Re: [TSVN] Comment on 1895

Will Dean wrote:

> I've just updated to 1895 - you've done exactly what I intended, but I
> do feel we should check that cache timeout mechanism in
> SVNFolderStatus::GetCachedItem. I am worried that
> excessive/uncontrolled cache lifetime is potentially a nasty problem,
> especially while we don't have any cache invalidation mechanism linked
> from TortoiseProc.
>
> But maybe you have a rationale for omitting the timeout test which I've
> missed...?

No. I'm adding the cache timeout (the same as in GetFullStatus()) later.
But right now I'm busy with other things.

> If you include the timeout test in GetCachedItem, then you can replace
> all the cache test code in GetFullStatus with a call to GetCachedItem.
>
> Indeed, if the GetCachedItem call was duplicated into ColumnProvider, we
> could drop it from GetFullStatus altogether.

I don't think we should do that. The same timeout is needed as in
GetFullStatus() so it will work correctly. Otherwise, the cache never
times out and we would have a problem.

> As you've said, this optimisation was for versioned files only, and is
> actually a slight pessimisation for unversioned files. We should
> probably be checking HasSVNAdminDir early as well, as this can run
> without doing any file-system access.

I think I have some time tomorrow to look into that.

Stefan

-- 
        ___
   oo  // \\      "De Chelonian Mobile"
  (_,\/ \_/ \     TortoiseSVN
    \ \_/_\_/>    The coolest Interface to (Sub)Version Control
    /_/   \_\     http://tortoisesvn.tigris.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tortoisesvn.tigris.org
For additional commands, e-mail: dev-help@tortoisesvn.tigris.org