Project Detail

Description

Hi,

I need to be able to capture the changes between 2 files and save them to a 3rd diff file. Note that only the changes should be saved to these diff file. The catch is that I need this to support both small and large files and both text and binary files.

Here is somemore details on how it should function...

Scenario:
souce file = d:testfile1.txt
target file = g:backupfile1.txt
versioned files = g:backupfile1.txt.b00, file1.txt.b01, file1.txt.b02, etc...

Block level Copy function
============================
Block.Copy(src, tgt, append)
src = File to process
tdt = path to store copy of src
append = True/False.
If True append changes from src to tgt
If False create separate diff files with ".bxxx
extention to store each
change. The xxx represents the version.

Example:
if g:backupfile1.txt does not exist Copy D:testfile1.txt to g:backupfile1.txt and subsequent changes to g:backupfile1.txt.b01, file1.txt.b02, etc...

Block.Copy(d:testfile1.txt, g:backupfile1.txt, false)

Block level Restore function
============================
Block.Restore(src, tgt, version)
src = The first copy of the original file
tgt = The path to recreate the original file in
version = the point in time version to restore the file to.

Example:
Restore the 3rdt version of g:backupfile1.txt to d:testfile1.txt

Block.Restore(g:backupfile1.txt, d:testfile1.txt, 3)

I would expect that the solution will handle any file type (text, binary, etc.) The sizes may range from a few KB to 10GB+.

The large files may be ISO images, outlook PST's, NT Backup .bk images, and a variety of other files.

If the parameter "Append" = False Then The differential files (*.bxxx) should contain only the changes. So if I change the original by inserting "Zenu" only Zenu and minimal pointers are saved to the diff file. This is important because as if it is not done I could end up using too much disk space for the changes.

If the parameter "Append" = True Then only append the changes to the target file and do not create additional diff files.

The algorythm should be able to process any files with minimal Memory, and Disk utilization but at the same time be efficient. CPU utilization should be considered as well.

Reagards,

Zenu