Package dk.netarkivet.harvester.indexserver

Class RawMetadataCache

java.lang.Object

dk.netarkivet.harvester.indexserver.FileBasedCache<Long>

dk.netarkivet.harvester.indexserver.RawMetadataCache

All Implemented Interfaces:

RawDataCache

Direct Known Subclasses:

CDXDataCache, CrawlLogDataCache

public class RawMetadataCache
extends FileBasedCache<Long>
implements RawDataCache

This is an implementation of the RawDataCache specialized for data out of metadata files. It uses regular expressions for matching URL and mime-type of ARC entries for the kind of metadata we want.

Field Summary

Fields

Modifier and Type	Field	Description
static Pattern	MATCH_ALL_PATTERN	A regular expression object that matches everything.

Fields inherited from class dk.netarkivet.harvester.indexserver.FileBasedCache

cacheDir

Constructor Summary

Constructors

Constructor	Description
RawMetadataCache(String prefix, Pattern urlMatcher, Pattern mimeMatcher)	Create a new RawMetadataCache.

Method Summary

Modifier and Type	Method	Description
protected Long	cacheData(Long id)	Actually cache data for the given ID.
File	getCacheFile(Long id)	Get the file potentially containing (cached) data for a single job.

Methods inherited from class dk.netarkivet.harvester.indexserver.FileBasedCache

cache, get, getCacheDir, getIndex

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface dk.netarkivet.harvester.indexserver.RawDataCache

cache, get

Field Detail

MATCH_ALL_PATTERN

public static final Pattern MATCH_ALL_PATTERN

A regular expression object that matches everything.

Constructor Detail

RawMetadataCache

public RawMetadataCache(String prefix,
                        Pattern urlMatcher,
                        Pattern mimeMatcher)

Create a new RawMetadataCache. For a given job ID, this will fetch and cache selected content from metadata files (<ID>-metadata-[0-9]+.arc). Any entry in a metadata file that matches both patterns will be returned. The returned data does not directly indicate which file they were from, though parts intrinsic to the particular format might.

Parameters:

prefix - A prefix that will be used to distinguish this cache's files from other caches'. It will be used for creating a directory, so it must not contain characters not legal in directory names.

urlMatcher - A pattern for matching URLs of the desired entries. If null, a .* pattern will be used.

mimeMatcher - A pattern for matching mime-types of the desired entries. If null, a .* pattern will be used.

Method Detail

getCacheFile

public File getCacheFile(Long id)

Get the file potentially containing (cached) data for a single job.

Specified by:

getCacheFile in class FileBasedCache<Long>

Parameters:

id - The job to find data for.

Returns:

The file where cache data for the job can be stored.

See Also:

FileBasedCache.getCacheFile(Object)

cacheData

protected Long cacheData(Long id)

Actually cache data for the given ID.

Specified by:

cacheData in class FileBasedCache<Long>

Parameters:

id - A job ID to cache data for.

Returns:

A File containing the data. This file will be the same as getCacheFile(ID);

See Also:

FileBasedCache.cacheData(Object)