[Qgis-developer] Aliases of features attributes in Mapserver's WMS GetProjectSettings and GetFeatureInfo

[Bernhard Ströbl]

Hi Marcel,

have you seen this [1]?
Clearly mapserver should stick to either name or alias in _all_ 
responses. As (I assume) name is unique (because it must be unique in 
the data source) and alias may not (because it is a user setting), 
mapserver should use the name and (as proposed by you) additionally 
provide the alias at least in GetProjectSettings. No idea if 
GetCapabilities and GetFeatureInfo should return the alias, too, because 
they are standardized requests.

Bernhard


[1] http://hub.qgis.org/issues/8727

Am 08.04.2014 08:05, schrieb Marcel Dancak:
> Hi, I found some issues in Mapserver when using aliases for features
> attributes.
>
> GetProjectSetting response has features attributes of vector layer in
> this form
>
> <Attribute typeName="INTEGER" precision="0" length="0"
> editType="LineEdit" type="qlonglong" comment="" name="PK_UID"/>
>
> where "name" property contains either attribute's alias (if defined) or
> its name. So you are loosing attribute's name when using aliases, and
> that make it impossible to perform searching by this attribute with
> GetFeatureInfo request (with FILTER parameter). Separate "alias"
> property would be much better.
>
> Another question is, what should be used as attribute name in
> GetFeatureInfo response, alias or name? Current mapserver returns prefer
> aliases (if defined) in plain XML format, but in GML format it always
> returns names. In my opinion, names should be used in both formats and
> aliases could be passed as additional info (if particular format allows
> that), or not used at all in GetFeatureInfo cause clients would be able
> to perform mapping from names to aliases by info from GetProjectSettings.
>
> And the last issue is that mapserver always returns all features
> attributes in GetFeatureInfo response is GML format, even if they are
> marked to be excluded in WMS server.
>
>
> I have all these issues fixed and I can create a pull requests, but
> first I want to discuss about it and maybe another solutions will came out.
>
>
> Marcel
>
>


__________ Information from ESET Security, version of virus signature database 9648 (20140407) __________

The message was checked by ESET Security.

    part000.txt - is OK

http://www.eset.com