Package org.syncany.operations.down

Class DatabaseFileReader

java.lang.Object

org.syncany.operations.down.DatabaseFileReader

All Implemented Interfaces:

java.util.Iterator<MemoryDatabase>

public class DatabaseFileReader
extends java.lang.Object
implements java.util.Iterator<MemoryDatabase>

The DatabaseFileReader provides a way to read a series of database files in a memory-efficient way, by converting them to a series of MemoryDatabases, none of which are too large.

Constructor Summary

Constructors

Constructor	Description
DatabaseFileReader(DatabaseXmlSerializer databaseSerializer, DatabaseBranch winnersApplyBranch, java.util.Map<DatabaseVersionHeader,java.io.File> databaseVersionLocations)

Method Summary

Modifier and Type	Method	Description
boolean	hasNext()
MemoryDatabase	next()	Loads the winner's database branch into the memory in a MemoryDatabase object, by using the already downloaded list of remote database files.
void	remove()

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.util.Iterator

forEachRemaining

Constructor Detail

DatabaseFileReader

public DatabaseFileReader(DatabaseXmlSerializer databaseSerializer,
                          DatabaseBranch winnersApplyBranch,
                          java.util.Map<DatabaseVersionHeader,java.io.File> databaseVersionLocations)

Method Detail

hasNext

public boolean hasNext()

Specified by:

hasNext in interface java.util.Iterator<MemoryDatabase>

next

public MemoryDatabase next()

Loads the winner's database branch into the memory in a MemoryDatabase object, by using the already downloaded list of remote database files.

Because database files can contain multiple DatabaseVersions per client, a range for which to load the database versions must be determined.

Example 1:

  db-A-0001   (A1)     Already known             Not loaded
  db-A-0005   (A2)     Already known             Not loaded
              (A3)     Already known             Not loaded
              (A4)     Part of winner's branch   Loaded
              (A5)     Purge database version    Ignored (only DEFAULT)
  db-B-0001   (A5,B1)  Part of winner's branch   Loaded
  db-A-0006   (A6,B1)  Part of winner's branch   Loaded
 

In example 1, only (A4)-(A5) must be loaded from db-A-0005, and not all four database versions.

Other example:

  db-A-0005   (A1)     Part of winner's branch   Loaded
  db-A-0005   (A2)     Part of winner's branch   Loaded
  db-B-0001   (A2,B1)  Part of winner's branch   Loaded
  db-A-0005   (A3,B1)  Part of winner's branch   Loaded
  db-A-0005   (A4,B1)  Part of winner's branch   Loaded
  db-A-0005   (A5,B1)  Purge database version    Ignored (only DEFAULT)
 

In example 2, (A1)-(A5,B1) [except (A2,B1)] are contained in db-A-0005 (after merging!), so db-A-0005 must be processed twice; each time loading separate parts of the file. In this case: First load (A1)-(A2) from db-A-0005, then load (A2,B1) from db-B-0001, then load (A3,B1)-(A4,B1) from db-A-0005, and ignore (A5,B1).

Specified by:

next in interface java.util.Iterator<MemoryDatabase>

Returns:

Returns a loaded memory database containing all metadata from the winner's branch

remove

public void remove()

Specified by:

remove in interface java.util.Iterator<MemoryDatabase>