def _is_monotonic_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
increasing = _read_arg_or_kwarg(args, 0, kwargs, 'increasing', None)
strict = _read_arg_or_kwarg(args, 1, kwargs, 'strict', False)
items = _read_arg_or_kwarg(args, 2, kwargs, 'items', {k: (increasing, strict) for k in dfcheck})
results = {}
for col, (increasing, strict) in items.items():
s = pd.Index(dfcheck[col])
if increasing:
good = getattr(s, 'is_monotonic_increasing')
elif increasing is None:
good = getattr(s, 'is_monotonic') | getattr(s, 'is_monotonic_decreasing')
else:
good = getattr(s, 'is_monotonic_decreasing')
if strict:
if increasing:
good = good & (s.to_series().diff().dropna() > 0).all()
elif increasing is None:
good = good & ((s.to_series().diff().dropna() > 0).all() |
(s.to_series().diff().dropna() < 0).all())
else:
good = good & (s.to_series().diff().dropna() < 0).all()
results[col] = not good
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _unique_index_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
ret_specd['obj'] = Counter(list(dfcheck.index))
ret_specd['ndframe'] = pd.Series(ret_specd['obj'])
ret_specd['bool'] = not dfcheck.index.is_unique
return _ret_proper_objects(_ret, ret_specd)
def _is_shape_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
shape = _read_required_arg_or_kwarg(args, 0, kwargs, 'shape')
ret_specd['obj'] = "is_shape has no output object"
ret_specd['ndframe'] = "is_shape has no output ndframe"
ret_specd['bool'] = not dfcheck.shape == shape
return _ret_proper_objects(_ret, ret_specd)
def _is_shape_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
shape = _read_required_arg_or_kwarg(args, 0, kwargs, 'shape')
ret_specd['obj'] = "is_shape has no output object"
ret_specd['ndframe'] = "is_shape has no output ndframe"
ret_specd['bool'] = not dfcheck.shape == shape
return _ret_proper_objects(_ret, ret_specd)
def _none_missing_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
columns = _read_arg_or_kwarg(args, 0, kwargs, 'columns', dfcheck.columns)
ret_specd['obj'] = dfcheck[columns].isnull()
ret_specd['ndframe'] = ret_specd['obj'].any()
ret_specd['bool'] = ret_specd['ndframe'].any()
return _ret_proper_objects(_ret, ret_specd)
def _equal_columns_sum_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
cola = _read_arg_or_kwarg(args, 0, kwargs, 'cola', dfcheck.columns[0])
colb = _read_arg_or_kwarg(args, 1, kwargs, 'colb', dfcheck.columns[1])
results = {cola : dfcheck[cola].dropna().sum(),
colb : dfcheck[colb].dropna().sum()}
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = not results[cola] == results[colb]
return _ret_proper_objects(_ret, ret_specd)
def _within_set_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
for k, v in items.items():
results[k] = not dfcheck[k].isin(v).all()
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _has_dtypes_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
dtypes = dfcheck.dtypes
for k, v in items.items():
results[k] = not dtypes[k] == v
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _within_range_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
for k, (lower, upper) in items.items():
results[k] = (lower > dfcheck[k]).any() or (upper < dfcheck[k]).any()
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _within_set_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
for k, v in items.items():
results[k] = not dfcheck[k].isin(v).all()
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _has_dtypes_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
dtypes = dfcheck.dtypes
for k, v in items.items():
results[k] = not dtypes[k] == v
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _within_range_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
items = _read_required_arg_or_kwarg(args, 0, kwargs, 'items')
results = {}
for k, (lower, upper) in items.items():
results[k] = (lower > dfcheck[k]).any() or (upper < dfcheck[k]).any()
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = any(list(results.values()))
return _ret_proper_objects(_ret, ret_specd)
def _within_n_std_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
n = _read_arg_or_kwarg(args, 0, kwargs, 'n', 3)
means = dfderive.mean()
stds = dfderive.std()
results = (pd.np.abs(dfcheck - means) < n * stds)
ret_specd['obj'] = results
ret_specd['ndframe'] = results
ret_specd['bool'] = not results.all().all()
return _ret_proper_objects(_ret, ret_specd)
def _within_n_std_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
n = _read_arg_or_kwarg(args, 0, kwargs, 'n', 3)
means = dfderive.mean()
stds = dfderive.std()
results = (pd.np.abs(dfcheck - means) < n * stds)
ret_specd['obj'] = results
ret_specd['ndframe'] = results
ret_specd['bool'] = not results.all().all()
return _ret_proper_objects(_ret, ret_specd)
def _equal_columns_sum_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
cola = _read_arg_or_kwarg(args, 0, kwargs, 'cola', dfcheck.columns[0])
colb = _read_arg_or_kwarg(args, 1, kwargs, 'colb', dfcheck.columns[1])
results = {
cola: dfcheck[cola].dropna().sum(),
colb: dfcheck[colb].dropna().sum()
}
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = not results[cola] == results[colb]
return _ret_proper_objects(_ret, ret_specd)
def _has_in_index_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
orig_obj_to_check = _read_required_arg_or_kwarg(args, 0, kwargs, 'obj')
obj_to_check = orig_obj_to_check
try_ix = _read_arg_or_kwarg(args, 1, kwargs, 'try_ix', False)
try_strftime = _read_arg_or_kwarg(args, 2, kwargs, 'try_strftime', False)
check_na = _read_arg_or_kwarg(args, 3, kwargs, 'check_na', False)
ans = obj_to_check in dfcheck.index
if try_strftime:
try:
obj_to_check = obj_to_check.strftime(try_strftime)
except:
pass
# at this point, if it's in the index, ans will be True
if try_ix:
try:
ans = ans or (len(dfcheck.ix[obj_to_check]) > 0)
except:
pass
# at this point, if it's in the index, or if it's string representation
# is in the index, ans will be True
if check_na:
try:
isna = dfcheck.ix[obj_to_check].isnull()[0]
except:
isna = True
ans = not ((not ans) or isna)
results = {orig_obj_to_check : ans}
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = not ans
return _ret_proper_objects(_ret, ret_specd)
def _has_in_index_ret(dforig, dfcheck, dfderive, *args, **kwargs):
_ret, ret_specd = _pull_out_ret(kwargs, dforig)
orig_obj_to_check = _read_required_arg_or_kwarg(args, 0, kwargs, 'obj')
obj_to_check = orig_obj_to_check
try_ix = _read_arg_or_kwarg(args, 1, kwargs, 'try_ix', False)
try_strftime = _read_arg_or_kwarg(args, 2, kwargs, 'try_strftime', False)
check_na = _read_arg_or_kwarg(args, 3, kwargs, 'check_na', False)
ans = obj_to_check in dfcheck.index
if try_strftime:
try:
obj_to_check = obj_to_check.strftime(try_strftime)
except:
pass
# at this point, if it's in the index, ans will be True
if try_ix:
try:
ans = ans or (len(dfcheck.ix[obj_to_check]) > 0)
except:
pass
# at this point, if it's in the index, or if it's string representation
# is in the index, ans will be True
if check_na:
try:
isna = dfcheck.ix[obj_to_check].isnull()[0]
except:
isna = True
ans = not ((not ans) or isna)
results = {orig_obj_to_check: ans}
ret_specd['obj'] = results
ret_specd['ndframe'] = pd.Series(results)
ret_specd['bool'] = not ans
return _ret_proper_objects(_ret, ret_specd)