continuing download of a file once you find a larger one with same name

It sometimes happens to me that I find a good video but it stops in the middle. Later I find one with exactly the same name but a larger file size.

I thought it would be great if the old file could be automatically detected and the new one appended to it, thus reducing download.

I've used gnotella before and used to manually add my old file in the downloads.txt file to trick it into doing this, but XoloX resumeinfo.txt file is just a bit more tricky, with all the parts and so on... :-)

Thanx,

Björn

[The Seeker]

With movies sometimes a larger file with the same name is not really the same... and with MP3s the feature you are requesting will almost always result in corruption...

\•/Disclaimer\•/
This is untested.
If it doens't work
Don't complain.
/¯\Disclaimer/¯\

If you want to trick XoloX into resuming , try adding the following to your resumeinfo.txt:


------------
JOB
# <- A job # that isn't in use.
1
blah blah blah <- Search query to find the file, eg: "matrix divx" (no quotes)
x:\path\xolox\downloads\filename.ext <- use exact file name to download
######### <- new (larger) file size
2
00000000 <- CRC (dissabled, so it's 0's)
# <- # of failed segments (set to 0, unless you like seeing a number there)
NONSAVE <- If other downloads have something else here, use that instead.
SEGMENT <- start segment info
1 <- segment 1
x:\Path\XoloX\temp\JOB-Segment#-1.xlx <- # = job number
0 <- denotes start of file
38731181 <- length of already downloaded segment (in bytes) minus 2048
------------

Then just copy/move your partial file to x:\path\XoloX\Temp\JOB-Segment#-1.xlx and start XoloX.

Thank you for explaining the resume-info.txt file to me!
It now does look much less cryptic!

Björn

Add date of last succesfull resuming to the resumeinfo.txt 'JOB' information (see format above, not included yet).

So a tweaking/maintaining tool for never finishing or never starting downloads would become possible. A few months old '0% downloads' waste requery traffic and flood the GUI at the moment.

[ThePowerTool]

Originally posted by The Seeker
With movies sometimes a larger file with the same name is not really the same... and with MP3s the feature you are requesting will almost always result in corruption...

On larger file, yep--I've seen that. Sometimes it's a completely different file. Sometimes it's exactly what you really want.

On the 2nd part I have to question your conclusion. If you have a partial file and a completed file where the partial file represents a portion of the complete file, you should be able to CRC blocks until failue and then pick up with a "resume" function. It wouldn't have to result in corruption.

This would also be extremely handy in cases where you are downloading a partially downloaded file and then find the complete version. The real trick would be modifing the UI so the user can indicate both files should be the same file and then selecting the filename to be used for the save.

ThePowerTool