I would suggest that at long as you can handle strict compliance, 
anything else you can handle should be deprecated, but allowed. I 
wouldnt even invest in documenting it as an allowed feature, and it 
definitely should not appear in any demos or how-tos.

It would be nice if there was feedback to clients sending non-compliant 
requests: e.g. always include a comment in the response indicating a 
loose interpretation was invoked?
It would be good to log in a noce way the client agent and source for 
non-compliant requests: this adds a tiny little bit of pain for the 
serivice owner to encourage clients to get fixed.

Rob

Jesse Eichar wrote:
> Hey Justin,
>
> I am making this "relaxed" compliance and "strict" compliance change  
> because otherwise I have to change all of Geotools and that is  
> completely out of the question... yet I need WFS to be able to be  
> compliant and still be able to communicate with Geoserver.  So right  
> there I need both compliance settings.  The WFS client is going to be  
> strategized as is the WMS client.  It is a lot of work for me but it  
> is less than changing all of Geotools just for this requirement.
>
> However, for trunk we have to decide what to do here.  Are we going  
> to have strict compliance or are we going to have 2+ levels of  
> compliance?  I honestly can't decide because both have their merits.
>
> Jesse
>
>
>
> On 4-Aug-06, at 10:02 AM, Justin Deoliveira wrote:
>
>   
>> Hi Jesse,
>>
>> Sounds like you are having fun with the filter spec :). I am  
>> working on gtxml bindings for filter 1.1 parsing so this somewhat  
>> applies to me as well.
>>
>> This is a tough call to make. I kind of feel uncomfortable forcing  
>> this type of logic onto the client. To have to decide to try and  
>> use a "relaxed" filter encoder or to use a standards compliant one  
>> is a lot of work. It sounds like you need to know before hand what  
>> type of server you are dealing with, I wouldn't think this is a  
>> decision that a general client would want to make. And it would be  
>> difficult to try it on the fly as well as I would imagine you would  
>> need to make the decision based on a failure, which is probably not  
>> the most reliable way to do it.
>>
>> Have you seen this behaviour in any other servers. If it pops up in  
>> multple places then I would say the hack might be justified, but if  
>> its just for a single server I wouldn't say so.
>>
>> I know its probably just a little change but i have seen the result  
>> of a "little hack" turn into a piece of code that our codebase  
>> becomes depenent on. Hence my purist attitude.
>>
>> Again touch call to make, definitley a place where the spec is a  
>> bit lacking. I am sure this thread will start the ever so popular  
>> strict standard vs lax standard war.
>>
>> -Justin
>>
>>
>> Jesse Eichar wrote:
>>     
>>> The same issue exists in Filter 1.1.  And using a function won't  
>>> help because third party servers won't support the function.  I  
>>> working with Ionic's Redspider server and it can't read our  
>>> filters.  I've made up a work around.  I'm adding a hint to the  
>>> XML Filter Encoding so that you can set the compliance level.   
>>> When communicating with Geoserver I use the default compliance  
>>> Ionic is another level (medium) and then there is strict.   Ionic  
>>> allows:
>>>       
>>>>> <Filter>
>>>>> <Or>
>>>>> <BBox> some box </BBox>
>>>>> </Or>
>>>>> <FeatureId fid="fid" />
>>>>> </Filter>
>>>>>           
>>> Which is still not legal but is very useful.   In the case of NOT 
>>> ( FidFilter ) I'm doing the processing on the client.
>>> We will have to think about how we want to handle this issue in  
>>> general though.
>>> Jesse
>>> On 4-Aug-06, at 12:50 AM, Jody Garnett wrote:
>>>       
>>>> Jesse can you please confirm against Filter 1.1? What you say  
>>>> sounds pretty silly, if needed we can work around it with a  
>>>> function :-(
>>>> Also note that Filter is available in a BNF form as part of the  
>>>> catalog specification - we really need to take a common ground,  
>>>> and police what is created through the use of specification  
>>>> specific Factories.
>>>>
>>>> I would also ask Justin for comment as he has spent a lot of time  
>>>> on the ground with Filter.
>>>> Jody
>>>>
>>>>         
>>>>> Hi,
>>>>>
>>>>> I've been browsing through the filter spec and to my dismay  
>>>>> have  learned that the following is not legal:
>>>>>
>>>>> <Filter>
>>>>> <Or>
>>>>> <BBox> some box </BBox>
>>>>> <FeatureId fid="fid" />
>>>>> </Or>
>>>>> </Filter>
>>>>>
>>>>> To be compliant with the spec it is supposed to be:
>>>>>
>>>>> <Filter>
>>>>> <Or>
>>>>> <BBox> some box </BBox>
>>>>> </Or>
>>>>> <FeatureId fid="fid" />
>>>>> </Filter>
>>>>>
>>>>>
>>>>> We also can not do the following:
>>>>>
>>>>> <Filter>
>>>>> <Not>
>>>>> <FeatureId fid="fid"/>
>>>>> </Not>
>>>>> </Filter>
>>>>>
>>>>>
>>>>> Our filter implementation allows both so we can't interoperate   
>>>>> properly with others because we can make illegal documents....
>>>>>
>>>>> Looking for comments on what should be done here.
>>>>>
>>>>> Jesse
>>>>>
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------- 
>>>>> ------
>>>>> Take Surveys. Earn Cash. Influence the Future of IT
>>>>> Join SourceForge.net's Techsay panel and you'll get the chance  
>>>>> to share your
>>>>> opinions on IT & business topics through brief surveys -- and  
>>>>> earn cash
>>>>> http://www.techsay.com/default.php? 
>>>>> page=join.php&p=sourceforge&CID=DEVDEV <http://www.techsay.com/ 
>>>>> default.php?page=join.php&p=sourceforge&CID=DEVDEV>
>>>>> _______________________________________________
>>>>> Geotools-devel mailing list
>>>>> Geotools-devel@lists.sourceforge.net <mailto:Geotools- 
>>>>> [EMAIL PROTECTED]>
>>>>> https://lists.sourceforge.net/lists/listinfo/geotools-devel
>>>>>
>>>>>
>>>>>           
>>>>         
>>> !DSPAM:1004,44d364d9186132207481331!
>>> --------------------------------------------------------------------- 
>>> ---
>>> --------------------------------------------------------------------- 
>>> ----
>>> Take Surveys. Earn Cash. Influence the Future of IT
>>> Join SourceForge.net's Techsay panel and you'll get the chance to  
>>> share your
>>> opinions on IT & business topics through brief surveys -- and earn  
>>> cash
>>> http://www.techsay.com/default.php? 
>>> page=join.php&p=sourceforge&CID=DEVDEV
>>> !DSPAM:1004,44d364d9186132207481331!
>>> --------------------------------------------------------------------- 
>>> ---
>>> _______________________________________________
>>> Geotools-devel mailing list
>>> Geotools-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/geotools-devel
>>> !DSPAM:1004,44d364d9186132207481331!
>>>       
>> -- 
>> Justin Deoliveira
>> The Open Planning Project
>> [EMAIL PROTECTED]
>>     
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Geotools-devel mailing list
> Geotools-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/geotools-devel
>   



-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Geotools-devel mailing list
Geotools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to