[Chameleon] Query Widget Bug?
pspencer at dmsolutions.ca
Thu Mar 10 13:03:43 EST 2005
I honestly think this is a problem in the postgis driver rather than a
mapserver or chameleon problem. Can you post the details on the
mapserver list to see if any of the refractions guys respond?
Corey Puffalt wrote:
> First a bit more information: I have a number of layers that do not
> exhibit this problem and a number that do. All of these layers are
> postgis layers... in fact, all the layers query the same table. The
> only difference is the FILTER parameter. (Each layer differs in that
> the data is specific to a particular company.) The only difference
> I've been able to identify between the layers that do work and the
> ones that don't are where the geometry data is located (Canada vs. US
> mostly)... the layers with lines in the US appear to have this eror.
> I've discovered sometimes no error is printed in the pop-up query box
> but what I get instead is a table with just the SHAPE_INDEX,
> TILE_INDEX & LAYER_INDEX columns (the layer attributes are all
> missing). In both cases, the map in the map window fails to render
> (broken image in the browser.)
> Mapscript is being used as a module AFAIK. (Default FGS setup.)
> PHP 4.3.10 (cgi) (built: Feb 9 2005 18:44:59)
> Copyright (c) 1997-2004 The PHP Group
> Zend Engine v1.3.0, Copyright (c) 1998-2004 Zend Technologies
> I notice that the MapQuery class extends Logger and uses it to log a
> bunch of information. How do I enable the logging and set the
> filename appropriately? Can I access this Logger and control it when
> I initialize my Chameleon subclass?
> Thanks for your help,
> On Thu, 10 Mar 2005 12:28:16 -0500, Paul Spencer
> <pspencer at dmsolutions.ca> wrote:
>>damn. Thats exactly my version.
>>do you know what layer is being queried when you get this error? At the
>>point that Chameleon gets to this line, MapScript has already reported a
>>certain number of results and this code is the correct way to read the
>>A couple of thoughts/ideas/questions:
>>* $oShape->free() should probably be called at the end of the loop (just
>>before // end for loop)
>>* what type of layer is being queried? Is it, perchance, a postgis
>>layer or some other spatial database layer? My thought here is that the
>>connection may be getting corrupted or timed out or something
>>* are you using MapScript as a cgi or a module?
>>* what version of PHP are you using
>>I've also cc'd Daniel Morissette to see if there is anything in the
>>MapServer core that would indicate why you would get this particular
>>error message from this particular mapscript call.
>>Corey Puffalt wrote:
>>>Yeah that's what I thought too... :P I've left the chameleon stuff
>>>untouched so it should be exactly what you have (at least as
>>>distributed by the FGS project.)
>>>At any rate, I've attached the file... the line referred to seems to
>>>be this one:
>>> $oShape = $oLayer->getShape($oResult->tileindex,
>>>On Thu, 10 Mar 2005 11:39:20 -0500, Paul Spencer
>>><pspencer at dmsolutions.ca> wrote:
>>>>Corey, this is strange. That line in my copy of map_query is nowhere
>>>>where this error could come from.
>>>>Can you send me your map_query.php please.
>>>>Corey Puffalt wrote:
>>>>>Every now and again I'm getting the following error appearing when
>>>>>using the Query widget. I'm not yet sure what's triggering the
>>>>>problem though it does appear specific to certain queryable layers.
>>>>>The following error message appears in the Query pop-up window:
>>>>>Fatal error: [MapServer Error]: msQueryByPoint(): No matching
>>>>>record(s) found. in
>>>>>on line 561
> Chameleon mailing list
> Chameleon at lists.maptools.org
|Paul Spencer pspencer at dmsolutions.ca |
|Applications & Software Development |
|DM Solutions Group Inc http://www.dmsolutions.ca/|
More information about the Chameleon