RE: [TSVN] RFC: New cache scheme

At 09:56 21/01/2005 +0000, you wrote:

>This would be misleading - I can imagine what would happen if tsvncache
>became unavailable (for whatever reason and my wc was all marked unversioned
>- you're asking for a potentially large number of 'bug' reports here.

'Marked unversioned' was just my confusing way of saying 'doesn't get an
overlay'

Anyway, if the cache has become unavailable, then that *is* a bug, so a bug
report doesn't seem out of place...

>If you do what Toby suggests and have tsvncache settings tab you can also
>fail gracefully if the cache is not present - if the cache is stopped or
>doesn't respond to it's ipc calls then you stop trying to use it and set the
>status of objects the slow but reliable way.

That would involve the shell extension having all the SVN status code in it
again though, which would be a pity.

My only point about surviving cache process stops and starts was not that
it's got any particular relevance to normal use, just that it makes
development of the caching software *enormously* easier than with the shell
extension.

I don't see any particular reason to think that it would be less reliable
than what's happening at the moment.

Will

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tortoisesvn.tigris.org
For additional commands, e-mail: dev-help@tortoisesvn.tigris.org
Received on Fri Jan 21 11:10:00 2005