def _count(self):
count_columns = self._non_group_columns()
if len(count_columns) == 0:
count_columns.append('_'.join(self.columns) + '_count')
groupby = {
"$group": {
"_id": {k: "$%s" % k for k in self.columns},
}
}
for k in count_columns:
groupby['$group']['%s' % k] = {"$sum": 1}
pipeline = self._amend_pipeline([groupby])
if self.should_sort:
sort = qops.SORT(**dict(qops.make_sortkey('_id')))
pipeline.append(sort)
return list(self.collection.aggregate(pipeline, allowDiskUse=True))
def _amend_pipeline(self, pipeline):
""" amend pipeline with default ops on coll.aggregate() calls """
if self.sort_order:
sort = qops.SORT(**dict(qops.make_sortkey(self.sort_order)))
pipeline.append(sort)
return pipeline
def merge(self, right, on=None, left_on=None, right_on=None,
how='inner', target=None, suffixes=('_x', '_y'),
sort=False, inspect=False):
"""
merge this dataframe with another dataframe. only left outer joins
are currently supported. the output is saved as a new collection,
target name (defaults to a generated name if not specified).
:param right: the other MDataFrame
:param on: the list of key columns to merge by
:param left_on: the list of the key columns to merge on this dataframe
:param right_on: the list of the key columns to merge on the other
dataframe
:param how: the method to merge. supported are left, inner, right.
Defaults to inner
:param target: the name of the collection to store the merge results
in. If not provided a temporary name will be created.
:param suffixes: the suffixes to apply to identical left and right
columns
:param sort: if True the merge results will be sorted. If False the
MongoDB natural order is implied.
:returns: the MDataFrame to the target MDataFrame
"""
# validate input
supported_how = ["left", 'inner', 'right']
assert how in supported_how, "only %s merges are currently supported" % supported_how
for key in [on, left_on, right_on]:
if key:
assert isinstance(
key, six.string_types), "only single column merge keys are supported (%s)" % key
if isinstance(right, Collection):
right = MDataFrame(right)
assert isinstance(
right, MDataFrame), "both must be MDataFrames, got right=%" % type(right)
if how == 'right':
# A right B == B left A
return right.merge(self, on=on, left_on=right_on, right_on=left_on,
how='left', target=target, suffixes=suffixes)
# generate lookup parameters
on = on or '_id'
right_name = self._get_collection_name_of(right, right)
target_name = self._get_collection_name_of(
target, '_temp.merge.%s' % uuid4().hex)
target_field = (
"%s_%s" % (right_name.replace('.', '_'), right_on or on))
lookup = qops.LOOKUP(right_name,
key=on,
left_key=left_on,
right_key=right_on,
target=target_field)
# unwind merged documents from arrays to top-level document fields
unwind = qops.UNWIND(target_field, preserve=how != 'inner')
# get all fields from left, right
project = {}
for left_col in self.columns:
source_left_col = left_col
if left_col == '_id':
project[left_col] = 1
continue
if left_col.startswith('_idx'):
continue
if left_col.startswith('_om#'):
continue
if left_col != (on or left_on) and left_col in right.columns:
left_col = '%s%s' % (left_col, suffixes[0])
project[left_col] = "$%s" % source_left_col
for right_col in right.columns:
if right_col == '_id':
continue
if right_col.startswith('_idx'):
continue
if right_col.startswith('_om#'):
continue
if right_col == (on or right_on) and right_col == (on or left_on):
# if the merge field is the same in both frames, we already
# have it from left
continue
if right_col in self.columns:
left_col = '%s%s' % (right_col, suffixes[1])
else:
left_col = '%s' % right_col
project[left_col] = '$%s.%s' % (target_field, right_col)
expected_columns = list(project.keys())
project = {"$project": project}
# store merged documents and return an MDataFrame to it
out = qops.OUT(target_name)
pipeline = [lookup, unwind, project]
if sort:
sort_cols = make_list(on or [left_on, right_on])
sort_key = qops.make_sortkey(sort_cols)
sort = qops.SORT(**dict(sort_key))
pipeline.append(sort)
pipeline.append(out)
if inspect:
result = pipeline
else:
result = self.collection.aggregate(pipeline, allowDiskUse=True)
result = MDataFrame(self.collection.database[target_name],
force_columns=expected_columns)
return result