added mypy settings

myresources/crocodile/core.py

@@ -108,7 +108,9 @@ def get_attributes(self, remove_base_attrs: bool = True, return_objects: bool =
         return List([getattr(self, x) for x in attrs]) if return_objects else List(attrs)
     def print(self, dtype: bool = False, attrs: bool = False, **kwargs: Any): return Struct(self.__dict__).update(attrs=self.get_attributes() if attrs else None).print(dtype=dtype, **kwargs)
     @staticmethod
-    def get_state(obj: Any, repr_func: Callable[[Any], str] = lambda x: x, exclude: Optional[list[str]] = None) -> dict[str, Any]: return repr_func(obj) if not any([hasattr(obj, "__getstate__"), hasattr(obj, "__dict__")]) else (tmp if type(tmp := obj.__getstate__() if hasattr(obj, "__getstate__") else obj.__dict__) is not dict else Struct(tmp).filter(lambda k, v: k not in (exclude or [])).apply2values(lambda k, v: Base.get_state(v, exclude=exclude, repr_func=repr_func)).__dict__)
+    def get_state(obj: Any, repr_func: Callable[[Any], dict[str, Any]] = lambda x: x, exclude: Optional[list[str]] = None) -> dict[str, Any]:
+        if not any([hasattr(obj, "__getstate__"), hasattr(obj, "__dict__")]): return repr_func(obj)
+        return (tmp if type(tmp := obj.__getstate__() if hasattr(obj, "__getstate__") else obj.__dict__) is not dict else Struct(tmp).filter(lambda k, v: k not in (exclude or [])).apply2values(lambda k, v: Base.get_state(v, exclude=exclude, repr_func=repr_func)).__dict__)
     def viz_composition_heirarchy(self, depth: int = 3, obj: Any = None, filt: Optional[Callable[[Any], None]] = None):
         install_n_import("objgraph").show_refs([self] if obj is None else [obj], max_depth=depth, filename=str(filename := Path(__import__("tempfile").gettempdir()).joinpath("graph_viz_" + randstr(noun=True) + ".png")), filter=filt)
         _ = __import__("os").startfile(str(filename.absolute())) if __import__("sys").platform == "win32" else None; return filename
@@ -142,11 +144,20 @@ def __call__(self, *args: Any, **kwargs: Any) -> 'List[Any]': return List([i.__c
     # ======================== Access Methods ==========================================
     def __setitem__(self, key: int, value: T) -> None: self.list[key] = value
     def sample(self, size: int = 1, replace: bool = False, p: Optional[list[float]] = None) -> 'List[T]':
-        return self[list(__import__("numpy").random.choice(len(self), size, replace=replace, p=p))]
+        import numpy as np
+        tmp = np.random.choice(len(self), size, replace=replace, p=p)
+        return List([self.list[item] for item in tmp.tolist()])
     def split(self, every: int = 1, to: Optional[int] = None) -> 'List[List[T]]':
         every = every if to is None else __import__("math").ceil(len(self) / to)
-        return List([(self[ix:ix + every] if ix + every < len(self) else self[ix:len(self)]) for ix in range(0, len(self), every)])
-    def filter(self, func: Callable[[T], bool], which: Optional[Callable[[int, T], T2]] = lambda _idx, x: x) -> 'List[T2]': return List([which(idx, x) for idx, x in enumerate(self.list) if func(x)])
+        res: list[List[T]] = []
+        for ix in range(0, len(self), every):
+            if ix + every < len(self):
+                tmp = self.list[ix:ix + every]
+            else:
+                tmp = self.list[ix:len(self)]
+            res.append(List(tmp))
+        return List(res)
+    def filter(self, func: Callable[[T], bool], which: Callable[[int, T], T2] = lambda _idx, x: x) -> 'List[T2]': return List([which(idx, x) for idx, x in enumerate(self.list) if func(x)])
     # ======================= Modify Methods ===============================
     def reduce(self, func: Callable[[T, T], T] = lambda x, y: x + y, default: Optional[T] = None) -> list[T]:
         args = (func, self.list) + ((default,) if default is not None else ())
@@ -159,7 +170,10 @@ def sort(self, key=None, reverse: bool = False) -> 'List[T]': self.list.sort(key
     def sorted(self, *args: list[Any], **kwargs: Any) -> 'List[T]': return List(sorted(self.list, *args, **kwargs))
     def insert(self, __index: int, __object: T): self.list.insert(__index, __object); return self
     # def modify(self, expr: str, other: Optional['List[T]'] = None) -> 'List[T]': _ = [exec(expr) for idx, x in enumerate(self.list)] if other is None else [exec(expr) for idx, (x, y) in enumerate(zip(self.list, other))]; return self
-    def remove(self, value: Optional[T] = None, values: Optional[list[T]] = None, strict: bool = True) -> 'List[T]': _ = [self.list.remove(a_val) for a_val in ((values or []) + ([value] if value else [])) if strict or value in self.list]; return self
+    def remove(self, value: Optional[T] = None, values: Optional[list[T]] = None, strict: bool = True) -> 'List[T]':
+        for a_val in ((values or []) + ([value] if value else [])):
+            if strict or value in self.list: self.list.remove(a_val)
+        return self
     def to_series(self): return __import__("pandas").Series(self.list)
     def to_list(self) -> list[T]: return self.list
     def to_numpy(self, **kwargs: Any) -> 'Any': import numpy as np; return np.array(self.list, **kwargs)
@@ -172,10 +186,16 @@ def __getitem__(self, key: Union[int, list[int], 'slice']) -> Union[T, 'List[T]'
         elif isinstance(key, int): return self.list[key]
         # assert isinstance(key, slice)
         return List(self.list[key])  # for slices  # type: ignore
-    def apply(self, func: Callable[[T], T2], *args: Any, other: Optional['List[T]'] = None, filt: Optional[Callable[[T], bool]] = lambda x: True, jobs: Optional[int] = None, prefer: Optional[str] = [None, 'processes', 'threads'][0], depth: int = 1, verbose: bool = False, desc: Optional[str] = None, **kwargs: Any) -> 'List[T2]':
+    def apply(self, func: Callable[[T], T2], *args: Any, other: Optional['List[T]'] = None, filt: Callable[[T], bool] = lambda x: True, jobs: Optional[int] = None, prefer: Optional[str] = [None, 'processes', 'threads'][0], depth: int = 1, verbose: bool = False, desc: Optional[str] = None, **kwargs: Any) -> 'List[T2]':
         if depth > 1: self.apply(lambda x: x.apply(func, *args, other=other, jobs=jobs, depth=depth - 1, **kwargs))
-        iterator = (self.list if not verbose else install_n_import("tqdm").tqdm(self.list, desc=desc)) if other is None else (zip(self.list, other) if not verbose else install_n_import("tqdm").tqdm(zip(self.list, other), desc=desc))
-        if jobs: from joblib import Parallel, delayed; return List(Parallel(n_jobs=jobs, prefer=prefer)(delayed(func)(x, *args, **kwargs) for x in iterator)) if other is None else List(Parallel(n_jobs=jobs, prefer=prefer)(delayed(func)(x, y) for x, y in iterator))
+        from tqdm import tqdm
+        if other is None:
+            iterator = (self.list if not verbose else tqdm(self.list, desc=desc))
+        else: iterator = (zip(self.list, other) if not verbose else tqdm(zip(self.list, other), desc=desc))
+        if jobs:
+            from joblib import Parallel, delayed
+            if other is None: return List(Parallel(n_jobs=jobs, prefer=prefer)(delayed(func)(x, *args, **kwargs) for x in iterator))  # type: ignore
+            return List(Parallel(n_jobs=jobs, prefer=prefer)(delayed(func)(x, y) for x, y in iterator))  # type: ignore
         return List([func(x, *args, **kwargs) for x in iterator if filt(x)]) if other is None else List([func(x, y) for x, y in iterator])
     def to_dataframe(self, names: Optional[list[str]] = None, minimal: bool = False, obj_included: bool = True):
         df = __import__("pandas").DataFrame(columns=(['object'] if obj_included or names else []) + list(self.list[0].__dict__.keys()))
@@ -189,7 +209,7 @@ def to_dataframe(self, names: Optional[list[str]] = None, minimal: bool = False,
 class Struct(Base):  # inheriting from dict gives `get` method, should give `__contains__` but not working. # Inheriting from Base gives `save` method.
     """Use this class to keep bits and sundry items. Combines the power of dot notation in classes with strings in dictionaries to provide Pandas-like experience"""
     def __init__(self, dictionary: Union[dict[Any, Any], Type[object], None] = None, **kwargs: Any):
-        if dictionary is None or isinstance(dictionary, dict): final_dict = dict() if dictionary is None else dictionary
+        if dictionary is None or isinstance(dictionary, dict): final_dict = {} if dictionary is None else dictionary
         else:
             final_dict = (dict(dictionary) if dictionary.__class__.__name__ == "mappingproxy" else dictionary.__dict__)  # type: ignore
         final_dict.update(kwargs)  # type ignore
@@ -254,7 +274,11 @@ def print(self, dtype: bool = True, return_str: bool = False, justify: int = 30,
         if not bool(self): res = f"Empty Struct."
         else:
             if as_yaml or as_config: res = __import__("yaml").dump(self.__dict__) if as_yaml else config(self.__dict__, justify=justify, **kwargs)
-            else: res = self._pandas_repr(justify=justify, return_str=False, limit=limit).drop(columns=[] if dtype else ["dtype"])
+            else:
+                import pandas as pd
+                tmp = self._pandas_repr(justify=justify, return_str=False, limit=limit)
+                assert isinstance(tmp, pd.DataFrame)
+                res = tmp.drop(columns=[] if dtype else ["dtype"])
         if not return_str:
             if ("DataFrame" in res.__class__.__name__ and install_n_import("tabulate")): install_n_import("rich").print(res.to_markdown())
             else: print(res)