Re: svn commit: r33863 - trunk/subversion/libsvn_wc

[Erik Huelsmann <eh...@gmail.com>]

On Thu, Oct 23, 2008 at 4:58 PM,  <hw...@tigris.org> wrote:
> Author: hwright
> Date: Thu Oct 23 07:58:58 2008
> New Revision: 33863
>
> Log:
> Followup to r33831: Just use the system temp directory for a temporary file.
> In this case, the file isn't being atomically moved anywhere, just opened and
> read again, so we don't really care where it lives.

But, why not just send it into a stringbuf directly? I mean, we're
diffing memory structures, writing it to disk in order to have it be
read back 15 statements later into a stringbuf (which is used for
later manipulation...). It would seem more efficient to skip the disk
step.

Bye,

Erik.

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

Re: svn commit: r33863 - trunk/subversion/libsvn_wc

["Hyrum K. Wright" <hy...@mail.utexas.edu>]

Erik Huelsmann wrote:
> On Thu, Oct 23, 2008 at 4:58 PM,  <hw...@tigris.org> wrote:
>> Author: hwright
>> Date: Thu Oct 23 07:58:58 2008
>> New Revision: 33863
>>
>> Log:
>> Followup to r33831: Just use the system temp directory for a temporary file.
>> In this case, the file isn't being atomically moved anywhere, just opened and
>> read again, so we don't really care where it lives.
> 
> But, why not just send it into a stringbuf directly? I mean, we're
> diffing memory structures, writing it to disk in order to have it be
> read back 15 statements later into a stringbuf (which is used for
> later manipulation...). It would seem more efficient to skip the disk
> step.

Agreed that it's possible, but more than I want to think about right now.  I'm
just a remove-the-deprecated-API monkey. ;)

-Hyrum