web2py and Spatial / GIS support

[Pierre]

Hi,

just to clarify things a bit.....

(1)Is there any plan in the direction of implementing additional GIS functions like:  st_transform or st_interpolate in web2py  ?
As is it's not impossible to do GIS in web2py but it's certainly not very comfortable because whenever an absent function is involved you can forget about using the DAL and fall back on executesql(query)
(2)If not do you guys consider it's my responsability or someone else in the same case to dig in web2py source code and figure out a solution ?
(3)Or maybe GIS is considered a marginal issue ?
I would not feel offended if it is case (2) though I consider focusing on my app and making it fully functional is the best way I can contribute to the overall web2py project

[Pierre]

nobody knows or nobody wants to answer...........?
well that's not encouraging for people interested in GIS.......
and it might cause an unpleasant feeling that this GIS-thing is down......correct me if this is a false assumption

am sorry if i did ask the wrong question or if i hurted someone's feelings
anyway programmers aren't made of marshmallow ? are they..........eventhough some of them sleep with a teddy bear


"Le réel tel qu'il est, c'est une idée d'âne"    G.Deleuze in Nietzsche et la philosophie p 208

[Niphlod]

nobody pushed GIS and frankly the support for it is scattered. There hasn't been an holistic approach in the implementation.

IMHO you should not rely on anyone but yourself to add support on pydal.

[Pierre]

too bad....
do you think the existing implementation is still usable in a production context or is it too dangerous / unstable ?
there's a python package called pyproj that might offer some workarounds to transform points coordinates........and "stay in touch" with the dal

[DenesL]

I believe the original web2py GIS sponsor (http://sahanafoundation.org/) uses it in their products, which are production ready.

Expanding it depends on:
- existence of desired functionality in the supported databases
- desirability of said functions for the community
- willingness to contribute, from developers and users

[Pierre]

I thought GIS/Geolocation is something fancy  / highly desirable ? It also implies some important aspects of life ..........
the GIS function i miss is st_transform (POSTGIS) which i can test in depth if implemented. It won't be a drama if it's not as there are workarounds via pyproj or gdal used by Geodjango. In the worst case I cannot put all this together, you can imagine what a "reasonable person" would be tempted to do........

[Niphlod]

I don't want to push anyone away but if the underlying statement is "I need this ready to go" and a library-framework-whatever doesn't contain it, then there's really nothing anyone can do. Nobody ever said that web2py is a silver bullet (nor is it django, flask, express, sails, rails, cakephp, caddy, iris, play, limonade, onion, sinatra, put_your_own_here). If you feel you project can come to life easier with another framework and you don't want to contribute to web2py, hinting a blackmail is not going to win you any added feature :P

[Ron Chatterjee]

I share your frustration Niphlod and I am with you on this. I see all these posts about os, boy with a new toy but barely about how we do certain things given what's already there.  

[Pierre]

I already said I wanted to contribute as a user aiming at designing a completely functional app. If you feel this isn't enough of a contribution then we disagree. I don't have the energy / time  and possibly the skills to participate in Web2py api development and at the same time to write/test/deploy/promote and so on my app. This might be frustrating but I can't transform myself into  "superman" just because of that.  Neither will I be frustrated in case no more feature is added. I'll do with what I have in my hands now and you have to admit that any lambda user must also keep a "sense of economy" in order to "survive".  It ain't easy to build an app as anyone can realize reading the continuous flow of user's posts............be it with Web2py / Django / Zope.......whatever......Then we must face  the complexity  of  promoting the app  to make it viable...etc.......that's why we should gather our forces not divide them. Writing applications is only  a small part of the overall challenge, publicity probably stands as the most important part. We should think about that too.......that is take height away from technical issues........

[Ron Chatterjee]

May be we can have a tab in web2py page or slice with "feature request" n have people submit ideas n others vote for it. If number of votes exceed 100 it goes to a priority list or something. Just some thoughts...