Re: java and svn_notify_func_t and large files

[Ich Selbst <ic...@gmx.ch>]

> Hi, I'm writting a java gui for subversion and would like to provide
> file transfer feedback to the user (e.g. when large files are transfered
> instead of simply showing the file name I would like to show the bytes
> being transfered as well).
> Currently the svn_client_checkout only calls the notify function once
> per file. In my previous life a wrote a CVS client and gui and we added
> a callback just for this purpose, to provide timely feedback for
> transfers of large files.

That would not only be good for a java GUI but for any other clients as
well.
I asked for that some time ago but since I asked on a weekend I got
no response - my fault :)

I would even go a step further and provide a two-way callback in which
the client could return FALSE to stop the command. Because right now
it's not possible to stop a command (at least without killing the thread
and leaving memory leaks).


> Are there plans to reduce the granularity of the callback and say have
> it called for every block size that is transfered? Or am I missing
> something in the API?

If something like that is in the API I haven't found it either.


Steve



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: java and svn_notify_func_t and large files

[Karl Fogel <kf...@newton.ch.collab.net>]

Jean-Michel Lemieux <fa...@lemieux-martel.ca> writes:
> I didn't find an issue for this, should I enter one?

I guess so!  To my surprise, a search for "cancel", "interrupt", and
"stop" in current open issues summaries yields nothing.

Thanks,
-K


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: java and svn_notify_func_t and large files

[Jean-Michel Lemieux <fa...@lemieux-martel.ca>]

> 
>
>I would even go a step further and provide a two-way callback in which
>the client could return FALSE to stop the command. Because right now
>it's not possible to stop a command (at least without killing the thread
>and leaving memory leaks).
>
I would go as far as saying that the ability to interrupt a command is 
more important than
increased callback frequency. When the "cancel" button is pressed the 
svn operation better
stop quickly :)

I didn't find an issue for this, should I enter one?
Jean-Michel


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: java and svn_notify_func_t and large files

[Karl Fogel <kf...@newton.ch.collab.net>]

"Ich Selbst" <ic...@gmx.ch> writes:
> > Are there plans to reduce the granularity of the callback and say have
> > it called for every block size that is transfered? Or am I missing
> > something in the API?
> 
> If something like that is in the API I haven't found it either.

I think there is a plan actually, and that Ben Collins-Sussman and Jon
Trowbridge may know more about it.  Guys, am I crazy?

-K

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org