def __init__(
self,
left: pd.DataFrame,
right: pd.DataFrame,
on=None,
left_on=None,
right_on=None,
merge_suffixes=get_option("operators.binary.column_suffixes"),
add_suffixes=False,
add_indexes=(None, None),
):
self.left = self.orig_left = left
self.right = self.orig_right = right
self.on = lltools.maybe_make_list(on)
self.left_on = lltools.maybe_make_list(left_on)
self.right_on = lltools.maybe_make_list(right_on)
self.merge_suffixes = merge_suffixes
self.add_suffixes = validatortools.validate_bool_kwarg(
add_suffixes, "add_suffixes")
self.add_indexes = add_indexes
self._validate_specification()
def replace_tag(self, old_tag, new_tag):
"""Replace ``old_tag`` with ``new_tag``."""
if set(old_tag).issubset(set(self._tags)):
old_tag = lltools.maybe_make_list(old_tag)
new_tag = lltools.maybe_make_list(new_tag)
for ot in old_tag:
self._tags.remove(ot)
self._tags.extend(new_tag)
def keep_cols(self, cols, inplace=False):
"""Keep specified columns (thus dropping the rest).
Note: `Key` columns will always be kept.
"""
cols_to_keep = self.key_cols
# preserve order of the key columns by sorting them according
# to existing order
cols_to_keep.sort(key=lambda i: self.columns.tolist().index(i))
cols = lltools.maybe_make_list(cols)
cols_to_keep.extend([c for c in cols if c not in cols_to_keep])
result = self[cols_to_keep]
if inplace:
return self._update_inplace(result)
else:
return result
def mappers(self, mappers):
self._mappers = []
mappers = lltools.maybe_make_list(mappers)
for mapper in mappers:
self.add_mapper(mapper)
def test_maybe_make_list():
assert lltools.maybe_make_list("a") == ["a"]
assert lltools.maybe_make_list(["a"]) == ["a"]
assert lltools.maybe_make_list(("a",)) == ("a",)
def __init__(
self,
left: pd.DataFrame,
right: pd.DataFrame,
id_on=None,
id_left_on=None,
id_right_on=None,
date_on=None,
date_left_on=None,
date_right_on=None,
get: str = "all",
when: str = "earlier_or_later",
days: int = 90,
left_link_id=None,
dropna: bool = False,
drop_duplicates: bool = False,
duplicates_indicator: bool = False,
merge="partial",
merge_suffixes=get_option("operators.binary.column_suffixes"),
prepend_levels=(None, None),
):
self.left = left
self.right = right
self.id_on = lltools.maybe_make_list(id_on)
self.id_left_on = lltools.maybe_make_list(id_left_on)
self.id_right_on = lltools.maybe_make_list(id_right_on)
self.date_on = date_on
self.date_left_on = date_left_on
self.date_right_on = date_right_on
self.get = get
self.when = when
self.days = days
self.left_link_id = left_link_id
self.dropna = validatortools.validate_bool_kwarg(dropna, "dropna")
self.drop_duplicates = validatortools.validate_bool_kwarg(
drop_duplicates, "drop_duplicates"
)
self.duplicates_indicator = duplicates_indicator
self.duplicates_indicator_name: Optional[str]
if isinstance(self.duplicates_indicator, str):
self.duplicates_indicator_name = self.duplicates_indicator
elif isinstance(self.duplicates_indicator, bool):
self.duplicates_indicator_name = (
get_option("column.system.duplicates") if self.duplicates_indicator else None
)
else:
raise ValueError("indicator option can only accept boolean or string arguments")
self.merge = merge
self.merge_suffixes = merge_suffixes
self.prepend_levels = prepend_levels
self._left_suffix = get_option("operators.binary.column_suffixes")[0]
self._right_suffix = get_option("operators.binary.column_suffixes")[1]
self._diff_days_col = get_option("column.system.diff_days")
self._abs_diff_days_col = get_option("column.system.abs_diff_days")
self._merge_indicator_col = get_option("column.system.merge")
self._validate_specification()