com.bigdata.bfs

Class FileMetadataSchema

java.lang.Object

com.bigdata.sparse.Schema

com.bigdata.bfs.FileMetadataSchema

All Implemented Interfaces:

Externalizable, Serializable

public class FileMetadataSchema
extends Schema

The schema for metadata about file versions stored in the BigdataFileSystem. Some well known properties are always defined, but any property may be stored - ideally within their own namespace!

Note: File version creation time and update times are available using the SparseRowStore, which stores and reports the timestamp for each property value. Convenience methods are available on RepositoryDocumentImpl to report those timestamps. Timestamps for file blocks can NOT be obtained.

Note: A content length property was deliberately NOT defined. The design is geared towards very large file and asynchronous read/write of file blocks. The length of short files may be readily computed by the expediency of sucking their contents into a buffer. Large files should always be processed using a stream-oriented technique or distributed to concurrent clients in block sized pieces.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static String	CONTENT_ENCODING The encoding, if any, used to convert the byte[] content to characters.
static String	CONTENT_TYPE The MIME type associated with the content (the same semantics as the HTTP Content-Type header).
static String	ID The content identifer is an arbitrary Unicode String whose value is defined by the client.
static String	INDEX_TEXT Optional boolean property.
static String	VERSION The file version number.

Constructor Summary

Constructors

Constructor and Description
FileMetadataSchema()

Method Summary

Methods inherited from class com.bigdata.sparse.Schema

appendPrimaryKey, fromKey, getKey, getName, getPrefix, getPrimaryKeyName, getPrimaryKeyType, getSchemaBytes, readExternal, toString, writeExternal

Methods inherited from class java.lang.Object

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

Field Detail

ID

public static final transient String ID

The content identifer is an arbitrary Unicode String whose value is defined by the client. This is generally a pathname, similar to a path in a file system.

CONTENT_TYPE

public static final transient String CONTENT_TYPE

The MIME type associated with the content (the same semantics as the HTTP Content-Type header).

CONTENT_ENCODING

public static final transient String CONTENT_ENCODING

The encoding, if any, used to convert the byte[] content to characters.

Note: This is typically deduced from an analysis of the MIME Type in Content-Type header and at times the leading bytes of the response body itself.

VERSION

public static final transient String VERSION

The file version number. Together the file ID and the file VERSION form the primary key for the data index. A null value is stored in this field when the file version is deleted.

INDEX_TEXT

public static final transient String INDEX_TEXT

Optional boolean property. When present, the text in the file version will be tokenized and indexed for free text search. Non-text files should not set this property. Instead, a workflow or map/reduce job should be written to generate a text-only variant for the indexer.

Constructor Detail

FileMetadataSchema

public FileMetadataSchema()