for any info/changes follow me: @nickmilon

mongoUtils.aggregation module

aggregation operations

class mongoUtils.aggregation.Aggregation(collection, pipeline=None, **kwargs)

Bases: object

a helper for constructing aggregation pipelines see: aggregation framework supports all aggregation operators

Parameters:

• collection (obj) – a pymongo collection object
• pipeline (list) – (optional) an initial pipeline list
• kwargs (dict) – (optional) any arguments

Returns:

an aggregation object

Example:

>>> from pymongo import MongoClient;from mongoUtils.configuration import testDbConStr  # import MongoClient
>>> db = MongoClient(testDbConStr).get_default_database()                              # get test database
>>> aggr_obj = Aggregation(db.muTest_tweets_users, allowDiskUse=True)                  # select users collection
>>> aggr_obj.help()                                                                    # ask for help
['project', 'match', 'redact', 'limit', .... ]                                         # available operators
>>> aggr_obj.match({'lang': 'en'})                                                     # match English speaking
>>> aggr_obj.group({'_id': None, "avg_followers": {"$avg": "$followers_count"}})       # get average followers
>>> print(aggr_obj.code(False))                                                        # print pipeline
[{"$match": {"lang": "en"}},{"$group": {"avg_followers":
{"$avg": "$followers_count"},"_id": null}}]
>>> next(aggr_obj())                                                                   # execute and get results
{u'avg_followers': 2943.8210227272725, u'_id': None})                                  # results

_operators = ['project', 'match', 'redact', 'limit', 'skip', 'sort', 'unwind', 'group', 'out', 'geoNear']

_frmt_str = '{}\nstage#= {:2d}, operation={}'

__init__(collection, pipeline=None, **kwargs)

classmethod construct_fields(fields_list=[])

a constructor for fields

classmethod construct_stats(fields_lst, _id=None, stats=['avg', 'max', 'min'], incl_count=True)

a constructor helper for group statistics

Parameters:

• fields_lst: (list) list of field names
• stats: (list) list of statistics
• incl_count: (Bool) includes a count if True

Example:

>>> specs_stats(['foo'])
{'max_foo': {'$max': '$foo'}, '_id': None, 'avg_foo': {'$avg': '$foo'}, 'min_foo': {'$min': '$foo'}}

pipeline

returns the pipeline (a list)

classmethod help(what='operators')

returns list of available operators

add(operator, value, position=None)

adds an operation at specified position in pipeline

search(operator, count=1)

returns (position, operator

save(file_pathname)

save pipeline list to file

remove(position)

remove an element from pipeline list given its position

code(verbose=True)

clear()

__call__(print_n=None, **kwargs)

perform the aggregation when called >>> Aggregation_object()

for kwargs see: aggregate

Parameters:

• print_n:

  • True: will print results and will return None
  • None: will cancel result printing
  • int: will print top n documents

• kwargs: if any of kwargs are specified override any arguments provided on instance initialization.

class mongoUtils.aggregation.AggrCounts(collection, field, match=None, sort={'count': -1}, **kwargs)

Bases: mongoUtils.aggregation.Aggregation

constructs a group count aggregation pipeline based on Aggregation class

Parameters:

• collection (obj) – a pymongo collection object
• field (str) – field name
• match (dict) – a query match expression, defaults to None
• sort (dict) – a sort expression defaults to {‘count’: -1}
• kwargs (dict) – optional arguments to pass to parent Aggregation

Example:

>>> from pymongo import MongoClient;from mongoUtils.configuration import testDbConStr  # import MongoClient
>>> db = MongoClient(testDbConStr).get_default_database()                              # get test database
>>> AggrCounts(db.muTest_tweets_users, "lang",  sort={'count': -1})(verbose=True)      # counts by language
{u'count': 352, u'_id': u'en'}
{u'count': 283, u'_id': u'ja'}
{u'count': 100, u'_id': u'es'}  ...

__init__(collection, field, match=None, sort={'count': -1}, **kwargs)