Re: Problem with TSVNCache when file is deleted

Jens Scheidtmann wrote:
> Stefan Küng <tortoisesvn@gmail.com> writes:
>
>>> Why does TSVNCache begin to crawl in the first place?
>> As soon as it detects a delete, it has to crawl again. The deleted file
>> might be versioned, and therefore it must know if it has to change the
>> status of the parent folder to 'missing'.
>
> Yeah, sure.
>
> But the thing is, that it starts on the file before
> that. Let me summarize what I see in the File Monitor log:
>
> 1. IDE queries <filename> (10 MB).
> 2. IDE starts saving <tmp-file> (takes a while)

TSVN gets a notification that a file got modified. It now has to check
* if the file is versioned
* if it is inside a versioned folder, it has to re-fetch the status of
that folder and all files in it (takes the same disk access and time as
fetching the status of a single file - there's an issue about this in
the Subversion issue tracker...)

> 3. TCache queries <tmp-file>

So the file exists. And it is inside a versioned folder. So crawling starts.

> 4. TCache starts crawling (multiple/all files in directory are
> accessed, Note: the directory contains many small files and 1-5
> large files)
> 5. TCache starts copying <filename> to .svn/tmp/<file>.x.tmp.tmp
> (could one get rid of this?)

Sorry, no. Subversion has to create a temp file of your working copy
file to check if there are modifications. The reason a temp file is
needed are the possible expanded keywords. So to get the status of that
file, the temp file is needed.

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