[Bitten] Bitten Sprint: 24 April, 10am GMT

[Simon Cross]

I'm coordinating a Bitten sprint this Saturday.

A group of us are getting together in Cape Town as part of a local PUG
meeting [1] but we'll be on #bitten on IRC and we'd be delighted to
have people from elsewhere join in the ticket stomping. Our local PUG
meet up should include at least two people with commit access (me and
davidfraser) and osimons has said he might be on IRC a bit so we
shouldn't be short on people to commit patches.

The focus will be on dealing with the outstanding Milestone 0.6
tickets [1] but feel free to suggest other tickets for attention.

I should arrive at the PUG meeting at around 10:00am GMT. Our meetings
tend to be of somewhat flexible extent but I'm guessing there still be
a few or us around to push on into the early evening (with perhaps a
break for dinner at some point).

If you won't be at the sprint but have something you'd like looked at,
please file a ticket (and if possible mark it with the 0.6 Milestone
or assign it to me, i.e. hodgestar).

[1] http://ctpug.org.za/wiki/Meeting20100424
[2] http://bitten.edgewall.org/query?status=assigned&status=new&status=reopened&group=status&milestone=0.6

Schiavo
Simon

--
You received this message because you are subscribed to the Google Groups "Bitten" group.
To post to this group, send email to bit...@googlegroups.com.
To unsubscribe from this group, send email to bitten+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/bitten?hl=en.

[osimons]

On Apr 22, 11:25 pm, Simon Cross <hodges...@gmail.com> wrote:
> I'm coordinating a Bitten sprint this Saturday.
>
> A group of us are getting together in Cape Town as part of a local PUG
> meeting [1] but we'll be on #bitten on IRC and we'd be delighted to
> have people from elsewhere join in the ticket stomping.  Our local PUG
> meet up should include at least two people with commit access (me and
> davidfraser) and osimons has said he might be on IRC a bit so we
> shouldn't be short on people to commit patches.
>
> The focus will be on dealing with the outstanding Milestone 0.6
> tickets [1] but feel free to suggest other tickets for attention.
>
> I should arrive at the PUG meeting at around 10:00am GMT. Our meetings
> tend to be of somewhat flexible extent but I'm guessing there still be
> a few or us around to push on into the early evening (with perhaps a
> break for dinner at some point).
>
> If you won't be at the sprint but have something you'd like looked at,
> please file a ticket (and if possible mark it with the 0.6 Milestone
> or assign it to me, i.e. hodgestar).
>
> [1]http://ctpug.org.za/wiki/Meeting20100424

> [2]http://bitten.edgewall.org/query?status=assigned&status=new&status=re...

Thanks Simon, and thanks also to all that participated. Lots of good
patches getting reviewed, adjusted and committed. I'm very impressed
by all the effort put into the project by everyone in just this one
short day.

Unfortunately, I did not really manage to follow the sprint and its
discussions and changes. For the most part, you all seemed to have
managed quite well without me - and perhaps even because I wasn't
around :-) Reviewing the code now I just have a few comments and
follow-ups:

1) Walter, a great collection of patches included in the sprint. Your
patches are of high quality and I usually never have issues with them,
but let us just say that I would not object if you would tweak your
personal workflow to raise issues and post patches to tickets
earlier... :-)

2) The new charting still need some work. On first glance it looks
like Lint charting don't work, and I'm seeing errors such as "'dumps'
not being defined" and "'json.txt' template is malformed". That is no
doubt easily fixed, and let us follow details up on the ticket itself
(#426). However, the whole issue that Bitten now requires a json
encoder/decoder is not as clear-cut to me. json module is present in
Python 2.6, and for earlier versions simplejson module must be
installed (requirement). A large part of our users needs to install a
new new module when upgrading. Do we make the requirement explicit in
setup.py, or do we make it optional and just disable all charting if
json is not available? Do we rewire the communication format - like
keep the pre-existing XML format and instead reformat it to json on
the client before plugging it into graphs? I'm not quite sure what was
has been discussed.

Again, a big thanks to all that participated!


:::simon

[Simon Cross]

On Tue, Apr 27, 2010 at 1:20 PM, osimons <odds...@gmail.com> wrote:
> 2) The new charting still need some work.

Agreed. This is why the ticket is still open. :) I encouraged David to
commit the patch anyway because:

1) I felt it was important to get this patch into 0.6 so that
distributions that might want to package 0.6 can do so without
worrying about the licensing issues raised by the old charts.

2) It's a fairly big change that ideally needs feedback early adopters
before the 0.6 release and I don't think that will happen if the only
place the patch is available is in the ticket.

> However, the whole issue that Bitten now requires a json
> encoder/decoder is not as clear-cut to me. json module is present in
> Python 2.6, and for earlier versions simplejson module must be
> installed (requirement). A large part of our users needs to install a
> new new module when upgrading. Do we make the requirement explicit in
> setup.py, or do we make it optional and just disable all charting if
> json is not available? Do we rewire the communication format - like
> keep the pre-existing XML format and instead reformat it to json on
> the client before plugging it into graphs? I'm not quite sure what was
> has been discussed.

I'm happy to require simplejson for Python 2.4 and 2.5 and to update
the required Python version for the master from 2.3 to 2.4 but if
someone feels up to writing the Javascript needed to pull in the data
from XML instead I'm fine with that too. A third option is to pull in
the JSON encoder from simplejson into Bitten (about 500 lines of code)
and use that if neither json nor simplejson is present (I'm willing to
do this if it's felt this is the right direction).

> Again, a big thanks to all that participated!

It was a lot of fun. Thanks for checking through all the patches!

Schiavo
Simon

[Remy Blank]

Simon Cross wrote:
> I'm happy to require simplejson for Python 2.4 and 2.5 and to update
> the required Python version for the master from 2.3 to 2.4 but if
> someone feels up to writing the Javascript needed to pull in the data
> from XML instead I'm fine with that too. A third option is to pull in
> the JSON encoder from simplejson into Bitten (about 500 lines of code)
> and use that if neither json nor simplejson is present (I'm willing to
> do this if it's felt this is the right direction).

(I haven't followed the JSON discussion closely, so this might be
irrelevant.)

Trac has a JSON encoder on trunk, which handles most simple cases. It is
used to serialize e.g. ticket field data for the custom query page. See
the bottom of (lines 282 and following):

http://trac.edgewall.org/browser/trunk/trac/util/presentation.py

Maybe this is good enough for this use case?

-- Remy

[Simon Cross]

On Tue, Apr 27, 2010 at 4:15 PM, Remy Blank <remy....@pobox.com> wrote:
> Trac has a JSON encoder on trunk, which handles most simple cases. It is
> used to serialize e.g. ticket field data for the custom query page. See
> the bottom of (lines 282 and following):
>
>  http://trac.edgewall.org/browser/trunk/trac/util/presentation.py
>
> Maybe this is good enough for this use case?

Reports send lists and dictionaries of floats and strings and it looks
like all those are supported by the Trac mini-json encoder. Thanks for
the heads up!