request, 3); } - - private static Response load(final Request request, int retryCount) throws IOException { + + private static Response load(final Request request, final int retryCount) throws IOException { if (retryCount < 0) { throw new IOException("Redirection counter exceeded for URL " + request.url().toString() + ". Processing aborted."); } - + final String host = request.url().getHost(); if (host == null || host.length() < 2) throw new IOException("host is not well-formed: '" + host + "'"); final String path = request.url().getFile(); int port = request.url().getPort(); final boolean ssl = request.url().getProtocol().equals("https"); if (port < 0) port = (ssl) ? 443 : 80; - + // check if url is in blacklist final String hostlow = host.toLowerCase(); if (Switchboard.urlBlacklist != null && Switchboard.urlBlacklist.isListed(Blacklist.BLACKLIST_CRAWLER, hostlow, path)) { throw new IOException("CRAWLER Rejecting URL '" + request.url().toString() + "'. URL is in blacklist."); } - + // take a file from the net Response response = null; - + // create a request header final RequestHeader requestHeader = new RequestHeader(); requestHeader.put(HeaderFramework.USER_AGENT, ClientIdentification.getUserAgent()); @@ -251,17 +252,17 @@ public final class HTTPLoader { if (responseBody != null && (code == 200 || code == 203)) { // the transfer is ok - + //statistics: ByteCount.addAccountCount(ByteCount.CRAWLER, responseBody.length); - + // we write the new cache entry to file system directly // create a new cache entry response = new Response( request, requestHeader, - header, + header, Integer.toString(code), null, responseBody @@ -277,16 +278,16 @@ public final class HTTPLoader { if (redirectionUrlString.length() == 0) { throw new IOException("CRAWLER Redirection of URL=" + request.url().toString() + " aborted. Location header is empty."); } - + // normalizing URL final DigestURI redirectionUrl = new DigestURI(MultiProtocolURI.newURL(request.url(), redirectionUrlString)); - + // if we are already doing a shutdown we don't need to retry crawling if (Thread.currentThread().isInterrupted()) { throw new IOException("CRAWLER Retry of URL=" + request.url().toString() + " aborted because of server shutdown."); } - + // retry crawling with new url request.redirectURL(redirectionUrl); return load(request, retryCount - 1); @@ -297,5 +298,5 @@ public final class HTTPLoader { } return response; } - + }