Skip to content

Commit

Permalink
Applied formatting with new version of the plugin
Browse files Browse the repository at this point in the history
Signed-off-by: Julien Nioche <julien@digitalpebble.com>
  • Loading branch information
jnioche committed Oct 29, 2023
1 parent 7092b62 commit e9d0ede
Show file tree
Hide file tree
Showing 4 changed files with 10 additions and 0 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -44,6 +44,7 @@
public class BasicURLNormalizer extends URLFilter {

private static final Logger LOG = LoggerFactory.getLogger(BasicURLNormalizer.class);

/** Nutch 1098 - finds URL encoded parts of the URL */
private static final Pattern unescapeRulePattern = Pattern.compile("%([0-9A-Fa-f]{2})");

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ public abstract class AbstractQueryingSpout extends BaseRichSpout {
* 30 secs.
*/
protected static final String StatusTTLPurgatory = "spout.ttl.purgatory";

/**
* Min time to allow between 2 successive queries to the backend. Value in msecs, default 2000.
*/
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -23,25 +23,31 @@ public class ProtocolResponse {
* implementation and if http.store.headers is true).
*/
public static final String REQUEST_HEADERS_KEY = "_request.headers_";

/** Key which holds the verbatim HTTP response headers in metadata. */
public static final String RESPONSE_HEADERS_KEY = "_response.headers_";

/**
* Key which holds the IP address of the server the request was sent to (response received from)
* in metadata.
*/
public static final String RESPONSE_IP_KEY = "_response.ip_";

/** Key which holds the request time (begin of request) in metadata. */
public static final String REQUEST_TIME_KEY = "_request.time_";

/**
* Key which holds the protocol version(s) used for this request (for layered protocols this
* field may hold multiple comma-separated values)
*/
public static final String PROTOCOL_VERSIONS_KEY = "_protocol_versions_";

/**
* Metadata key which holds a boolean value in metadata whether the response content is trimmed
* or not.
*/
public static final String TRIMMED_RESPONSE_KEY = "http.trimmed";

/**
* Metadata key which holds the reason why content has been trimmed, see {@link
* TrimmedContentReason}.
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -50,8 +50,10 @@ public class WARCRecordFormat implements RecordFormat {
// http://iipc.github.io/warc-specifications/specifications/warc-format/warc-1.1/#warc-record-types
/** WARC record type to hold a HTTP request */
protected static final String WARC_TYPE_REQUEST = "request";

/** WARC record type to hold a HTTP response */
protected static final String WARC_TYPE_RESPONSE = "response";

/**
* WARC record type to hold any other resource, including a HTTP response with no HTTP headers
* available
Expand Down

0 comments on commit e9d0ede

Please sign in to comment.