Bdtjtk

I have mongodb data like:

{'word': 'good', 'info': [{'tbl_id': 'd1', 'term_freq': 2}, {'tbl_id': 'd2', 'term_freq': 56}, {'tbl_id': 'd3', 'term_freq': 3}]}

{'word': 'spark', 'info': [{'tbl_id': 'd1', 'term_freq': 6}, {'tbl_id': 'd3', 'term_freq': 11}, {'tbl_id': 'd4', 'term_freq': 10}]}

{'word': 'good', 'info': [{'tbl_id': 'd4', 'term_freq': 12}, {'tbl_id': 'd5', 'term_freq': 8}, {'tbl_id': 'd8', 'term_freq': 7}]}

{'word': 'spark', 'info': [{'tbl_id': 'd5', 'term_freq': 6}, {'tbl_id': 'd6', 'term_freq': 11}, {'tbl_id': 'd7', 'term_freq': 10}]}

and I want to use pymongo to process it, the result should be:

{'word': 'good',

 'info': [{'tbl_id': 'd1', 'term_freq': 2}, {'tbl_id': 'd2', 'term_freq': 56}, {'tbl_id': 'd3', 'term_freq': 3},

          {'tbl_id': 'd4', 'term_freq': 12}, {'tbl_id': 'd5', 'term_freq': 8}, {'tbl_id': 'd8', 'term_freq': 7}]}

{'word': 'spark',

 'info': [{'tbl_id': 'd1', 'term_freq': 6}, {'tbl_id': 'd3', 'term_freq': 11}, {'tbl_id': 'd4', 'term_freq': 10},

          {'tbl_id': 'd5', 'term_freq': 6}, {'tbl_id': 'd6', 'term_freq': 11}, {'tbl_id': 'd7', 'term_freq': 10}]}

I use group in pymongo:

a = mycol.aggregate([{"$group": {"_id":"$word", 'infos': {"$concatArrays": 1}}}])

for i in a:

    print(i)

It went wrong: pymongo.errors.OperationFailure: unknown group operator '$concatArrays'.
and I use group keyword:

a = mycol.group(key='word',condition=None, initial={'infos': }, reduce={"$concatArrays": "info"})

for i in a:

    print(i)

It also went wrong:

Traceback (most recent call last):File "F:/programs/SearchEngine/test.py", line 167, in <module> a = mycol.group(key='word',condition=None, initial={'infos': }, reduce={"$concatArrays": "info"})  File "C:Usersll.virtualenvsSearchEnginelibsite-packagespymongocollection.py", line 2550, in group  group["$reduce"] = Code(reduce)  File "C:Usersll.virtualenvsSearchEnginelibsite-packagesbsoncode.py", line 54, in __new__  "instance of %s" % (string_type.__name__))

TypeError: code must be an instance of str

The reason you are getting this error message is because the $concatArrays operator is an expression operator not a $group accumulator.

That being said, you can do this with the following pipeline:

[

    {

        "$group": {

            "_id": "$word",

            "info": {

                "$push": "$info"

            }

        }

    },

    {

        "$project": {

            "_id": 0,

            "word": "$_id",

            "info": {

                "$reduce": {

                    "input": "$info",

                    "initialValue": [



                    ],

                    "in": {

                        "$concatArrays": [

                            "$$value",

                            "$$this"

                        ]

                    }

                }

            }

        }

    }

]

We create a 2d list of info in the $group stage with the $push operator then in the another $project stage you flatten the list using the $reduce and $concatArrays.