#3552 NORM First D: unfinished downloads leave broken entries in journal

Zarro Boogs per Child bugtracker at laptop.org
Tue Sep 25 11:17:49 EDT 2007


#3552: unfinished downloads leave broken entries in journal
-------------------------------+--------------------------------------------
  Reporter:  danw              |       Owner:  tomeu                 
      Type:  defect            |      Status:  new                   
  Priority:  normal            |   Milestone:  First Deployment, V1.0
 Component:  journal-activity  |     Version:                        
Resolution:                    |    Keywords:                        
  Verified:  0                 |  
-------------------------------+--------------------------------------------

Comment(by tomeu):

 Replying to [comment:3 Eben]:
 > Does this mean that we have support for resumable downloads now?  Will
 resuming the Browse session the entry was created by allow the download to
 continue from its previous state?

 Nope, not yet :(

 > If yes:  Awesome, this is what we are shooting for.  We'll need to
 design it so that it can be determined that it isn't currently in
 progress, and also create a way to resume it (thus resuming, I'd assume,
 the Browse entry that created it).  Perhaps we'll need a pause/play button
 in addition to the cancel button within the entry for this.

 We should have another ticket with this spec'ed.

 > If no: We'll definitely not want to leave the broken download lying
 around if it can't be resumed.  Browse should take care of removing
 incomplete downloads automatically (see caveat below).

 Well, in this case, the Journal will ask the DataStore to remove the
 entry, Browse would notice, and the download would be cancelled. Same
 thing from the user POV, I think.

 > In either case: We should use a non-modal dialog to indicate that there
 are incomplete downloads when stopping a Browse session.  We could provide
 options to "Pause download" (assuming we have resume support), "Cancel
 download", and "Continue download."  The wording for these isn't good
 enough, but this illustrates the ideas.

 We were thinking in #3551 about letting the Browse process running in the
 background until all downloads finish, but close the activity window
 immediately after the user requests it. We need to see if there are no
 security concerns with that and how that could distort the memory feedback
 in the donut.

-- 
Ticket URL: <https://dev.laptop.org/ticket/3552#comment:4>
One Laptop Per Child <https://dev.laptop.org>
OLPC bug tracking system



More information about the Bugs mailing list