def infer_collection_ordered_helper(l, cands):
for vo in l:
tv = type(vo)
if is_ordered(tv):
if is_list(vo) and is_dict(type(v[0])):
cands.append(vo)
else:
infer_collection_ordered_helper(vo, cands)
elif is_dict(tv):
infer_collection_dict_helper(vo, cands)
def infer_collection_dict_helper(d, cands):
for k, vd in d.iteritems():
tv = type(vd)
if is_ordered(tv):
if is_list(tv) and is_dict(type(vd[0])):
cands.append(vd)
else:
infer_collection_ordered_helper(vd, cands)
elif is_dict(tv):
infer_collection_dict_helper(vd, cands)
def infer_collection(d):
if is_list(type(d)):
return d
cands = []
for k, v in d.iteritems():
tv = type(v)
if is_ordered(tv):
if is_list(tv) and len(v) > 0 and is_dict(type(v[0])):
cands.append(v)
elif is_dict(tv):
infer_collection_dict_helper(v, cands)
if len(cands) > 0:
return sorted(map(lambda c: (c, relsize(c)), cands), key=itemgetter(1), reverse=True)[0][0]
else:
return []
def standardize_tags_helper(d):
nd = {}
for k, v in d.iteritems():
if is_dict(type(v)):
cd = standardize_tags_helper(v)
else:
cd = v
nd[remove_standards_prefix(k)] = cd
return nd
def describe(self):
print "\n"
print "TABLENAME: %s, # RECORDS: %d" % (self.name, self.__len__())
if len(self.rows) == 0:
return
c = self.rows[0]
for k, v in c.iteritems():
print "\tOUTER KEY:", k
if is_dict(type(v)):
print "\t INNER KEYS:"
print "\t ", v.keys(), "\n"
def unnest_value(row):
"""
For data collections which have nested value parameters (like RSS)
this function will unnest the value to the higher level.
For example, say the row is {"title":{"value":"yahoo wins search"}}
This function will take that row and return the following row {"title": "yahoo wins search"}
"""
nr = {}
for k, v in row.iteritems():
if is_dict(type(v)) and "value" in v:
nr[k] = v["value"]
else:
nr[k] = v
return nr
def unnest_value(row):
"""
For data collections which have nested value parameters (like RSS)
this function will unnest the value to the higher level.
For example, say the row is {"title":{"value":"yahoo wins search"}}
This function will take that row and return the following row {"title": "yahoo wins search"}
"""
nr = {}
for k, v in row.iteritems():
if is_dict(type(v)) and "value" in v:
nr[k] = v["value"]
else:
nr[k] = v
return nr
def select(udf, name="", url=None, table=None, data=None, keep_standards_prefix=False):
if table is not None:
tb = table
keep_standards_prefix = True
if len(name) == 0:
name = tb.name
else:
tb = create(name, data=data, url=url, keep_standards_prefix=keep_standards_prefix)
results = []
for d in tb.rows:
try:
value = udf(d)
except KeyError:
try:
value = udf(strip_row(d)[0])
except KeyError:
continue
if is_dict(type(value)):
results.append(value)
return create(name=name, data=results, keep_standards_prefix=keep_standards_prefix)