Mailing List Archive: Python: Python

Keg - A python web framework

Index | Next | Previous | View Threaded

eghansah at gmail

Aug 3, 2008, 2:57 PM

Post #1 of 7 (358 views)
Permalink

Keg - A python web framework

Hi,

I've been working on a python web framework which I think might be of
interest to you.
Details may be found at http://code.google.com/p/keg/wiki/Concept.

All suggestions or comments will be greatly appreciated.

Thank you.

fetchinson at googlemail

Aug 4, 2008, 1:13 PM

Post #2 of 7 (334 views)
Permalink

Re: Keg - A python web framework [In reply to]

> I've been working on a python web framework which I think might be of
> interest to you.
> Details may be found at http://code.google.com/p/keg/wiki/Concept.
>
>
> All suggestions or comments will be greatly appreciated.

I fail to see what the advantages of your framework are over django or
turbogears. The description you give on the above url doesn't really
help as all the goals you state for your project are solved by both
django and turbogears.

I wouldn't want to be too discouraging but it seems you are led by the
NIH principle which is not a really useful one in the open source
world. You can expect exactly 0 users and no appreciation for your
efforts which will/can lead to frustration and bad health.
Contributing to an already existing and mature framework like django
and turbogears can/will be on the other hand rewarding.

Cheers,
Daniel
--
Psss, psss, put it down! - http://www.cafepress.com/putitdown
--
http://mail.python.org/mailman/listinfo/python-list

workitharder at gmail

Aug 4, 2008, 3:18 PM

Post #3 of 7 (321 views)
Permalink

Re: Keg - A python web framework [In reply to]

On Aug 4, 1:13 pm, "Daniel Fetchinson" <fetchin...@googlemail.com>
wrote:
> > I've been working on a python web framework which I think might be of
> > interest to you.
> > Details may be found athttp://code.google.com/p/keg/wiki/Concept.
>
> > All suggestions or comments will be greatly appreciated.
>
> I fail to see what the advantages of your framework are over django or
> turbogears. The description you give on the above url doesn't really
> help as all the goals you state for your project are solved by both
> django and turbogears.
>
> I wouldn't want to be too discouraging but it seems you are led by the
> NIH principle which is not a really useful one in the open source
> world. You can expect exactly 0 users and no appreciation for your
> efforts which will/can lead to frustration and bad health.
> Contributing to an already existing and mature framework like django
> and turbogears can/will be on the other hand rewarding.
>
> Cheers,
> Daniel
> --
> Psss, psss, put it down! -http://www.cafepress.com/putitdown

I'd need to see a couple walkthroughs to be interested. Suggested
titles: "Easy things made easy" and "Hard things made possible". Links
to current sites using the platform also helps.
--
http://mail.python.org/mailman/listinfo/python-list

kay.schluehr at gmx

Aug 4, 2008, 8:36 PM

Post #4 of 7 (341 views)
Permalink

Re: Keg - A python web framework [In reply to]

> You can expect exactly 0 users and no appreciation for your
> efforts which will/can lead to frustration and bad health.

Is it really so bad to have no users and not becoming a "rock star
programmer" but rather being a "poor poet programmer" (PPP) instead?

--
http://mail.python.org/mailman/listinfo/python-list

straton at lampsacos

Aug 5, 2008, 4:37 AM

Post #5 of 7 (331 views)
Permalink

Re: Keg - A python web framework [In reply to]

Daniel Fetchinson wrote:
>> I've been working on a python web framework which I think might be of
>> interest to you.
>> Details may be found at http://code.google.com/p/keg/wiki/Concept.
>>
>>
>> All suggestions or comments will be greatly appreciated.
>
> I fail to see what the advantages of your framework are over django or
> turbogears. The description you give on the above url doesn't really
> help as all the goals you state for your project are solved by both
> django and turbogears.
>
> I wouldn't want to be too discouraging but it seems you are led by the
> NIH principle which is not a really useful one in the open source
> world. You can expect exactly 0 users and no appreciation for your
> efforts which will/can lead to frustration and bad health.
> Contributing to an already existing and mature framework like django
> and turbogears can/will be on the other hand rewarding.
>
> Cheers,
> Daniel

The goal I would like for your webapp project is:

"Do in Python what 'Apache Cocoon' does in Java"

There is already a python cocoon project, in hibernation.
Perhaps you could join it and wake it up!

http://pypi.python.org/pypi/pycoon
--
http://mail.python.org/mailman/listinfo/python-list

eghansah at gmail

Aug 8, 2008, 2:46 PM

Post #6 of 7 (324 views)
Permalink

Re: Keg - A python web framework [In reply to]

Hi,

Thank you for the comments so far.

To be honest with you I didn't know about pycoon until Bukzor mentioned it.
There appears to be some similarities between the two projects. However, I
think I'd have to take a closer look at it to be sure that I'm not
duplicating the efforts made in that project.

As to the question on how different this is from other frameworks, I think
there are certainly many similarities. As I admitted in the writeup, it
draws from other projects including django. However, there is one new idea I
haven't seen anywhere . . . not yet at least. In keg, I try to use URL
pattern matching to run one or more functions required to generate the page
the URL is pointing to. For instance, when you try to access a page like
http://www.python.org Keg will run all functions whose URL regex matches the
URL requested. Their output is then combined to generate the resulting page.
With this approach, we could have functions that generate menus, those that
generate page content and those that manage logins. These could all be
separately maintained. Keg ties their outputs all together to generate the
page. This means you could work on a menu system and not worry about how you
will generate ads for the page. The possibilities are endless . . . at least
in theory.

Also, each function receives the same input. This means that the execution
of one function does not really affect the execution of any others.
Hopefully this makes debugging much easier. Another good effect of this idea
is that all functions can be run in parallel since they are independent.

tjreedy at udel

Aug 8, 2008, 4:00 PM

Post #7 of 7 (319 views)
Permalink

Re: Keg - A python web framework [In reply to]

eghansah wrote:

> As to the question on how different this is from other frameworks, I
> think there are certainly many similarities. As I admitted in the
> writeup, it draws from other projects including django. However, there
> is one new idea I haven't seen anywhere . . . not yet at least. In keg,
> I try to use URL pattern matching to run one or more functions required
> to generate the page the URL is pointing to. For instance, when you try
> to access a page like http://www.python.org <http://www.python.org/> Keg
> will run all functions whose URL regex matches the URL requested. Their
> output is then combined to generate the resulting page. With this
> approach, we could have functions that generate menus, those that
> generate page content and those that manage logins. These could all be
> separately maintained. Keg ties their outputs all together to generate
> the page. This means you could work on a menu system and not worry about
> how you will generate ads for the page. The possibilities are endless .
> . . at least in theory.
>
> Also, each function receives the same input. This means that the
> execution of one function does not really affect the execution of any
> others. Hopefully this makes debugging much easier. Another good effect
> of this idea is that all functions can be run in parallel since they are
> independent.

So make your project an add-on to Django or other frameworks. You tell
Django to send all requests to key.py. Keg.py runs the pattern matcher,
call the functions, and re-assembles the result to pass back to Django
for delivery. The functions then have all other components of Django
available.

In other words, don't reinvent the wheel, invent a new wheel cover*.

Terry Jan Reedy

This is possibly a new version of an old saying. Other endings I found
on Google (first 2000 hits) are '', 'improve it', 'improve on it', 'just
add new tyres', 'just identify a colleague (who has done it)', 'Patch,
extend or subclass an existing module', 're-invent its use!', 'write
something new', and '(or worse, a flat tire)'.

--
http://mail.python.org/mailman/listinfo/python-list

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads