On Tue, Feb 14, 2012 at 9:43 AM, Marcus D. Hanwell
mhanwell@gmail.com wrote:
On Tue, Feb 14, 2012 at 9:08 AM, David Lonie loniedavid@gmail.com wrote:
On Mon, Feb 13, 2012 at 4:16 PM, Marcus D. Hanwell
mhanwell@gmail.com wrote:
I think it would be great to get a 1.1.0 release out sometime in the
near future. I was looking at a couple of compilation failures on the
dashboards, and have addressed some of them here,
http://review.source.kitware.com/#/t/36/
Albert just submitted a patch I merged to address something introduced
by the terachem input deck generator. The Mac dashboard looks similar,
with a fix in the code I submitted. I was going to revisit the extra
attributes being removed from the fragment wrapping, but I think the
rest of that patch is ready to go in.
Are there other outstanding problems that need to be resolved? I need
to take a look at the Windows submission as it is reporting a
conflict, it would be good to get some new binaries out for Windows.
I have a topic branch I’d like to merge in before the next release:
http://review.source.kitware.com/#/t/66/
Sounds reasonable - I was just taking a look at this topic. I was
hoping to work through all of the commits, and decide if they should
be merged, abandoned or fixed up and merged.
I weeded out the changes that needed to be abandoned (most of that
topic was part of an older topic that would have been redundant).
It modifies how the initial viewpoint is set for crystals and
orthographic projections.
It’s also about time for bug-fixing weekend to catch up on the
tracker. Does anyone have time for this in the near future, e.g.
pre-release? I might be able to find a few hours this weekend to work
on this.
I am traveling this weekend unfortunately, but will try to make some
time before I leave. I have been going through a few of the open bugs
as and when I have had time.
Ok, it looks like I’ll have some time to spend on this over the
weekend. Just make sure that the bugs you’ve been working on are
assigned to you so we don’t duplicate effort.
There was also a post about adding a “submit backtrace on crash”
feature – this may be a good project to focus on, too.
Agreed, but I would rather not delay a release for it. I think we are
well overdue. It would be good for us to make more regular releases
and get feedback. I keep showing off Avogadro, and having people ask
why they can’t use feature x, and it is because we haven’t made a
1.1.0 release with binaries. I need to get back on the Windows
packaging too, which could do with being rewritten but at least needs
a little patching up. That would allow for the easy creation of
nightly binaries on Windows at least.
Yes, this certainly shouldn’t delay a release. Let’s get one out ASAP,
we can always release again in a month or so if we’re ready. I’ve been
building my XtalOpt windows packages against master and pulling most
of the open topics on Gerrit, so windows builds should still be fairly
easy.
Last I checked, Windows was still the one with the largest user base,
and the most need for testing as it has more crashers lurking than the
other two platforms.
Agreed – I like Geoff’s idea of “fuzz” testing the application with
random user events, just to see what can be broken. IMO we should
start to dedicate more time to writing unit tests and focusing on
stability. We can do this as we clear out the tracker by adding
failing test cases for the bugs so we can regression test, but this is
of course difficult for bugs that require GUI interaction.
How does the QtTesting code you added work? Is there an example of how
to use it? IIRC that should help test GUI bugs.
Dave