/**
* Combine the status stored in the index and the underlying status.
*
* @param h status stored in the index
* @param cache caching the underlying file statuses
* @return the combined file status
* @throws IOException
*/
private FileStatus toFileStatus(HarStatus h, Map<String, FileStatus> cache) throws IOException {
FileStatus underlying = null;
if (cache != null) {
underlying = cache.get(h.partName);
}
if (underlying == null) {
final Path p = h.isDir ? archivePath : new Path(archivePath, h.partName);
underlying = fs.getFileStatus(p);
if (cache != null) {
cache.put(h.partName, underlying);
}
}
long modTime = 0;
int version = metadata.getVersion();
if (version < 3) {
modTime = underlying.getModificationTime();
} else if (version == 3) {
modTime = h.getModificationTime();
}
return new FileStatus(
h.isDir() ? 0L : h.getLength(),
h.isDir(),
underlying.getReplication(),
underlying.getBlockSize(),
modTime,
underlying.getAccessTime(),
underlying.getPermission(),
underlying.getOwner(),
underlying.getGroup(),
makeRelative(this.uri.getPath(), new Path(h.name)));
}
// get the version of the filesystem from the masterindex file
// the version is currently not useful since its the first version
// of archives
public int getHarVersion() throws IOException {
if (metadata != null) {
return metadata.getVersion();
} else {
throw new IOException("Invalid meta data for the Har Filesystem");
}
}
private String decodeFileName(String fname) throws UnsupportedEncodingException {
int version = metadata.getVersion();
if (version == 2 || version == 3) {
return decodeString(fname);
}
return fname;
}
/**
* Initialize a Har filesystem per har archive. The archive home directory is the top level
* directory in the filesystem that contains the HAR archive. Be careful with this method, you do
* not want to go on creating new Filesystem instances per call to path.getFileSystem(). the uri
* of Har is har://underlyingfsscheme-host:port/archivepath. or har:///archivepath. This assumes
* the underlying filesystem to be used in case not specified.
*/
@Override
public void initialize(URI name, Configuration conf) throws IOException {
// initialize the metadata cache, if needed
initializeMetadataCache(conf);
// decode the name
URI underLyingURI = decodeHarURI(name, conf);
// we got the right har Path- now check if this is
// truly a har filesystem
Path harPath = archivePath(new Path(name.getScheme(), name.getAuthority(), name.getPath()));
if (harPath == null) {
throw new IOException("Invalid path for the Har Filesystem. " + name.toString());
}
if (fs == null) {
fs = FileSystem.get(underLyingURI, conf);
}
uri = harPath.toUri();
archivePath = new Path(uri.getPath());
harAuth = getHarAuth(underLyingURI);
// check for the underlying fs containing
// the index file
Path masterIndexPath = new Path(archivePath, "_masterindex");
Path archiveIndexPath = new Path(archivePath, "_index");
if (!fs.exists(masterIndexPath) || !fs.exists(archiveIndexPath)) {
throw new IOException(
"Invalid path for the Har Filesystem. " + "No index file in " + harPath);
}
metadata = harMetaCache.get(uri);
if (metadata != null) {
FileStatus mStat = fs.getFileStatus(masterIndexPath);
FileStatus aStat = fs.getFileStatus(archiveIndexPath);
if (mStat.getModificationTime() != metadata.getMasterIndexTimestamp()
|| aStat.getModificationTime() != metadata.getArchiveIndexTimestamp()) {
// the archive has been overwritten since we last read it
// remove the entry from the meta data cache
metadata = null;
harMetaCache.remove(uri);
}
}
if (metadata == null) {
metadata = new HarMetaData(fs, masterIndexPath, archiveIndexPath);
metadata.parseMetaData();
harMetaCache.put(uri, metadata);
}
}
/**
* Get block locations from the underlying fs and fix their offsets and lengths.
*
* @param file the input file status to get block locations
* @param start the start of the desired range in the contained file
* @param len the length of the desired range
* @return block locations for this segment of file
* @throws IOException
*/
@Override
public BlockLocation[] getFileBlockLocations(FileStatus file, long start, long len)
throws IOException {
HarStatus hstatus = getFileHarStatus(file.getPath());
Path partPath = new Path(archivePath, hstatus.getPartName());
FileStatus partStatus = metadata.getPartFileStatus(partPath);
// get all part blocks that overlap with the desired file blocks
BlockLocation[] locations =
fs.getFileBlockLocations(partStatus, hstatus.getStartIndex() + start, len);
return fixBlockLocations(locations, start, len, hstatus.getStartIndex());
}