PostGIS error on 4.8b2

[Sean Gillies]

Ken,

Cc yourself to bug 1443. I committed a fix for it and would  
appreciate confirmation.

Sean

On Dec 10, 2005, at 8:05 PM, Ken Lord wrote:

> Here's a temporary fix that may work for certain users for the USING
> error for PostgreSQL 8.0.3 - PostGIS / mapserver 4.8b2 until the
> problem is patched.
>
> If your dataset is in the same map projection as your output map,
> simply remove the 'USING srid=3Dxxxxx' from the end of your SQL  
> query in
> your DATA string.
>
> You will again be able to use nested select statements for datasets
> based on on-the-fly spatial queries etc.
>
> As Daryl Herzmann appears to have discovered, the current bug prevents
> mapserver from dealing with more than one USING tag in your SQL.
>
> Paul - Instead of waiting til Monday, I replicated the problem by
> upgrading my cariboo-gold.no-ip.info website to MS 4.8b2, to try your
> suggestion of turning on statement logging.
>
> - The log shows the same error seen on the webpage  '.... syntax error
> at or near "USING" at character ...'  So the database is receiving the
> request.
>
> And without the 'USING srid=3Dxxxxx' in the SQL statement, each map
> request crams the log with ~30 lines reffering to the 'find_srid'
> function ... but the map generates properly.
>
> Cheers,
> Ken Lord
> Vancouver BC
>
>
>
> On 12/10/05, Ken Lord <kenlord at gmail.com> wrote:
>> Thanks Daryl,
>>
>> I saw your patch there,  unfortunately I'm one of those hobbled  
>> people
>> living in the windows world, who hasn't learned how to compile the
>> software.
>>
>> So it will be 4.8b3 or beyond for me :)
>>
>> Cheers,
>> Ken Lord
>> Vancouver BC
>>
>> On 12/10/05, Daryl Herzmann <akrherz at iastate.edu> wrote:
>>> On Sat, 10 Dec 2005, Ken Lord wrote:
>>>
>>>> Thanks Howard,
>>>>
>>>> That bug does seem to cover my problem,  I've added a post to the
>>>> report to confirm that it affects more than just postgresql 8.1 +
>>>> mapserver 4.8b2.
>>>>
>>>> Paul, I'll try your suggestion on Monday. I had simplified my SQL
>>>> statement when trying to make it work, evidently I could have
>>>> simplified it more :)
>>>>
>>>> On 12/10/05, Howard Butler <hobu at iastate.edu> wrote:
>>>>> Is <http://mapserver.gis.umn.edu/bugs/show_bug.cgi?id=3D1443>
>>>>> still relevant to this issue (issues with using *both* UNIQUE  
>>>>> and SR=
> ID)?
>>>
>>> Hi Ken,
>>>
>>> I am nearly positive if you apply my patch included in that bug,  
>>> Postgi=
> s
>>> with "USING..." in the data statement will work...
>>>
>>> Hopefully that patch, in some form, is accepted before the next  
>>> release=
> ...
>>>
>>> daryl
>>>
>>> --
>>> /**
>>>  * Daryl Herzmann (akrherz at iastate.edu)
>>>  * Program Assistant -- Iowa Environmental Mesonet
>>>  * http://mesonet.agron.iastate.edu
>>>  */
>>>
>>>
>>

---
Sean Gillies
sgillies at frii dot com
http://zcologia.com/news