Shawn Walker wrote:
On 02/11/10 10:55 AM, John Rice wrote:
Shawn Walker wrote:
...
So one thing that isn't clear to me is how the background threads PM
uses to retrieve package information via api_obj.info() work.
Specifically, if it tries to retrieve some package information, and it
can't because an ImageLockedError() gets raised (which isn't totally
accurate, what's really locked is the ImageInterface because there's
another transport operation in progress, but it's effectively the same
to consumers), how soon will it retry again or will it just give up
completely?
It just gives up, this is a low value operation, if it fails its not
catastrophic to the user. We do raise the image lock exception for the
user if they are trying to do something such as search, install, remove
and so on which is more important to flag directly to them. We are also
making judicious use of the busy cursor to try and prevent the user
getting into this situation in the first place, such as when you are
clicking on a row in the main list view.
So then, if it fails, that background thread will not try and try
again (or the caller)?
No it shouldn't, it should fail in the __show_info call(), catch the
exception and the __update_package_info() should output a suitable
message to the Details panel:
"Information not available for this package..."
JR
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
