On Wed, Apr 14, 2010 at 15:53, Stefan Sperling <s...@elego.de> wrote:
> On Wed, Apr 14, 2010 at 12:21:02PM -0400, Greg Stein wrote:
>> On Tue, Apr 13, 2010 at 15:45, <s...@apache.org> wrote:
>> >...
>> > +++
>> > subversion/branches/svn-patch-improvements/subversion/libsvn_client/patch.c
>> > Tue Apr 13 19:45:17 2010
>> >...
>> > + /* Finally, delete empty directories. */
>> > + for (i = 0; i < empty_dirs->nelts; i++)
>> > + {
>> > + const char *empty_dir;
>> > +
>> > + svn_pool_clear(iterpool);
>> > +
>> > + if (ctx->cancel_func)
>> > + SVN_ERR(ctx->cancel_func(ctx->cancel_baton));
>
> We do it here ^^^
>
>> > +
>> > + empty_dir = APR_ARRAY_IDX(empty_dirs, i, const char *);
>> > + if (ctx->notify_func2)
>> > + {
>> > + svn_wc_notify_t *notify;
>> > +
>> > + notify = svn_wc_create_notify(empty_dir, svn_wc_notify_delete,
>> > + iterpool);
>> > + (*ctx->notify_func2)(ctx->notify_baton2, notify, iterpool);
>> > + }
>> > + if (! dry_run)
>> > + SVN_ERR(svn_wc_delete4(ctx->wc_ctx, empty_dir, FALSE, FALSE,
>> > + NULL, NULL, /* suppress cancellation */
>> > + NULL, NULL, /* suppress notification */
>> > + iterpool));
>>
>> I can see suppressing notification, but why cancellation?
>
> See above.
>
> It's a bit arbitrary how cancellation is done in the patch code.
> It may need a bit of review in this area.
I saw the above, but svn_wc_delete4() also has plenty of loops.
Passing along the cancellation will provide for faster cancellation.
Cheers,
-g
