[mapserver-dev] RFC 102: Styleitem Javascript Plugin

Alan Boudreault aboudreault at mapgears.com
Fri Aug 30 12:44:23 PDT 2013


All,

I've put some prototype code in my github fork to demonstrate that this 
doesn't affect many mapserver parts. The main changes are:

- 
https://github.com/aboudreault/mapserver/blob/styleitem_javascript/mapv8.cpp
- 
https://github.com/aboudreault/mapserver/blob/styleitem_javascript/maplayer.c#L844

Note that this is still in development. There are still work to do and 
exposing an object properly to do.

Alan

On 13-08-20 12:09 PM, thomas bonfort wrote:
> (this is mostly a summary of a discussion I had with Alan)
>
> While this seems like a promising and powerful feature, I think it
> would be beneficial if we didn't move too fast on this one:
>
> - using STYLEITEM is self-contained, but does have some
> non-negligeable limitations feature-wise (typically: the shape's
> attributes alone give no way of accessing the current map scale, which
> would seem like a minimum in order to be able to do dynamic styling)
> - future needs, such as generic templating, manipulating geometries,
> accessing the whole resultset (in order to apply content-dependant
> thematic mapping), ..., are not addressed
> - while this is tied deeper into the mapserver guts than mapscripts,
> it might still be beneficial to think of this in the scope of a public
> mapserver API, i.e. what members/getters/setters do we expose for
> outside manipulation.
>
> While I somewhat resent the fact of being the one trying to slow
> things down, I feel this needs a longer discussion, and would be a
> very interesting subject for a codesprint. However Alan has a
> commercial need behind this to move faster and imposing a 6 month wait
> might also bury the idea completely.
>
> Provided that the impact on the codebase is limited to some
> v8/mapserver wrapper code along with styleitem plugging, I would like
> to recommend that the RFC be amended to add that this feature is
> experimental and that backwards compatibility will likely/probably be
> broken in the not-so-short term if and when we decide to move deeper
> into its possibilities. (Alan confirmed that this is acceptable for
> him and his client).
>
> thoughts?
>
> best regards,
> thomas
>
>
> On 19 August 2013 18:17, Alan Boudreault <aboudreault at mapgears.com> wrote:
>> Devs,
>>
>> Please take a look at RFC 102 which propose the addition of a styleitem
>> javascript plugin as an alternative to style our features.
>>
>> http://mapserver.org/development/rfc/ms-rfc-102.html
>>
>> I'll be calling for vote on it next week.
>>
>> Thanks,
>> Alan
>>
>> --
>> Alan Boudreault
>> http://www.mapgears.com/
>> _______________________________________________
>> mapserver-dev mailing list
>> mapserver-dev at lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/mapserver-dev


-- 
Alan Boudreault
http://www.mapgears.com/


More information about the mapserver-dev mailing list