[postgis-devel] Parser/Emitter Changeover

[Paul Ramsey]

Actually, I think it's easy enough to continue to emit the old
extended form (just testing a change that does that now), which for
backwards compatibility makes sense, even if hardly anyone (or no one)
uses it. A bigger backwards issue will be my desire to emit ISO WKT
instead of extended.

P.

On Mon, Dec 6, 2010 at 8:14 AM, Paul Ramsey <pramsey at opengeo.org> wrote:
> Well, the SQL/MM Part 3 spec BNF seems to require the dimensionality
> flags on all typed subparts as well as the parent part. For example:
>
> <geometrycollection text representation> ::= GEOMETRYCOLLECTION [ <z
> m> ] <geometrycollection text>
>
> <geometrycollection text> ::= <empty set>
> | <left paren> <well-known { <comma> <well-known text representation>
> text representation> }... <right paren>
>
> <well-known text representation> ::= <point text representation> |
> <curve text representation>
> | <surface text representation> | <collection text representation>
>
> <point text representation> ::= POINT [ <z m> ] <point text>
>
> On Mon, Dec 6, 2010 at 3:55 AM, strk <strk at keybit.net> wrote:
>> On Sat, Dec 04, 2010 at 12:15:40AM -0800, Paul Ramsey wrote:
>>
>>> - In EWKT, collections of M-dimensioned objects have the M flag on all
>>>   subcomponents as well as parent components.
>>>
>>>   eg: CURVEPOLYGONM(CIRCULARSTRINGM(..))
>>
>> Is there any standard out specifying this one ?
>>
>> --strk;
>>
>>  ()   Free GIS & Flash consultant/developer
>>  /\   http://strk.keybit.net/services.html
>> _______________________________________________
>> postgis-devel mailing list
>> postgis-devel at postgis.refractions.net
>> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>>
>