Using .map file as a raster layer source

[Tamas Szekeres]

Steve,

I'm assuming that the polygon of the city in the detailed map completely
covers the area of the city in itself, and the surrounding area is set
with transparent color.
For an example see 'www.utvonalterv.hu' and zoom in a smaller city. A
detailed map of the city will be drawn over the map of the country.


Tamas

On Thu, 25 Aug 2005 17:17:21 -0400, Stephen Woodbridge
<woodbri at SWOODBRIDGE.COM> wrote:

>Tamas,
>
>In a mapfile, if you have multiple overlaping layers then they all get
>drawn one over the next to build up the image.
>
>How would your mapfile layer work? the same way?
>
>So if I have a bunch of mapfiles representing the world and in same
>areas I have a city mapfiles that represent a city in detail and I have
>VMAP0 basemap mapfile layer, then it would draw the VMAP0 layer first
>and overlay it with the city layer, assuming the MIN/MAXSCALE of the
>city layer was on.
>
>Is this your intent for this feature?
>
>-Steve W.
>
>Tamas Szekeres wrote:
>> Frank,
>>
>> Certainly this solution should also work in the conventional way, when
the
>> DATA field contains the name of the mapfile. However that modification
>> would also address the case when we have an overall map (like a Country)
>> and a high number of detailed maps for a lower scale (cityes and
>> villages), and if the user is zooming in an area the corresponding high
>> detailed map should be drawn over the overall map. Tileindex is an
option
>> to prefilter the maps falling into the zooming area.
>> Up to present I haven't found a built in alternative to achieve this
>> issue, because for WMS, all of the hundreds of maps should be
>> predownloaded and written into the filesystem though only one or two
maps
>> should be really drawn. Suppose that the detailed maps are with
completely
>> different layer structure, so layer level tileindexing does not work.
>>
>> However anyone could write a script to realize this, but I think that it
>> is not practical to duplicate many of the functions already exist in the
>> system.
>>
>> Using mapfile as a data source (like inline WMS) is a different approach
>> which requires to realize many other features like selection of the
>> layers, with much of work. In this case the "CONNECTIONTYPE MAPFILE"
>> pseudo configuration would be syntactically more convenient.
>>
>>
>> Tamas Szekeres
>>
>>
>>
>> On Thu, 25 Aug 2005 11:07:31 -0400, Frank Warmerdam
<fwarmerdam at GMAIL.COM>
>> wrote:
>>
>>
>>>On 8/25/05, Sean Gillies <sgillies at frii.com> wrote:
>>>
>>>>Neither of you say it, but to me this seems to be sorta like an inline
>>>>WMS. I like the idea, and we should see about fitting it into the
>>>>proposed new layer architecture.
>>>
>>>Sean,
>>>
>>>It is indeed a bit like an inlined WMS, though it also, potentially,
>>>is benefiting from the tile index.
>>>
>>>However, I see it as orthagonal to the layer plugin proposal which
>>>is really about making the layer *feature* access plugable.   It does
>>>not address other aspects of layers, such as the Draw interface
>>>which is still based on the layer TYPE.   In this case this is a
>>>special case in the TYPE RASTER tileindex processing.
>>>
>>>Actually, Tamas, does the mapfile name have to come from a
>>>tileindex?  I don't see why it shouldn't be possible to have a single
>>>mapfile listed in the DATA argument instead of having to use the
>>>tileindex.  This would be in keeping with how other files work.
>>>
>>>Best regards,
>>>--
>>>---------------------------------------+--------------------------------
--
>>
>> ----
>>
>>>I set the clouds in motion - turn up   | Frank Warmerdam,
>>
>> warmerdam at pobox.com
>>
>>>light and sound - activate the windows | http://pobox.com/~warmerdam
>>>and watch the world go round - Rush    | Geospatial Programmer for Rent
>>
>>