Sorry Juliusz, I don't think this issue is completely dead.

We're not trying to take that lock anymore, but it still fails when trying to
create the tentative pending file (why would it be trying to do that? it's a dry
run).

I think this means there's a flaw in my test script.  Assigning to myself to
look at the test :-/

On Tue, May 20, 2008 at 11:22 AM, Eric Kow <bugs@darcs.net> wrote:

>
> New submission from Eric Kow <eric.kow@gmail.com>:
>
> Sorry Juliusz, I don't think this issue is completely dead.
>
> We're not trying to take that lock anymore, but it still fails when trying
> to
> create the tentative pending file (why would it be trying to do that? it's
> a dry
> run).

In a dry run I might expect to get a summary of what works and what
doesn't.  That is, failure is noted but not fatal.  So I might expect darcs
to tell me that a lock cannot be taken and that I would need it for a wet
run (is that the right term?).

It probably makes your changes more difficult, but I thought I'd mention it.

Jason

Attachments

• unnamed

Test case re-submitted/corrected.  It now fails like it should, and Eric
understands shell scripts a little bit better.

Ah! David fixed my fix.
The following patch updated the status of issue855 (and issue864) to be resolved:

* resolve issue855: fix bug where we were calling revertRepository when running
with --dry-run.