The Plone Collective

On 1/11/07, Lennart Regebro <regebro@...> wrote:
> It turns out that VEVENTS doesn't need either duration or dtend. I
> made a quick fix to the Calendaring marshaller to set the end of
> events to the same as the start if the vevent doesn't have a dtend.

Could you please add a test so it does not regress on the next
refactoring? Thanks!

-- 
Sidnei da Silva
Enfold Systems                http://enfoldsystems.com
Fax +1 832 201 8856     Office +1 713 942 2377 Ext 214

On 1/11/07, Sidnei da Silva <sidnei@...> wrote:
> On 1/11/07, Lennart Regebro <regebro@...> wrote:
> > It turns out that VEVENTS doesn't need either duration or dtend. I
> > made a quick fix to the Calendaring marshaller to set the end of
> > events to the same as the start if the vevent doesn't have a dtend.
>
> Could you please add a test so it does not regress on the next
> refactoring? Thanks!

Well, I tried, but the fact is most tests fail for me, and the test I
added for this doesn't run at all. I don't know if I run them the
wrong way or something...

   bin/zopectl test -m Products.Calendaring

Should work, right?

Zope 2.9.4, Plone more or less trunk.

-- 
Lennart Regebro: Python, Zope, CPS, Plone.
+33 661 58 14 64

Yeah, I doubt the tests will run with Plone trunk. That's not to say
they can't be fixed. It looks like you just need to update simple
stuff, like loginPortalOwner -> loginAsPortalOwner. Would you mind
doing that?

On 1/17/07, Lennart Regebro <regebro@...> wrote:
> On 1/11/07, Sidnei da Silva <sidnei@...> wrote:
> > On 1/11/07, Lennart Regebro <regebro@...> wrote:
> > > It turns out that VEVENTS doesn't need either duration or dtend. I
> > > made a quick fix to the Calendaring marshaller to set the end of
> > > events to the same as the start if the vevent doesn't have a dtend.
> >
> > Could you please add a test so it does not regress on the next
> > refactoring? Thanks!
>
> Well, I tried, but the fact is most tests fail for me, and the test I
> added for this doesn't run at all. I don't know if I run them the
> wrong way or something...
>
>    bin/zopectl test -m Products.Calendaring
>
> Should work, right?
>
> Zope 2.9.4, Plone more or less trunk.
>
> --
> Lennart Regebro: Python, Zope, CPS, Plone.
> +33 661 58 14 64
>


-- 
Sidnei da Silva
Enfold Systems                http://enfoldsystems.com
Fax +1 832 201 8856     Office +1 713 942 2377 Ext 214

On 1/17/07, Sidnei da Silva <sidnei@...> wrote:
> Yeah, I doubt the tests will run with Plone trunk. That's not to say
> they can't be fixed. It looks like you just need to update simple
> stuff, like loginPortalOwner -> loginAsPortalOwner. Would you mind
> doing that?

OK, I'll look at it.

-- 
Lennart Regebro: Python, Zope, CPS, Plone.
+33 661 58 14 64

Well, everything that has anything to do with timezones fail for me, I
don't understand why, and there are some sort of permission problem I
don't understand.

But the test I added works now. I changed loginPortalOwner to
loginAsPortalOwner, and I also changed all the tests that expected
VERSION:1.0 to expect VERSION:2.0.

-- 
Lennart Regebro: Python, Zope, CPS, Plone.
+33 661 58 14 64

On 1/17/07, Lennart Regebro <regebro@...> wrote:
> Well, everything that has anything to do with timezones fail for me, I
> don't understand why, and there are some sort of permission problem I
> don't understand.
>
> But the test I added works now. I changed loginPortalOwner to
> loginAsPortalOwner, and I also changed all the tests that expected
> VERSION:1.0 to expect VERSION:2.0.

Sounds good enough for me. Nate had complained about the timezones as
well I believe. Someone might even have submitted a patch, but I can't
find.

Thank you very much!

-- 
Sidnei da Silva
Enfold Systems                http://enfoldsystems.com
Fax +1 832 201 8856     Office +1 713 942 2377 Ext 214

On 1/17/07, Sidnei da Silva <sidnei@...> wrote:
> Sounds good enough for me. Nate had complained about the timezones as
> well I believe. Someone might even have submitted a patch, but I can't
> find.

After digging deeper, it's possible that the errors in timezone is
because pretty much ever timezone test assumes that you are in Brazil.
It's also impossible to change the local timezone during the tests,
because the dateutils module initialize the timezone during
initializing, in several different places, and there seem to be no
easy way to override it.

So, all in all, it's for the moment impossible to know if there is a
bug in this or not. :-)

I hate timezones. I propose the whole world start using a timekeeping
system that has no relation to the sun, so we don't have to bother
about this crap. Metric time rules! :)

-- 
Lennart Regebro: Python, Zope, CPS, Plone.
+33 661 58 14 64

> I hate timezones. I propose the whole world start using a timekeeping
> system that has no relation to the sun, so we don't have to bother
> about this crap. Metric time rules! :)

Yeah, the Internet Time:
http://www.timeanddate.com/time/internettime.html
-- 
Sidnei da Silva
Enfold Systems                http://enfoldsystems.com
Fax +1 832 201 8856     Office +1 713 942 2377 Ext 214

On 1/17/07, Sidnei da Silva <sidnei@...> wrote:
> > I hate timezones. I propose the whole world start using a timekeeping
> > system that has no relation to the sun, so we don't have to bother
> > about this crap. Metric time rules! :)
>
> Yeah, the Internet Time:
> http://www.timeanddate.com/time/internettime.html

Still based on days. No good. :-)

-- 
Lennart Regebro: Python, Zope, CPS, Plone.
+33 661 58 14 64

On 1/17/07, Lennart Regebro <regebro@...> wrote:
> On 1/17/07, Sidnei da Silva <sidnei@...> wrote:
> > > I hate timezones. I propose the whole world start using a timekeeping
> > > system that has no relation to the sun, so we don't have to bother
> > > about this crap. Metric time rules! :)
> >
> > Yeah, the Internet Time:
> > http://www.timeanddate.com/time/internettime.html
>
> Still based on days. No good. :-)
>

Also clearly a marketing ploy by Swatch, although it has some sensible
points, this site makes several good points, e.g. that ".beat" is a
stupid name for a unit of time, and has no relation to seconds, which
are the basis of metric time.

The crowning glory is that it also does not correspond to GMT in any
sensible way, so although it would be easier to convert between
internet time(s), converting a normal time to internet time sounds
like nothing but hell. ;)

-- 
Justizin
Director of Application Architecture
Auxilium Group / Gnudyne(tm)