Download.pas

   1 unit Download;
   2 {manage downloads}
   3
   4 INTERFACE
   5 {todo}
   6
   7 IMPLEMENTATION
   8 {
   9  Same idea here as upmgr. Have an tAggr for each peer reporting speed and
  10  tJob for each file request saving to file and reqesting missed segments.
  11  The aggr should have limit of paralele jobs. Jobs will first be linked in
  12  tAggr queue and then started as slots become available.
  13
  14  After node restart, notify requester with a No-Source error. But should
  15  not be forgotten. More advanced DM could consult CHK or Category the file
  16  was found.
  17 }
  18
  19
  20
  21 END.