
July 14th, 2005
|
Software Developer | | Join Date: November 4th, 2002 Location: New York
Posts: 1,366
| |
It's information about ranges of the file. For each "leaf" on the tree, LimeWire can figure out if a part of the file is valid or corrupt and know whether or not it should erase that part (and try to redownload it).
For example, if the tree looked like: Code: A
/ \
B C
/ \ / \
D E F G and the file was 400KB, then the 'D' leaf might give info about the first 100k, 'E' might give 100k-200k, 'F' might describe 200k-300k, and 'G' would tell about 300k-400k.
If LimeWire had a corrupt byte at 304k, then the 'G' leaf would let LimeWire know that only that 100k was corrupt (instead of the whole file), and LimeWire would redownload that 100k. |