Suggestion for a AggregateDocument function

[rodrigomurta]

Hi Zach,
Do you have plans to create a AggregateDocument function?
It would be really cool if I could do things like:

AggregateDocument[ myColletion
        ,{"$project" ->  {"author" -> 1}}
        ,{"$group" -> {"_id" -> "$author", "count" -> {"$sum" -> 1}}}
        ,{"$sort" -> {"count" -> -1}}
        ,{"$limit" -> 5}
]

tks
Murta

[zbjornson]

Hadn't gotten that far, but yes, thanks for the suggestion. I'll play with some designs soon.

(If you need that functionality immediately, you can access the the underlying java object for the collection with collection["collection"] and use jlink to call the aggregation commands that you need.)

Semi-related, I keep forgetting to add Sort options to all the query functions (just made #7 for that).

[zbjornson]

Feedback from Mike H on syntax:

Change the OP example to a list of stages:

AggregateDocuments[ myColletion,{
        ,{"$project" ->  {"author" -> 1}}
        ,{"$group" -> {"_id" -> "$author", "count" -> {"$sum" -> 1}}}
        ,{"$sort" -> {"count" -> -1}}
        ,{"$limit" -> 5}
        }
]

or more generally AggregateDocuments[ myColletion, myPipeline], where myPipeline is a list containing one or more pipeline steps which are lists. Each pipeline stage takes the form {$operator -> operand} where the operand is normally a list.