Referencing previous iterations while in p:for-each

> I have an XPL question I've been scratching my head over for a while,
> and I wonder if anyone can help.
>
> I have an OPS-based app, MVC-based, that assembles its model by
> executing a "model" document that looks something like this:
>
> <model>  <!-- contains 1..n "datasource" elements -->
>     <datasource>
>         <!-- grab something from a web service -->
>     </datasource>
>     <datasource>
>         <!-- use the SQL processor to do a database query -->
>     </datasource>
> </model>
>
> This document gets fed to a pipeline built around this loop:
>
>     <p:for-each href="#model-doc" select="/model/datasource" root="data"
> id="aggregated-data">
>
>         <!-- Execute the code in each "datasource" element; the details
> don't matter here -->
>         <p:processor name="oxf:pipeline">
>             <p:input href="current()#xpointer(/datasource/p:config)"/>
>             <!-- other inputs -->
>             <p:output ref="aggregated-data"/>
>         </p:processor>
>
> </p:for-each>
>
> Here's my problem:
>
> Occasionally, I have a case where I need the results of one <datasource>
>  in order to execute the next one. For example, I might do a search
> against a web service in the first <datasource>, and then need something
> from the search results to use as a key in a database query in the next
> <datasource>.
>
> However, once an iteration of the <p:for-each> loop is completed, I
> don't see any way to get at its results when I move on to the next
> iteration. I apparently can't take the results of my web service call,
> for example, and present it to the SQL query as an input.
>
> Does anyone see a way to do this? Have I missed something?
>
> The only solution I've come up with so far is to use the scope
> serializer at the end of each iteration to write the results of each
> iteration into the request, and then use the scope generator at the
> beginning of each iteration to retrieve all the data from all the
> previous iterations. This could probably be made to work, but it would
> be very inefficient, since I really only need previous-iteration data
> about 10% of the time, and I probably can't afford that much of a
> performance hit.
>
> So... can anyone suggest anything?

> Jeff Jones wrote:
>> I have an XPL question I've been scratching my head over for a while,
>> and I wonder if anyone can help.
>>
>> I have an OPS-based app, MVC-based, that assembles its model by
>> executing a "model" document that looks something like this:
>>
>> <model>  <!-- contains 1..n "datasource" elements -->
>>     <datasource>
>>         <!-- grab something from a web service -->
>>     </datasource>
>>     <datasource>
>>         <!-- use the SQL processor to do a database query -->
>>     </datasource>
>> </model>
>>
>> This document gets fed to a pipeline built around this loop:
>>
>>     <p:for-each href="#model-doc" select="/model/datasource"
>> root="data" id="aggregated-data">
>>
>>         <!-- Execute the code in each "datasource" element; the
>> details don't matter here -->
>>         <p:processor name="oxf:pipeline">
>>             <p:input href="current()#xpointer(/datasource/p:config)"/>
>>             <!-- other inputs -->
>>             <p:output ref="aggregated-data"/>
>>         </p:processor>
>>
>> </p:for-each>
>>
>> Here's my problem:
>>
>> Occasionally, I have a case where I need the results of one
>> <datasource>  in order to execute the next one. For example, I might
>> do a search against a web service in the first <datasource>, and then
>> need something from the search results to use as a key in a database
>> query in the next <datasource>.
>>
>> However, once an iteration of the <p:for-each> loop is completed, I
>> don't see any way to get at its results when I move on to the next
>> iteration. I apparently can't take the results of my web service call,
>> for example, and present it to the SQL query as an input.
>>
>> Does anyone see a way to do this? Have I missed something?
>>
>> The only solution I've come up with so far is to use the scope
>> serializer at the end of each iteration to write the results of each
>> iteration into the request, and then use the scope generator at the
>> beginning of each iteration to retrieve all the data from all the
>> previous iterations. This could probably be made to work, but it would
>> be very inefficient, since I really only need previous-iteration data
>> about 10% of the time, and I probably can't afford that much of a
>> performance hit.
>>
>> So... can anyone suggest anything?
>
> You are right that in each iteration, current() returns a new XML
> document which does not provide access to previous iterations.
>
> You could work around this by first transforming your model document to
> contain things like:
>
> <model>
>   <iteration>
>     <datasource>
>
>     </datasource>
>     <datasource>
>
>     </datasource>
>   </iteration>
>   <iteration>
>     <datasource>
>
>     </datasource>
>     <datasource>
>
>     </datasource>
>   </iteration>
>   ...
> </model>
>
> Then you would iterate over <iteration>, and therefore each iteration
> would have all the data needed.
>
> -Erik
>
>
> ------------------------------------------------------------------------
>
>
> --
> You receive this message as a subscriber of the [hidden email] mailing list.
> To unsubscribe: mailto:[hidden email]
> For general help: mailto:[hidden email]?subject=help
> ObjectWeb mailing lists service home page: http://www.objectweb.org/wws