A contributer appears

4 views
Skip to first unread message

Jesse Hallett

unread,
Dec 7, 2008, 1:45:53 PM12/7/08
to za...@googlegroups.com, Igal Koshevoy
We have an active contributer! You can see Simon Chiang's zaml for at:

http://github.com/bahuvrihi/zaml/tree/master

or you can see a graph of all branches of zaml, at this point, just the two, at:

http://github.com/hallettj/zaml/network

I am about to look at Simon's additions and will work on merging stuff.


---------- Forwarded message ----------
From: GitHub <nor...@github.com>
Date: Sun, Dec 7, 2008 at 8:04 AM
Subject: [GitHub] bahuvrihi sent you a message
To: hall...@gmail.com


bahuvrihi wants you to pull from bahuvrihi/zaml at master

Body: Jesse - I was messing around with ZAML and thought you might
like a look at my changes. I added a test suite and made the output
identical to YAML for all the test cases. Originally I had hoped to
make some small quick loader as well, but it looked like more than I
wanted to tackle. Nice project. Regards,
- Simon

View repository: http://github.com/bahuvrihi/zaml/tree/master

Markus

unread,
Dec 7, 2008, 2:15:21 PM12/7/08
to simon.a...@gmail.com, Igal Koshevoy, za...@googlegroups.com
Simon --

> Originally I had hoped to make some small quick loader as well, but
> it looked like more than I wanted to tackle.

What was the motivation for writing a loader? Specifically, are you
running into performance problems with YAML's loader, or cases where it
doesn't do what you want, or?

The reason I asked is that the way we wandered into re-writing the
emitter portion to produce ZAML was to address a specific use case which
was causing problems for another project (Ian McIntosh's Luz, see his
blog at: http://gnomecoder.wordpress.com/luz/).

If you've got a particular use case in mind for a loader, we may be able
to help, offer ideas, or at least commiserate.

-- MarkusQ

P.S. Thanks for writing the chemical constants thing (I'm pretty sure it
was you). I grabbed it a while back for a quick prototype I was
throwing together and everything worked just like I'd expect it to.
Very intuitive API--good job.


Jesse Hallett

unread,
Dec 7, 2008, 11:31:01 PM12/7/08
to za...@googlegroups.com, simon.a...@gmail.com, Igal Koshevoy
Simon,

Thanks for putting in all of that work on Zaml! I have merged your
code into a branch in our project called 'proposed':

http://github.com/hallettj/zaml/commits/proposed

I would like to give the other Zaml contributers a chance to look at
the changes before they are pushed to the main branch. I added some of
my own changes to the same branch, such as bringing back Time and Date
serialization.

Also, if you would like to be involved in Zaml discussions, you are
welcome to join our mailing list:

http://groups.google.com/group/zaml

Cheers,
Jesse

Markus

unread,
Dec 8, 2008, 1:52:12 AM12/8/08
to za...@googlegroups.com
Jesse --

> ...bringing back Time and Date serialization.

Could you refresh my memory on this? Way Time and Date in and taken out
or...?

-- Markus

P.S. Proposed looks interesting. It also seems to be about 25% faster,
which is extra coolness. I'd like to look at it a bit more when I'm
awake, but it looks pretty good.

Jesse Hallett

unread,
Dec 8, 2008, 2:04:28 AM12/8/08
to za...@googlegroups.com
I think Simon commented out the Time and Date stuff in his branch.
Probably because the tests for those methods were failing because the
output didn't match YAML's. I uncommented the code and the tests, and
I made some modifications to the code so that the serialized output
from ZAML matches that of YAML. I removed the '!timestamp' prefix from
Date serialization because that reads back into YAML as a Time object.
And the fractional seconds and timezone offset didn't seem to be
working on my machine for Time serialization, so I rewrote that part.
Reply all
Reply to author
Forward
0 new messages