This is an extremely good suggestion.

As everybody knows, there are some file servers that have restrictions about the time one (or Jdownloader in this case) has/have to wait between every file download in a pile of files to be downloaded, in order to continue to the next file. And also the restriction about you are only able to download file by file.

I found really useful to have many mirrors by diferent file servers. When you download file.01.rar, the file.02.rar could not be download from the same server until a several minutes have lapsed, lets say 5 minutes or 10 minutes (depending on the files server rules). Now the good thing about having more mirrors is that Jdownloader can automatically go to the next mirror to try download the same file.02.rar instead of wait the time to get the file from the same server as previous downloaded file.

So, Jdownloader as you say, can manage mirrors when they are in the same package. I have been using this very well as I explained before. However I think there are some inconvenients that let me suggest the mirrors could be better managed by Jdownloader.

The first inconvenient is that when you add files from different file servers, the package sums the total size. So for a package weight of 5 GB, if you add 2 mirrors for every file in the package, it will show a weight of 15 GB (5 GB + 5 GB * 2 mirrors). Now the second inconvenient is that the size will also modify the estimaded left time to complete the download. If a 5 GB package could be downloaded with a 2 Mbps connection in ~ 6 hours, with 2 aditional mirrors, Jdownloader will show ~ 18 hours as estimated left time to complete the package download.

A best way to manage mirrors, I think could be in the form of a "mirrors tab" for every file, instead of the same file showed many times (for every files server) in the downloading package files list.

I hope this could help to evaluate the applicability of the function I suggest to manage mirrors.

Uriel