Nils was kind enough to provide details (and I imagine help along the
way as well): http://bugs.code.downthemall.net/trac/ticket/413#comment:4
Replying to ant:
Nils, will you still accept a patch for this?
Sure.
If so, could you provide some guidance on where to start?
The verification code lives in trunk/chrome/content/dta/manager/verificator.js
There is already whole file checksumming support, so is
checksumming each chunk the difficult part, or is it changing DTA to
re-request a chunk when the checksums don't match?
There are multiple parts needed to implemented:
* Parsing the metalink trunk/chrome/content/dta/manager/metalinker.js
* Storing stuff, so that is survives between sessions. This has to
be backwards-compatible, i.e. loading old queue items lacking the new
stuff has to be successful. See the .toSource methods
trunk/chrome/content/dta/manager.js
* Calculating the checksums trunk/chrome/content/dta/manager/verificator.js
* Asking the user what he'd like to do in case of error.
* Setting up the retry chunks in case of error and starting the
download again(in the quest of uncoupling please do not mess with
QueueItem? data from verificator or similar, but instead provide some
public methods in QueueItem? that can be used for this matter).
I know that esp. the verificator/metalinker code is not very well
designed and too tightly coupled. Any implementor should try not to
add even more coupling, however.
See CreatingPatches.
It would be great if a patch would do the following, however this is optional:
* split verificator into a re-usable js module and some "user" code.
* split metalinker into a re-usable js module (parser/validation),
interface code (the selection dialog) and manager.js stuff.
On Fri, Jul 3, 2009 at 9:14 AM, Hampus Wessman<[email protected]> wrote:
> True. I'll look into it! Shouldn't be that hard.
> / Hampus
>
>
> Ant Bryan skrev:
>
> thanks Hampus! I think it's one of the most useful too! so much so,
> that I think it may be more important than ME2. not to knock ME2! :)
> but ME1 is great enough and gets us by fine (not that the new features
> would be good too).
>
> I wonder how much work this actually is?
>
> On May 9, 4:29 am, Hampus Wessman <[email protected]> wrote:
>
>
> That would be really nice. Chunk checksums is one of the coolest
> metalink features, IMO!
>
> I could do this after I've finished Metalink Editor 2.0, but that might
> take some time. It would be even better if someone else did it before!
>
> Hampus Wessman
>
> On Fri, 8 May 2009 17:31:15 -0400
>
> Anthony Bryan <[email protected]> wrote:
>
>
>
> I don't have any numbers to back this up, but I'd guess that most
> metalink users are taking advantage of them with DownThemAll!.
>
>
> DTA is easy to use, pleasant looking, & just a Firefox extension install
> away.
>
>
> the only metalink related feature it lacks is repairing a download if
> the metalink has chunk checksums. it already verifies checksums for
> the whole file.
>
>
> I think this should be one of our higher priorities, because of the
> amount of people that DTA reaches. for instance, openSUSE uses
> metalinks with chunk checksums, but if a downloader uses a metalink
> client that doesn't support them then the downloader could get
> frustrated if there's an error in their big download...
>
>
> we have this on our ideas page & here's the request in the DTA bugtracker
>
>
> http://groups.google.com/group/metalink-discussion/web/gsoc-ideas
> http://bugs.code.downthemall.net/trac/ticket/413
>
>
> does anyone with JS skills want to work on this? Nils had said he
> would accept a patch.
> maybe someone from the openSUSE community could do it?
>
>
> --
> (( Anthony Bryan ... Metalink [http://www.metalinker.org]
> )) Easier, More Reliable, Self Healing Downloads
>
>
>
>
> >
>
--
(( Anthony Bryan ... Metalink [ http://www.metalinker.org ]
)) Easier, More Reliable, Self Healing Downloads
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Metalink Discussion" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/metalink-discussion?hl=en
-~----------~----~----~----~------~----~------~--~---
