Re: [PATCH v2] Reject checkouts to existing directory

On Fri, Oct 28, 2016 at 02:49:04AM +0200, bert_at_qqmail.nl wrote:
> Just checking the log message: This changes 90% of our testsuite to use ‘svn co --force’, by changing the standard checkout function in actions.py that is used by virtual all tests.
>
> If that is really necessary I would call this a breaking change.
>
> It also changes a few of the most core basic tests… That tells me that some very common usecases will change.
>
>
> If we have to change the behavior of 90% of our tests for this, I don’t think we can just commit it as any basic minor test.
>
> In the summary I don’t see the support for restartability of a broken checkout. See the documentation of svn checkout in the svn-book. (http://svnbook.red-bean.com/nightly/en/svn.ref.svn.c.checkout.html)
> [[
> If you interrupt a checkout (or something else interrupts your checkout, such as loss of connectivity, etc.), you can restart it either by issuing the identical checkout command again or by updating the incomplete working copy:
> $ svn checkout file:///var/svn/repos/test mine
> A mine/a
> A mine/b
> ^C
> svn: E200015: Caught signal
> $ svn checkout file:///var/svn/repos/test mine
> A mine/c
> ^C
> svn: E200015: Caught signal
> $ svn update mine
> Updating 'mine':
> A mine/d
> Updated to revision 20.
> $
> ]]
> Resuming will now require a –force, which also triggers different behavior in other parts of checkout.
>
> Which would require me to update dozens of buildscripts that worked just fine since 1.0, if released.

I guess it makes sense to allow for resuming canceled checkouts.
After all it should not be that hard to implement. Will implement
in the next revision.

[snip]

Regards
Patrick

-- 
Patrick Steinhardt, Entwickler
elego Software Solutions GmbH, http://www.elego.de
Gebäude 12 (BIG), Gustav-Meyer-Allee 25, 13355 Berlin, Germany
Sitz der Gesellschaft: Berlin, USt-IdNr.: DE 163214194
Handelsregister: Amtsgericht Charlottenburg HRB 77719
Geschäftsführer: Olaf Wagner