[mapserver-users] PostGIS problems in nightly

[Havard Tveite]

David Fuhry has already made a patch:
http://trac.osgeo.org/mapserver/ticket/2613

After applying the patch, the behaviour seems to be
back to normal.

Regards,

Håvard

Stephen Woodbridge wrote:
> Havard,
> 
> This sounds like a bug, can you create a bug on this so it gets resolved
> for 5.2. If not let me know, and I'll write on.
> 
> Thanks,
>    -Steve W
> 
> Havard Tveite wrote:
>> Sorry for not testing more, Steve.  Answers to your questions
>> below.
>>
>> Stephen Woodbridge wrote:
>>> Håvard,
>>>
>>> Try:
>>>
>>> trim(both from direction)
>> works
>>
>>> which does the same thing as:
>>>
>>> trim(both ' ' from direction)
>>>
>>> this looks like it might be an unescaped ' and in single quotes string.
>>>
>>> you could also try:
>>>
>>> trim(both \' \' from direction)
>> does not work (still "crash").
>>
>>> or
>>>
>>> trim(both \\' \\' from direction)
>> does not work (still "crash").
>>
>>
>> Regards,
>>
>> Håvard
>>
>>> and report if either of these work.
>>>
>>> This would help narrow down the issue and maybe give you a work around.
>>>
>>> -Steve W
>>>
>>> Håvard Tveite wrote:
>>>> I want to report a small problem I faced with a PostGIS
>>>> connection with the latest nightly build.
>>>>
>>>> With the latest build I have problems with one of my
>>>> PostGIS connections.
>>>> The problem goes away when I switch back to the nightly
>>>> build from March 4th 2008.  Other PostGIS connections
>>>> seem to work fine on both builds.
>>>>
>>>> The mapserver_log reports:
>>>> ******************************************************
>>>> Fri May  9 17:08:42 2008,2747,128.39.236.206,okart,0,-19379.849812
>>>> 26500.000000 19379.849812 57500.000000,-1.000000 -1.000000,o-kartpoint
>>>> o-kartpolygon o-kartline ,msDrawMap(): Image handling error. Failed to
>>>> draw layer named 'o-kartpoint'. <br>
>>>> prepare_database(): Query error. Error declaring cursor: ERROR:  current
>>>> transaction is aborted, commands ignored until end of transaction block
>>>>
>>>> With query string: DECLARE mycursor BINARY CURSOR FOR SELECT
>>>> "symbol"::text,"direction"::text,asbinary(force_collection(force_2d(the_geom)),'XDR'),OID::text
>>>>
>>>> from (select oid, symbol, trim(both ' ' from direction) as direction,
>>>> the_geom from okartpoint order by sequence) as okart WHERE the_geom &&
>>>> setSRID('BOX3D(-19379.8498122653 26500,19379.8498122653 57500)'::BOX3D,
>>>> -1 )
>>>>  <br>
>>>> msPOSTGISLayerRetrievePK(): Query error. Error executing POSTGIS
>>>> statement (msPOSTGISLayerRetrievePK():select attname from pg_attribute,
>>>> pg_constraint, pg_class where pg_constraint.conrelid = pg_class.oid and
>>>> pg_class.oid = pg_attribute.attrelid and pg_constraint.contype = 'p' and
>>>> pg_constraint.conkey[1] = pg_attribute.attnum and pg_class.relname =
>>>> '(select oid, symbol, trim(both ' ' from direction) as direction,
>>>> the_geom from okartpoint order by sequence) as okart' and
>>>> pg_table_is_visible(pg_class.oid) and pg_constraint.conkey[2] is null
>>>> <br>
>>>> ******************************************************
>>>>
>>>> PostgreSQL log:
>>>> ******************************************************
>>>> 2008-05-09 17:08:42 MET DST okart guest 128.39.175.213ERROR:  syntax
>>>> error at or near "' from direction) as direction, the_geom from
>>>> okartpoint order by sequence) as okart'" at character 283
>>>> ******************************************************
>>>>
>>>> MapServer version 5.1-dev OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG OUTPUT=WBMP
>>>> OUTPUT=PDF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=AGG SUPPORTS=FREETYPE
>>>> SUPPORTS=ICONV SUPPORTS=WMS_SERVER SUPPORTS=WMS_CLIENT
>>>> SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT SUPPORTS=WCS_SERVER INPUT=EPPL7
>>>> INPUT=SDE INPUT=POSTGIS INPUT=OGR INPUT=GDAL INPUT=SHAPEFILE
>>>>
>>>> PostgreSQL 8.1.5, POSTGIS="1.3.2" GEOS="3.0.0-CAPI-1.4.1" PROJ="Rel.
>>>> 4.5.0, 22 Oct 2006" USE_STATS
>>>> I am on Solaris 2.7
>>>>
>>>> If I remove the trim statement ("trim(both ' ' from direction)
>>>> as direction"), and replace it with "direction" (untrimmed) there
>>>> is no error situation in the latest nightly.
>>>>
>>>> Is this a bug?
>>>>
>>>>
>>>> Regards,
>>>>
> 

-- 
--
Håvard Tveite
Department of Mathematical Sciences and Technology
Norwegian University of Life Sciences