geoalchemy integration

[Eric Lemoine-2]

Hi

(sorry if it's not the proper forum for this discussion, but I know
there may interested people listening here :-)

I'd like that MapFish rely on GeoAlchemy, to share code and development.

For this, we need that GeoAlchemy reads database geometry values as
objects implementing the __geo_interface__.

The easiest would be to make GeoAlchemy create Shapely geometry
objects. But I don't think it'd be a good idea to make GeoAlchemy
depend on Shapely, because Shapely isn't a pure-Python lib.

So I've been thinking about adding the boolean option "shape" to
GeoAlchemy's TypeEngine (GeometryBase). If this option is true
GeoAlchemy will create a Shapely geometry object using
shapely.wrk.loads() and will add as a property to the
PersistentSpatialElement object.

What do you think?

Thanks,
--
Eric
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sanjiv]

On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
<eric.lemoine@...> wrote:
> Hi
>
> (sorry if it's not the proper forum for this discussion, but I know
> there may interested people listening here :-)

We are all part of the gispython "community", aren't we ;)

>
> I'd like that MapFish rely on GeoAlchemy, to share code and development.
>
> For this, we need that GeoAlchemy reads database geometry values as
> objects implementing the __geo_interface__.
>
> The easiest would be to make GeoAlchemy create Shapely geometry
> objects. But I don't think it'd be a good idea to make GeoAlchemy
> depend on Shapely, because Shapely isn't a pure-Python lib.
>
> So I've been thinking about adding the boolean option "shape" to
> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
> GeoAlchemy will create a Shapely geometry object using
> shapely.wrk.loads() and will add as a property to the
> PersistentSpatialElement object.
>
> What do you think?
>

That would be great.
Is it possible to implement this as a separate package? That would
ease up dependency management.

regards
Sanjiv

> Thanks,
> --
> Eric
> _______________________________________________
> Community mailing list
> Community@...
> http://lists.gispython.org/mailman/listinfo/community
>
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Eric Lemoine-2]

On Sat, Sep 26, 2009 at 2:50 PM, Sanjiv Singh <singhsanjivk@...> wrote:

> On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
> <eric.lemoine@...> wrote:
>> Hi
>>
>> (sorry if it's not the proper forum for this discussion, but I know
>> there may interested people listening here :-)
>
> We are all part of the gispython "community", aren't we ;)
>
>>
>> I'd like that MapFish rely on GeoAlchemy, to share code and development.
>>
>> For this, we need that GeoAlchemy reads database geometry values as
>> objects implementing the __geo_interface__.
>>
>> The easiest would be to make GeoAlchemy create Shapely geometry
>> objects. But I don't think it'd be a good idea to make GeoAlchemy
>> depend on Shapely, because Shapely isn't a pure-Python lib.
>>
>> So I've been thinking about adding the boolean option "shape" to
>> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
>> GeoAlchemy will create a Shapely geometry object using
>> shapely.wrk.loads() and will add as a property to the
>> PersistentSpatialElement object.
>>
>> What do you think?
>>
>
> That would be great.
> Is it possible to implement this as a separate package? That would
> ease up dependency management.

Could you elaborate?

See attached patch to understand what I'm suggesting.

--
Eric Lemoine

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac, Cedex

Tel : [phone]
Mail : eric.lemoine@...
http://www.camptocamp.com

[geoalchemy.diff]

---

diff -r 3ef69b5712ce geoalchemy/geometry.py
--- a/geoalchemy/geometry.py Tue Sep 01 23:21:34 2009 +0530
+++ b/geoalchemy/geometry.py Sat Sep 26 15:44:33 2009 +0200
@@ -24,14 +24,22 @@
     def result_processor(self, dialect):
         def process(value):
             if value is not None:
+                elt = None
                 if isinstance(dialect, PGDialect):
-                    return PGPersistentSpatialElement(value)
+                    elt = PGPersistentSpatialElement(value)
                 if isinstance(dialect, SQLiteDialect):
-                    return SQLitePersistentSpatialElement(value)
+                    elt = SQLitePersistentSpatialElement(value)
                 if isinstance(dialect, MySQLDialect):
-                    return MySQLPersistentSpatialElement(value)
+                    elt = MySQLPersistentSpatialElement(value)
                 else:
                     raise NotImplementedError
+                if self.shape:
+                    # create a shapely geometry from the value
+                    # and add as a property in the spatial
+                    # element
+                    from shapely.wkb import loads
+                    elt.shape = loads(value.decode('hex'))
+                return elt
             else:
                 return value
         return process

---

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Eric Lemoine-2]

On Sat, Sep 26, 2009 at 3:45 PM, Eric Lemoine
<eric.lemoine@...> wrote:

> On Sat, Sep 26, 2009 at 2:50 PM, Sanjiv Singh <singhsanjivk@...> wrote:
>> On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
>> <eric.lemoine@...> wrote:
>>> Hi
>>>
>>> (sorry if it's not the proper forum for this discussion, but I know
>>> there may interested people listening here :-)
>>
>> We are all part of the gispython "community", aren't we ;)
>>
>>>
>>> I'd like that MapFish rely on GeoAlchemy, to share code and development.
>>>
>>> For this, we need that GeoAlchemy reads database geometry values as
>>> objects implementing the __geo_interface__.
>>>
>>> The easiest would be to make GeoAlchemy create Shapely geometry
>>> objects. But I don't think it'd be a good idea to make GeoAlchemy
>>> depend on Shapely, because Shapely isn't a pure-Python lib.
>>>
>>> So I've been thinking about adding the boolean option "shape" to
>>> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
>>> GeoAlchemy will create a Shapely geometry object using
>>> shapely.wrk.loads() and will add as a property to the
>>> PersistentSpatialElement object.
>>>
>>> What do you think?
>>>
>>
>> That would be great.
>> Is it possible to implement this as a separate package? That would
>> ease up dependency management.
>
> Could you elaborate?
>
> See attached patch to understand what I'm suggesting.

and FYI I got the read part of MapFish to work with GeoAlchemy with
only the attached patch to GeoAlchemy.

--
Eric Lemoine

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac, Cedex

Tel : [phone]
Mail : eric.lemoine@...
http://www.camptocamp.com

[geoalchemy.diff]

---

diff -r 3ef69b5712ce geoalchemy/base.py
--- a/geoalchemy/base.py Tue Sep 01 23:21:34 2009 +0530
+++ b/geoalchemy/base.py Sat Sep 26 15:55:35 2009 +0200
@@ -32,6 +32,7 @@
     
     def __init__(self, desc):
         self.desc = desc
+        self.shape = None
 
 class WKTSpatialElement(SpatialElement, expression.Function):
     """Represents a Geometry value expressed within application code; i.e. in
@@ -72,10 +73,11 @@
     
     name = 'GEOMETRY'
     
-    def __init__(self, dimension=None, srid=4326, spatial_index=True, **kwargs):
+    def __init__(self, dimension=None, srid=4326, spatial_index=True, shape=False, **kwargs):
         self.dimension = dimension
         self.srid = srid
         self.spatial_index = spatial_index
+        self.shape = shape
         super(GeometryBase, self).__init__(**kwargs)
     
     def bind_processor(self, dialect):
@@ -89,7 +91,11 @@
     def result_processor(self, dialect):
         def process(value):
             if value is not None:
-                return PersistentSpatialElement(value)
+                elt = PersistentSpatialElement(value)
+                if self.shape:
+                    from shapely.wkb import loads
+                    elt.shape = loads(value.decode('hex'))
+                return elt
             else:
                 return value
         return process
diff -r 3ef69b5712ce geoalchemy/geometry.py
--- a/geoalchemy/geometry.py Tue Sep 01 23:21:34 2009 +0530
+++ b/geoalchemy/geometry.py Sat Sep 26 15:55:35 2009 +0200
@@ -24,14 +24,22 @@
     def result_processor(self, dialect):
         def process(value):
             if value is not None:
+                elt = None
                 if isinstance(dialect, PGDialect):
-                    return PGPersistentSpatialElement(value)
-                if isinstance(dialect, SQLiteDialect):
-                    return SQLitePersistentSpatialElement(value)
-                if isinstance(dialect, MySQLDialect):
-                    return MySQLPersistentSpatialElement(value)
+                    elt = PGPersistentSpatialElement(value)
+                elif isinstance(dialect, SQLiteDialect):
+                    elt = SQLitePersistentSpatialElement(value)
+                elif isinstance(dialect, MySQLDialect):
+                    elt = MySQLPersistentSpatialElement(value)
                 else:
                     raise NotImplementedError
+                if self.shape:
+                    # create a shapely geometry from the value
+                    # and add as a property in the spatial
+                    # element
+                    from shapely.wkb import loads
+                    elt.shape = loads(value.decode('hex'))
+                return elt
             else:
                 return value
         return process

---

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sanjiv]

On Sat, Sep 26, 2009 at 7:27 PM, Eric Lemoine
<eric.lemoine@...> wrote:

> On Sat, Sep 26, 2009 at 3:45 PM, Eric Lemoine
> <eric.lemoine@...> wrote:
>> On Sat, Sep 26, 2009 at 2:50 PM, Sanjiv Singh <singhsanjivk@...> wrote:
>>> On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
>>> <eric.lemoine@...> wrote:
>>>> Hi
>>>>
>>>> (sorry if it's not the proper forum for this discussion, but I know
>>>> there may interested people listening here :-)
>>>
>>> We are all part of the gispython "community", aren't we ;)
>>>
>>>>
>>>> I'd like that MapFish rely on GeoAlchemy, to share code and development.
>>>>
>>>> For this, we need that GeoAlchemy reads database geometry values as
>>>> objects implementing the __geo_interface__.
>>>>
>>>> The easiest would be to make GeoAlchemy create Shapely geometry
>>>> objects. But I don't think it'd be a good idea to make GeoAlchemy
>>>> depend on Shapely, because Shapely isn't a pure-Python lib.
>>>>
>>>> So I've been thinking about adding the boolean option "shape" to
>>>> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
>>>> GeoAlchemy will create a Shapely geometry object using
>>>> shapely.wrk.loads() and will add as a property to the
>>>> PersistentSpatialElement object.
>>>>
>>>> What do you think?
>>>>
>>>
>>> That would be great.
>>> Is it possible to implement this as a separate package? That would
>>> ease up dependency management.
>>
>> Could you elaborate?

Basically, if someone installs geoalchemy without Shapely and then
turns shape=True he will
run into problems. Although when someone turns shape=True, he would
know what he is doing
and will most likely have shapely installed. But the possibility of
having a dependency issue does exist.

But again we need not have a separate package if it makes it too complex.

>>
>> See attached patch to understand what I'm suggesting.
>
> and FYI I got the read part of MapFish to work with GeoAlchemy with
> only the attached patch to GeoAlchemy.

Thats awesome. Should I grant you commit to the bitbucket repo?
Or you may like to fork it and send me pull request.

regards
Sanjiv

>
> --
> Eric Lemoine
>
> Camptocamp France SAS
> Savoie Technolac, BP 352
> 73377 Le Bourget du Lac, Cedex
>
> Tel : [phone]
> Mail : eric.lemoine@...
> http://www.camptocamp.com
>
> _______________________________________________
> Community mailing list
> Community@...
> http://lists.gispython.org/mailman/listinfo/community
>
>

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 26, 2009, at 6:34 PM, Sanjiv Singh wrote:

> On Sat, Sep 26, 2009 at 7:27 PM, Eric Lemoine
> <eric.lemoine@...> wrote:
>> On Sat, Sep 26, 2009 at 3:45 PM, Eric Lemoine
>> <eric.lemoine@...> wrote:
>>> On Sat, Sep 26, 2009 at 2:50 PM, Sanjiv Singh <singhsanjivk@...
>>> > wrote:
>>>> On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
>>>> <eric.lemoine@...> wrote:
>>>>> Hi
>>>>>
>>>>> (sorry if it's not the proper forum for this discussion, but I  
>>>>> know
>>>>> there may interested people listening here :-)
>>>>
>>>> We are all part of the gispython "community", aren't we ;)
>>>>
>>>>>
>>>>> I'd like that MapFish rely on GeoAlchemy, to share code and  
>>>>> development.
>>>>>
>>>>> For this, we need that GeoAlchemy reads database geometry values  
>>>>> as
>>>>> objects implementing the __geo_interface__.
>>>>>
>>>>> The easiest would be to make GeoAlchemy create Shapely geometry
>>>>> objects. But I don't think it'd be a good idea to make GeoAlchemy
>>>>> depend on Shapely, because Shapely isn't a pure-Python lib.
>>>>>
>>>>> So I've been thinking about adding the boolean option "shape" to
>>>>> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
>>>>> GeoAlchemy will create a Shapely geometry object using
>>>>> shapely.wrk.loads() and will add as a property to the
>>>>> PersistentSpatialElement object.
>>>>>
>>>>> What do you think?
>>>>>
>>>>
>>>> That would be great.
>>>> Is it possible to implement this as a separate package? That would
>>>> ease up dependency management.
>>>
>>> Could you elaborate?
>
> Basically, if someone installs geoalchemy without Shapely and then
> turns shape=True he will
> run into problems. Although when someone turns shape=True, he would
> know what he is doing
> and will most likely have shapely installed. But the possibility of
> having a dependency issue does exist.
>
> But again we need not have a separate package if it makes it too  
> complex.
>
>>>
>>> See attached patch to understand what I'm suggesting.
>>
>> and FYI I got the read part of MapFish to work with GeoAlchemy with
>> only the attached patch to GeoAlchemy.
>
> Thats awesome. Should I grant you commit to the bitbucket repo?
> Or you may like to fork it and send me pull request.
>
> regards
> Sanjiv

Do either of you think that a cross-project mailing list would be  
helpful for discussing these kinds of interoperability issues?  
Something like's Python's Web-SIG list, but for Python and GIS? I'm  
not saying this discussion is off-topic (you're both Shapely users),  
but it might reach a wider audience on a list that wasn't associated  
with a particular bunch of software projects.

Cheers,

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Eric Lemoine-2]

On Saturday, September 26, 2009, Sanjiv Singh <singhsanjivk@...> wrote:

> On Sat, Sep 26, 2009 at 7:27 PM, Eric Lemoine
> <eric.lemoine@...> wrote:
>> On Sat, Sep 26, 2009 at 3:45 PM, Eric Lemoine
>> <eric.lemoine@...> wrote:
>>> On Sat, Sep 26, 2009 at 2:50 PM, Sanjiv Singh <singhsanjivk@...> wrote:
>>>> On Sat, Sep 26, 2009 at 5:13 PM, Eric Lemoine
>>>> <eric.lemoine@...> wrote:
>>>>> Hi
>>>>>
>>>>> (sorry if it's not the proper forum for this discussion, but I know
>>>>> there may interested people listening here :-)
>>>>
>>>> We are all part of the gispython "community", aren't we ;)
>>>>
>>>>>
>>>>> I'd like that MapFish rely on GeoAlchemy, to share code and development.
>>>>>
>>>>> For this, we need that GeoAlchemy reads database geometry values as
>>>>> objects implementing the __geo_interface__.
>>>>>
>>>>> The easiest would be to make GeoAlchemy create Shapely geometry
>>>>> objects. But I don't think it'd be a good idea to make GeoAlchemy
>>>>> depend on Shapely, because Shapely isn't a pure-Python lib.
>>>>>
>>>>> So I've been thinking about adding the boolean option "shape" to
>>>>> GeoAlchemy's TypeEngine (GeometryBase). If this option is true
>>>>> GeoAlchemy will create a Shapely geometry object using
>>>>> shapely.wrk.loads() and will add as a property to the
>>>>> PersistentSpatialElement object.
>>>>>
>>>>> What do you think?
>>>>>
>>>>
>>>> That would be great.
>>>> Is it possible to implement this as a separate package? That would
>>>> ease up dependency management.
>>>
>>> Could you elaborate?
>
> Basically, if someone installs geoalchemy without Shapely and then
> turns shape=True he will
> run into problems. Although when someone turns shape=True, he would
> know what he is doing
> and will most likely have shapely installed. But the possibility of
> having a dependency issue does exist.

Right. I just don't see any simple way to address this at this point.

>
> But again we need not have a separate package if it makes it too complex.
>
>>>
>>> See attached patch to understand what I'm suggesting.
>>
>> and FYI I got the read part of MapFish to work with GeoAlchemy with
>> only the attached patch to GeoAlchemy.
>
> Thats awesome. Should I grant you commit to the bitbucket repo?

I'd be honored to. In any way, I'll send you my patches for review
before any commit.

Thanks,

> Or you may like to fork it and send me pull request.
>
> regards
> Sanjiv
>
>>
>> --
>> Eric Lemoine
>>
>> Camptocamp France SAS
>> Savoie Technolac, BP 352
>> 73377 Le Bourget du Lac, Cedex
>>
>> Tel : [phone]
>> Mail : eric.lemoine@...
>> http://www.camptocamp.com
>>
>> _______________________________________________
>> Community mailing list
>> Community@...
>> http://lists.gispython.org/mailman/listinfo/community
>>
>>
> _______________________________________________
> Community mailing list
> Community@...
> http://lists.gispython.org/mailman/listinfo/community
>

--
Eric Lemoine

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac, Cedex

Tel : [phone]
Mail : eric.lemoine@...
http://www.camptocamp.com
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Eric Lemoine-2]

> Do either of you think that a cross-project mailing list would be
> helpful for discussing these kinds of interoperability issues?
> Something like's Python's Web-SIG list, but for Python and GIS? I'm
> not saying this discussion is off-topic (you're both Shapely users),
> but it might reach a wider audience on a list that wasn't associated
> with a particular bunch of software projects.

Hi Sean.

As you see it, would this list be a gispython list, i.e. be listed
here <http://lists.gispython.org/mailman/listinfo/>?

Thanks,
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 28, 2009, at 9:28 AM, Eric Lemoine wrote:

>> Do either of you think that a cross-project mailing list would be
>> helpful for discussing these kinds of interoperability issues?
>> Something like's Python's Web-SIG list, but for Python and GIS? I'm
>> not saying this discussion is off-topic (you're both Shapely users),
>> but it might reach a wider audience on a list that wasn't associated
>> with a particular bunch of software projects.
>
> Hi Sean.
>
> As you see it, would this list be a gispython list, i.e. be listed
> here <http://lists.gispython.org/mailman/listinfo/>?

I think separated would be better. Using Google or Yahoo groups would  
make it easy to share administration.

Cheers,

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Eric Lemoine-2]

On Monday, September 28, 2009, Sean Gillies <sean.gillies@...> wrote:

> On Sep 28, 2009, at 9:28 AM, Eric Lemoine wrote:
>
>>> Do either of you think that a cross-project mailing list would be
>>> helpful for discussing these kinds of interoperability issues?
>>> Something like's Python's Web-SIG list, but for Python and GIS? I'm
>>> not saying this discussion is off-topic (you're both Shapely users),
>>> but it might reach a wider audience on a list that wasn't associated
>>> with a particular bunch of software projects.
>>
>> Hi Sean.
>>
>> As you see it, would this list be a gispython list, i.e. be listed
>> here <http://lists.gispython.org/mailman/listinfo/>?
>
> I think separated would be better. Using Google or Yahoo groups would
> make it easy to share administration.

+1. Thanks,

>
> Cheers,
>
> --
> Sean
>
> _______________________________________________
> Community mailing list
> Community@...
> http://lists.gispython.org/mailman/listinfo/community
>

--
Eric Lemoine

Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac, Cedex

Tel : [phone]
Mail : eric.lemoine@...
http://www.camptocamp.com
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 30, 2009, at 7:36 AM, Eric Lemoine wrote:

> On Monday, September 28, 2009, Sean Gillies <sean.gillies@...>  
> wrote:
>> On Sep 28, 2009, at 9:28 AM, Eric Lemoine wrote:
>>
>>>> Do either of you think that a cross-project mailing list would be
>>>> helpful for discussing these kinds of interoperability issues?
>>>> Something like's Python's Web-SIG list, but for Python and GIS? I'm
>>>> not saying this discussion is off-topic (you're both Shapely  
>>>> users),
>>>> but it might reach a wider audience on a list that wasn't  
>>>> associated
>>>> with a particular bunch of software projects.
>>>
>>> Hi Sean.
>>>
>>> As you see it, would this list be a gispython list, i.e. be listed
>>> here <http://lists.gispython.org/mailman/listinfo/>?
>>
>> I think separated would be better. Using Google or Yahoo groups would
>> make it easy to share administration.
>
> +1. Thanks,

Okay, let's do it. Your +1 vote makes you a group administrator ;)

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Graham Carlyle-3]

Sean Gillies wrote:
>
> Do either of you think that a cross-project mailing list would be  
> helpful for discussing these kinds of interoperability issues?  
> Something like's Python's Web-SIG list, but for Python and GIS? I'm  
> not saying this discussion is off-topic (you're both Shapely users),  
> but it might reach a wider audience on a list that wasn't associated  
> with a particular bunch of software projects.

Sorry to butt in but I also think that would be a good idea.

I initially thought this list was the (possibly defacto through its
name) cross-project mailing list for python and GIS. But as you've just
re-stated its for the gispython hosted projects.

So I think a general purpose GIS and python mailing list would be good.
Unless you're suggesting a smaller remit, possibly one just relating to
cross project issues?

cheers,
Graham
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 30, 2009, at 10:08 AM, Sean Gillies wrote:

>
> On Sep 30, 2009, at 7:36 AM, Eric Lemoine wrote:
>
>> On Monday, September 28, 2009, Sean Gillies <sean.gillies@...>
>> wrote:
>>> On Sep 28, 2009, at 9:28 AM, Eric Lemoine wrote:
>>>
>>>>> Do either of you think that a cross-project mailing list would be
>>>>> helpful for discussing these kinds of interoperability issues?
>>>>> Something like's Python's Web-SIG list, but for Python and GIS?  
>>>>> I'm
>>>>> not saying this discussion is off-topic (you're both Shapely
>>>>> users),
>>>>> but it might reach a wider audience on a list that wasn't
>>>>> associated
>>>>> with a particular bunch of software projects.
>>>>
>>>> Hi Sean.
>>>>
>>>> As you see it, would this list be a gispython list, i.e. be listed
>>>> here <http://lists.gispython.org/mailman/listinfo/>?
>>>
>>> I think separated would be better. Using Google or Yahoo groups  
>>> would
>>> make it easy to share administration.
>>
>> +1. Thanks,
>
> Okay, let's do it. Your +1 vote makes you a group administrator ;)
>

Done: http://groups.google.com/group/python-gis-sig

Here's what I wrote for an introduction: """

This is a list for software developers, programmers, and architects,  
for discussion of general issues around the Python language (and  
variants), platforms, and geographic information systems. "Help"  
questions specific to particular software products are off-topic and  
better addressed to that software product's support lists and forums.

"""

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 30, 2009, at 10:13 AM, Graham Carlyle wrote:

> Sean Gillies wrote:
>>
>> Do either of you think that a cross-project mailing list would be
>> helpful for discussing these kinds of interoperability issues?
>> Something like's Python's Web-SIG list, but for Python and GIS? I'm
>> not saying this discussion is off-topic (you're both Shapely users),
>> but it might reach a wider audience on a list that wasn't associated
>> with a particular bunch of software projects.
>
> Sorry to butt in but I also think that would be a good idea.
>
> I initially thought this list was the (possibly defacto through its
> name) cross-project mailing list for python and GIS. But as you've  
> just
> re-stated its for the gispython hosted projects.
>
> So I think a general purpose GIS and python mailing list would be  
> good.
> Unless you're suggesting a smaller remit, possibly one just relating  
> to
> cross project issues?

What do you think of the description/mission of http://groups.google.com/group/python-gis-sig?
  Too narrow?

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Graham Carlyle-3]

Sean Gillies wrote:
>
> What do you think of the description/mission of http://groups.google.com/group/python-gis-sig?
>   Too narrow?

This description seems fine to me but I'm not sure if the "Help"
questions sentence is necessary.

Its probably inevitable that people will go to this list for at least
pointers to suitable projects to help them do stuff. I would see that as
a benefit of the list. Do you think it will get overwhelmed by those
sorts of questions and so be less interesting for more technical
discussions?

Graham
_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Sep 30, 2009, at 10:43 AM, Graham Carlyle wrote:

> Sean Gillies wrote:
>>
>> What do you think of the description/mission of http://groups.google.com/group/python-gis-sig?
>>  Too narrow?
>
> This description seems fine to me but I'm not sure if the "Help"
> questions sentence is necessary.
>
> Its probably inevitable that people will go to this list for at least
> pointers to suitable projects to help them do stuff. I would see  
> that as
> a benefit of the list. Do you think it will get overwhelmed by those
> sorts of questions and so be less interesting for more technical
> discussions?
>
> Graham

A grizzled Python and GIS veteran assures me that we'll get lots of  
ESRI-specific geoprocessing questions unless we're very clear that  
they are off-topic.

While my own agenda is to get developers to talk and get some cross-
project fertilization, I agree with you that the list should answer  
"How do I get started?" questions. Ideally, we'd produce an FAQ as we  
go and refer people to it.

Sean

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Timmie]

> So I think a general purpose GIS and python mailing list would be good.
> Unless you're suggesting a smaller remit, possibly one just relating to
> cross project issues?
+1
this list here appeared from its description to be such a place. Maybe
because it was one of the first projects...

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Timmie]

> Done: http://groups.google.com/group/python-gis-sig
>
> Here's what I wrote for an introduction: """
>
> This is a list for software developers, programmers, and architects,  
> for discussion of general issues around the Python language (and  
> variants), platforms, and geographic information systems. "Help"  
> questions specific to particular software products are off-topic and  
> better addressed to that software product's support lists and forums.
>
> """

I have one suggestion:
Could the Maillist Owner please add the list to Gmane?
This is very convenient.
I have had the experience that the subscription reuqests via Gmane do
not work.
One has to actively put the address of Gmane to list members.

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Sean Gillies-2]

On Oct 1, 2009, at 10:06 PM, Tim Michelsen wrote:

>> Done: http://groups.google.com/group/python-gis-sig
>>
>> Here's what I wrote for an introduction: """
>>
>> This is a list for software developers, programmers, and architects,
>> for discussion of general issues around the Python language (and
>> variants), platforms, and geographic information systems. "Help"
>> questions specific to particular software products are off-topic and
>> better addressed to that software product's support lists and forums.
>>
>> """
> I have one suggestion:
> Could the Maillist Owner please add the list to Gmane?
> This is very convenient.
> I have had the experience that the subscription reuqests via Gmane do
> not work.
> One has to actively put the address of Gmane to list members.

I tried to subscribe it to gmane.comp.python.gis.sig, but didn't get  
any response back from Gmane yet. What do you mean by "actively put  
the address of Gmane to list members"?

--
Sean

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community

[Timmie]

> I tried to subscribe it to gmane.comp.python.gis.sig, but didn't get  
> any response back from Gmane yet.
It takes some days.

> What do you mean by "actively put  
> the address of Gmane to list members"?
I think in case they cannot subscribe automatically, you'd receive a
mail with instructions.

Thanks for your efforts!

_______________________________________________
Community mailing list
Community@...
http://lists.gispython.org/mailman/listinfo/community