bool-hybrid-array 9.10.21__tar.gz → 9.11.0__tar.gz

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bool-hybrid-array
-Version: 9.10.21
+Version: 9.11.0
 Summary: 一个高效的布尔数组（密集+稀疏混合存储，节省内存）
 Home-page: https://github.com/BKsell/bool-hybrid-array
 Author: 蔡靖杰
@@ -17,6 +17,8 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: Implementation :: CPython
+Classifier: Programming Language :: Python :: Implementation :: PyPy
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
@@ -571,6 +573,12 @@ except OverflowError as e:
 * **9.10.19**：和9.10.18.post4相同，正式版本发布
 * **9.10.20**：优化性能，增加BHA_jit_log日志
 * **9.10.21**：优化Ask_BHA，移除BHA_jit_log日志
+* **9.10.22**：进一步优化Ask_BHA的性能
+* **9.11.0**：新增对7.3.10以上的版本的PyPy解释器的支持
+
+
+
+
 
 
 ## **彩蛋：**
@@ -585,7 +593,11 @@ except OverflowError as e:
 * A：BoolHybridArray是本库中的**核心类**，所有函数都是围绕他进行的，但需要`split_index`,`size`,`is_sparse`；
   BoolHybridArr是一个**函数**，用于把一个可迭代对象转为`BoolHybridArray`类
 * Q：为什么不建议把太大的本类型数组打印？
-* A：虽然本类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+* A：虽然BoolHybridArray类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+- Q：为什么使用在Windows终端使用PyPy时建议先调用`chcp 65001`再启动PyPy？
+* A：因为本库有很多的的中文报错和打印，项目里也可能会有。Windows终端里的PyPy默认是GBK编码，有中文时会乱码。chcp 65001可以切换到UTF-8编码，防止乱码。
+
+
 
 ## 源代码和原理
 

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/README.md

@@ -531,6 +531,12 @@ except OverflowError as e:
 * **9.10.19**：和9.10.18.post4相同，正式版本发布
 * **9.10.20**：优化性能，增加BHA_jit_log日志
 * **9.10.21**：优化Ask_BHA，移除BHA_jit_log日志
+* **9.10.22**：进一步优化Ask_BHA的性能
+* **9.11.0**：新增对7.3.10以上的版本的PyPy解释器的支持
+
+
+
+
 
 
 ## **彩蛋：**
@@ -545,7 +551,11 @@ except OverflowError as e:
 * A：BoolHybridArray是本库中的**核心类**，所有函数都是围绕他进行的，但需要`split_index`,`size`,`is_sparse`；
   BoolHybridArr是一个**函数**，用于把一个可迭代对象转为`BoolHybridArray`类
 * Q：为什么不建议把太大的本类型数组打印？
-* A：虽然本类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+* A：虽然BoolHybridArray类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+- Q：为什么使用在Windows终端使用PyPy时建议先调用`chcp 65001`再启动PyPy？
+* A：因为本库有很多的的中文报错和打印，项目里也可能会有。Windows终端里的PyPy默认是GBK编码，有中文时会乱码。chcp 65001可以切换到UTF-8编码，防止乱码。
+
+
 
 ## 源代码和原理
 

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/bool_hybrid_array/__init__.py

@@ -1,9 +1,11 @@
+# -*- coding: utf-8 -*-
 import sys
 from types import ModuleType,FunctionType
 from . import core
 from .core import __builtins__,builtins
-from . import int_array
-__version__ = "9.10.21"
+try:from . import int_array
+except:pass
+__version__ = "9.11.0"
 public_objects = []
 for name in dir(core):
     if not name.startswith("_"):
@@ -16,11 +18,11 @@ globals().update({
     for name in public_objects
 })
 try:
-    __dict__ = ProtectedBuiltinsDict(globals())
-    sys.modules[__name__+'.int_array'] = ProtectedBuiltinsDict(int_array.__dict__)
-    sys.modules[__name__+'.core'] = ProtectedBuiltinsDict(core.__dict__)
     sys.modules[__name__] = ProtectedBuiltinsDict(globals())
-    sys.modules[__name__].name = 'bool_hybrid_array'
+    sys.modules[__name__].name = __name__
+    sys.modules[__name__+'.core'] = ProtectedBuiltinsDict(core.__dict__,name = f'{__name__}.core')
+    __dict__ = ProtectedBuiltinsDict(globals())
+    sys.modules[__name__+'.int_array'] = ProtectedBuiltinsDict(int_array.__dict__,name = __name__+'.int_array')
     core.__dict__ = ProtectedBuiltinsDict(core.__dict__)
 except:
     pass

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/bool_hybrid_array/__main__.py

@@ -1,4 +1,8 @@
+# -*- coding: utf-8 -*-
 from .__init__ import *
+import io
+
+sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding='utf-8', errors='ignore')
 
 # 创建实例
 

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/bool_hybrid_array/core.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 import builtins
+from types import MappingProxyType
 import array,bisect,numpy as np
 from collections.abc import MutableSequence,Iterable,Generator,Iterator,Sequence
 import itertools,copy,sys,math,weakref,random,mmap,os
@@ -8,6 +9,14 @@ import operator,ctypes,gc,abc,types
 from functools import lru_cache
 from typing import Union,_GenericAlias
 hybrid_array_cache = []
+try:
+    msvcrt = ctypes.CDLL('msvcrt.dll')
+    memcpy = msvcrt.memcpy
+except:
+    libc = ctypes.CDLL('libc.so.6')
+    memcpy = libc.memcpy
+memcpy.argtypes = (ctypes.c_void_p, ctypes.c_void_p, ctypes.c_size_t)
+memcpy.restype = ctypes.c_void_p
 if 'UnionType' in types.__dict__:
     class Union:
         def __getitem__(self,*args):
@@ -29,12 +38,23 @@ class ResurrectMeta(abc.ABCMeta,metaclass=abc.ABCMeta):
         super_cls.__setattr__('bases', bases)
         super_cls.__setattr__('namespace', namespace)
         super_cls.__setattr__('original_dict', dict(obj.__dict__))
-        del obj.original_dict["__abstractmethods__"]
-        del obj.original_dict["_abc_impl"]
+        try:del obj.original_dict["__abstractmethods__"]
+        except:pass
+        try:del obj.original_dict["_abc_impl"]
+        except:pass
+        try:del obj.original_dict['_abc_registry']
+        except:pass
+        try:del obj.original_dict['_abc_cache']
+        except:pass
+        try:del obj.original_dict['_abc_negative_cache']
+        except:pass
+        try:del obj.original_dict['_abc_negative_cache_version']
+        except:pass
+        super_cls.__setattr__('original_dict', MappingProxyType(obj.original_dict))
         return obj
     @lru_cache
     def __str__(cls):
-        return super().__repr__()[8:][:-2]
+        return f'{cls.__module__}.{cls.name}'
     @lru_cache
     def __repr__(cls,detailed = False):
         if detailed:
@@ -79,6 +99,7 @@ class ResurrectMeta(abc.ABCMeta,metaclass=abc.ABCMeta):
         original_dict["__or__"] = __or__
     except:
         pass
+    original_dict = MappingProxyType(original_dict)
 ResurrectMeta.__class__ = ResurrectMeta
 class BHA_Function(metaclass=ResurrectMeta):
     def __init__(self,v):
@@ -703,11 +724,11 @@ class BHA_Iterator(Iterator,metaclass=ResurrectMeta):
         return arr.copy() if copy else arr.view()
     __rand__,__ror__,__rxor__ = __and__,__or__,__xor__
 class ProtectedBuiltinsDict(dict,metaclass=ResurrectMeta):
-    def __init__(self, *args, protected_names = ["T", "F", "BHA_Bool", "BHA_List", "BoolHybridArray", "BoolHybridArr",
+    def __init__(self, *args, protected_names = ("T", "F", "BHA_Bool", "BHA_List", "BoolHybridArray", "BoolHybridArr",
                                 "TruesArray", "FalsesArray", "ProtectedBuiltinsDict", "builtins",
                                 "__builtins__", "__dict__","ResurrectMeta","math",
                                 "np","protected_names","BHA_Function",
-                                "__class__","Ask_BHA","Create_BHA","Ask_arr","numba_opt"],
+                                "__class__","Ask_BHA","Create_BHA","Ask_arr","numba_opt"),
                  name = 'builtins', **kwargs):
         super().__init__(*args, **kwargs)
         if name == 'builtins':
@@ -715,20 +736,23 @@ class ProtectedBuiltinsDict(dict,metaclass=ResurrectMeta):
             super().__setattr__('builtins',self)
             super().__setattr__('__builtins__',self)
         self.name = name
-        self.protected_names = protected_names
+        super().__setattr__("protected_names",protected_names)
     def __setitem__(self, name, value):
-        if name in ["T", "F"]:
-            current_T = self.get("T")
-            current_F = self.get("F")
-            if isinstance(current_T, BHA_bool) and isinstance(current_F, BHA_bool):
-                is_swap = (name == "T" and isinstance(value, BHA_bool) and value.value == current_F.value)or(name == "F" and isinstance(value, BHA_bool) and value.value == current_T.value)
-                if is_swap:
-                    print(f"""警告：禁止交换内置常量 __{self.name}__["{name}"] 和 __builtins__["{'F' if name == 'T' else 'T'}"]！""")
-                    raise AttributeError(f"""禁止交换内置常量 __{self.name}__["{name}"] 和 __{self.name}__["{'F' if name == 'T' else 'T'}"]""")
-        if name in self.protected_names and name not in ["T", "F"]:
-            print(f"警告：禁止修改内置常量 __{self.name}__['{name}']！")
-            raise AttributeError(f"禁止修改内置常量 __{self.name}__['{name}']")
-        super().__setitem__(name, value)
+        if not hasattr(self,"protected_names"):super().__setitem__(name, value)
+        try:
+            if name in ["T", "F"]:
+                current_T = self.get("T")
+                current_F = self.get("F")
+                if isinstance(current_T, BHA_bool) and isinstance(current_F, BHA_bool):
+                    is_swap = (name == "T" and isinstance(value, BHA_bool) and value.value == current_F.value)or(name == "F" and isinstance(value, BHA_bool) and value.value == current_T.value)
+                    if is_swap:
+                        print(f"""警告：禁止交换内置常量 __{self.name}__["{name}"] 和 __builtins__["{'F' if name == 'T' else 'T'}"]！""")
+                        raise AttributeError(f"""禁止交换内置常量 __{self.name}__["{name}"] 和 __{self.name}__["{'F' if name == 'T' else 'T'}"]""")
+            if name in self.protected_names and name not in ["T", "F"]:
+                print(f"警告：禁止修改内置常量 __{self.name}__['{name}']！")
+                raise AttributeError(f"禁止修改内置常量 __{self.name}__['{name}']")
+        except:pass
+        finally:super().__setitem__(name, value)
     def __delitem__(self, name):
         if name in self.protected_names:
             print(f"警告：禁止删除内置常量 __builtins__['{name}']！")
@@ -741,10 +765,12 @@ class ProtectedBuiltinsDict(dict,metaclass=ResurrectMeta):
         else:
             del self[name]
     def __getattr__(self, name):
-        if name in self:
-            return self[name]
-        else:
-            raise AttributeError(f"module 'builtins' has no attribute '{name}'")
+        try:
+            return super().__getattribute__(name)
+        except AttributeError:
+            if name in self:
+                return self[name]
+            raise AttributeError(f"module 'builtins' has no attribute '{name}'") from None
     def __setattr__(self,name,value):
         try:protected = self.protected_names
         except Exception:protected = self
@@ -786,18 +812,14 @@ def Ask_BHA(path):
             temp = mm.read().decode('utf-8').strip()
         temp = temp.split()
         temp2 = lambda x: BoolHybridArr(
-        reduce(
-        lambda acc, k: acc.append(0 if k < lead_zero else
-        (n >> ((total_len - 1) - k)) & 1
-        ) or acc,
-        range(total_len),
-        array.array('B',[])),
-        hash_=F
-        )if(
-            n := int(x, base=16),
-            lead_zero := len(x) - len(x.lstrip('0')),
-            total_len := lead_zero + (n.bit_length() if n != 0 else 1)
-        )else None
+        (
+        bit_stream := bytes(0 if k < lead_zero else (n >> ((total_len - 1) - k)) & 1 for k in range(total_len)),
+        arr := array.array('B', FalsesArray(total_len)),
+        memcpy(arr.buffer_info()[0], bit_stream, total_len),arr)[-1]
+        if(n := int(x, base=16),
+        lead_zero := len(x) - len(x.lstrip('0')),
+        total_len := lead_zero + (n.bit_length() if n else 1))
+        else array.array('B'))
         temp = BHA_List(map(temp2,temp))
         if len(temp) == 1:
             return temp[0]

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/bool_hybrid_array/int_array/__init__.py

@@ -1,7 +1,7 @@
+# -*- coding: utf-8 -*-
 from __future__ import annotations
 from ..core import *
 import builtins
-
 class IntBitTag(BHA_bool, metaclass=ResurrectMeta):
     def __str__(self):
         return "'-1'" if (hasattr(self, 'is_sign_bit') and self.is_sign_bit and self) else "'1'" if self else "'0'"
@@ -80,7 +80,7 @@ class IntHybridArray(BoolHybridArray):
                 block_end = block_start + self.bit_length
                 if block_end > self.size:
                     raise IndexError("索引超出范围")
-                bit_chunk = [super().__getitem__(j) for j in range(block_start, block_end)]
+                bit_chunk = [super(self.__class__, self).__getitem__(j) for j in range(block_start, block_end)]
                 num = self.to_int(bit_chunk)
                 result.append(num)
             return IntHybridArray(result, self.bit_length)
@@ -91,7 +91,7 @@ class IntHybridArray(BoolHybridArray):
         block_end = block_start + self.bit_length
         if block_end > self.size:
             raise IndexError("索引超出范围")
-        bit_chunk = [super().__getitem__(j) for j in range(block_start, block_end)]
+        bit_chunk = [super(self.__class__, self).__getitem__(j) for j in range(block_start, block_end)]
         return self.to_int(bit_chunk)
 
     def __setitem__(self, key, value):
@@ -145,14 +145,10 @@ class IntHybridArray(BoolHybridArray):
                     carry = 0 if num_bits[j] else 1
         bool_data = [sign_bit] + num_bits
         for bit_idx in range(self.bit_length):
-            super().__setitem__(block_start + bit_idx, bool_data[bit_idx])
+            super(self.__class__, self).__setitem__(block_start + bit_idx, bool_data[bit_idx])
 
     def __iter__(self):
-        for i in range(0, self.total_bits, self.bit_length):
-            if i + self.bit_length > self.size:
-                break
-            bit_chunk = [super().__getitem__(j) for j in range(i, i + self.bit_length)]
-            yield self.to_int(bit_chunk)
+        return map(self.__getitem__,range(len(self)))
 
     def __str__(self):
         return f"IntHybridArray([{', '.join(map(str, self))}])"
@@ -173,7 +169,7 @@ class IntHybridArray(BoolHybridArray):
     def index(self, value):
         value = int(value)
         x = f"{value} 不在 IntHybridArray 中"
-        for idx in range(len(self)+2>>2):
+        for idx in range(len(self)+1>>1):
             if self[idx] == value:
                 return idx
             elif self[-idx] == value:
@@ -184,7 +180,7 @@ class IntHybridArray(BoolHybridArray):
     def rindex(self, value):
         value = int(value)
         x = f"{value} 不在 IntHybridArray 中"
-        for idx in range(len(self)+2>>2):
+        for idx in range(len(self)+1>>1):
             if self[-idx] == value:
                 return -idx
             elif self[idx] == value:
@@ -200,4 +196,5 @@ class IntHybridArray(BoolHybridArray):
             len_ = len(iterable)
         self.total_bits += len_*self.bit_length
         for i,j in zip(range(len_),iterable):
-            self[-i-1] = j
+            self[-i-1] = j
+__all__ = list(globals())

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/bool_hybrid_array.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bool-hybrid-array
-Version: 9.10.21
+Version: 9.11.0
 Summary: 一个高效的布尔数组（密集+稀疏混合存储，节省内存）
 Home-page: https://github.com/BKsell/bool-hybrid-array
 Author: 蔡靖杰
@@ -17,6 +17,8 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: Implementation :: CPython
+Classifier: Programming Language :: Python :: Implementation :: PyPy
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.8
@@ -571,6 +573,12 @@ except OverflowError as e:
 * **9.10.19**：和9.10.18.post4相同，正式版本发布
 * **9.10.20**：优化性能，增加BHA_jit_log日志
 * **9.10.21**：优化Ask_BHA，移除BHA_jit_log日志
+* **9.10.22**：进一步优化Ask_BHA的性能
+* **9.11.0**：新增对7.3.10以上的版本的PyPy解释器的支持
+
+
+
+
 
 
 ## **彩蛋：**
@@ -585,7 +593,11 @@ except OverflowError as e:
 * A：BoolHybridArray是本库中的**核心类**，所有函数都是围绕他进行的，但需要`split_index`,`size`,`is_sparse`；
   BoolHybridArr是一个**函数**，用于把一个可迭代对象转为`BoolHybridArray`类
 * Q：为什么不建议把太大的本类型数组打印？
-* A：虽然本类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+* A：虽然BoolHybridArray类型数组省内存，但字符串不省内存，一个`True`要4字节，一个`False`要5字节，连逗号都要占1字节（`numpy.ndarray`：我都比字符串省内存）
+- Q：为什么使用在Windows终端使用PyPy时建议先调用`chcp 65001`再启动PyPy？
+* A：因为本库有很多的的中文报错和打印，项目里也可能会有。Windows终端里的PyPy默认是GBK编码，有中文时会乱码。chcp 65001可以切换到UTF-8编码，防止乱码。
+
+
 
 ## 源代码和原理
 

{bool_hybrid_array-9.10.21 → bool_hybrid_array-9.11.0}/setup.py

@@ -8,7 +8,7 @@ def get_long_description():
     return "一个高效的布尔数组（密集+稀疏混合存储，节省内存）"
 setup(
     name="bool-hybrid-array",
-    version="9.10.21",
+    version="9.11.0",
     author="蔡靖杰",
     extras_require={"int_array":[]},
     author_email="1289270215@qq.com",
@@ -27,11 +27,13 @@ setup(
         'Programming Language :: Python :: 3.12',
         'Programming Language :: Python :: 3.13',
         'Programming Language :: Python :: 3.14',
+        'Programming Language :: Python :: Implementation :: CPython',
+        'Programming Language :: Python :: Implementation :: PyPy', 
         "License :: OSI Approved :: MIT License",
         "Operating System :: OS Independent",
     ],
     keywords="boolean array, compact storage",
-    package_data={"": ["README.md", "LICENSE",'temp.py', 'temp.cmd','BHA_Opener.zip'],"bool_hybrid_array":["*"]},
+    package_data={"": ["README.md", "LICENSE",'temp.py', 'temp.cmd','BHA_Opener.zip'],"bool_hybrid_array":["*"],"bool_hybrid_array/__pycache__":['*.pyc']},
     include_package_data=True,
     url="https://github.com/BKsell/bool-hybrid-array",
     project_urls={