add to set" 在Java中返回布尔值 - 那Python呢？

18 投票

7 回答

14780 浏览

提问于 2025-04-17 07:14

在Java中，我喜欢使用“添加到集合”操作返回的布尔值来检查这个元素是否已经在集合里了。

if (set.add("x")) {
    print "x was not yet in the set";
}

我想问的是，Python中有没有类似方便的方法？我试了：

 z = set()
 if (z.add(y)):
     print something

但是它什么都没有打印出来。我是不是漏掉了什么？

集合操作编程语言数据结构 java 布尔值方法比较元素检查

7 个回答

显然，set().add(elem) 这个方法总是返回 None，而且它的类型是 NoneType，这是根据以下内容得出的：

$ python
Python 3.10.7 (main, Sep  7 2022, 01:54:01) [GCC 12.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> type(set().add(12345))
<class 'NoneType'>

如果你想避免重复查找，可以使用 len() 来检测元素是否被添加，通过在添加之前和之后检查集合的长度：

a_set = set()
#...
something=12345
pre_len = len(a_set)
a_set.add(something) #always returns None
if pre_len != len(a_set):
    print(f"The element({something}) was added therefore it was not already in the set.")
else:
    print(f"The element({something}) was not added because it was already in the set.")

我找到了一些关于 len() 的信息：

len() 是怎么工作的？
len() 的工作时间是 O(1)，因为集合是一个对象，并且有一个成员来存储它的大小。以下是来自 Python 文档的 len() 描述。

返回一个对象的长度（项目数量）。参数可以是一个序列（比如字符串、字节、元组、列表或范围）或一个集合（比如字典、集合或冻结集合）。

来源: https://www.geeksforgeeks.org/find-the-length-of-a-set-in-python/

我原本期待长度会以某种方式被缓存，上面的内容似乎证实了这一点。

此外（感谢在 liberachat IRC 的 #python 频道提供的信息和确认），我自己也通过查看源代码确认了这一点，所以我现在可以肯定 len() 返回的是缓存的结果。

还有这个页面在 Python 维基上也证实了这一点：获取长度（对于 list，但影响所有集合）的操作时间复杂度是 O(1)（即使在最坏情况下），而 x in s 的操作时间复杂度在 O(1) 和 O(n)（最坏情况）之间，其中 'n' 是当前容器中的元素数量。

下面我尝试查看时间差，但我不知道自己是否做对了，可能我遗漏了什么，但 len() 的变体似乎稍微慢一些（它比 x in set 变体多花了 9.4% 或 51.7% 的时间）。无论如何，这里是：

what =========== time it took in seconds (underscores for readability/matching)
-----
init_set() highest time==========================5.374_412_298_202_514_648_438_
init_set() average time==========================5.374_412_298_202_514_648_438_
init_set() lowest time===========================5.374_412_298_202_514_648_438_
re_set() highest time============================1.283_961_534_500_122_070_312_
re_set() average time============================1.233_813_919_126_987_457_275_
re_set() lowest time=============================1.027_379_512_786_865_234_375_
use_len_and_always_add() highest time============0.000_089_168_548_583_984_375_
use_len_and_always_add() average time============0.000_002_022_699_288_719_765_*
use_len_and_always_add() lowest time=============0.000_001_668_930_053_710_938_
double_lookup_never_add() highest time===========0.000_107_288_360_595_703_125_
double_lookup_never_add() average time===========0.000_001_333_327_164_114_879_*
double_lookup_never_add() lowest time============0.000_000_953_674_316_406_250_
double_lookup_always_add() highest time==========0.000_087_261_199_951_171_875_
double_lookup_always_add() average time==========0.000_001_681_037_948_399_603_*
double_lookup_always_add() lowest time===========0.000_001_192_092_895_507_812_
double_lookup_never_add2() highest time==========0.000_120_401_382_446_289_062_
double_lookup_never_add2() average time==========0.000_001_423_196_238_256_303_*
double_lookup_never_add2() lowest time===========0.000_001_192_092_895_507_812_
using_len_many() highest time===================10.652_944_326_400_756_835_938_
using_len_many() average time===================10.642_948_746_681_213_378_906_
using_len_many() lowest time====================10.632_953_166_961_669_921_875_
double_lookup_always_add_many() highest time====10.278_126_478_195_190_429_688_
double_lookup_always_add_many() average time====10.234_028_577_804_565_429_688_
double_lookup_always_add_many() lowest time=====10.189_930_677_413_940_429_688_
double_lookup_always_add_many2() highest time===10.584_211_587_905_883_789_062_
double_lookup_always_add_many2() average time===10.565_821_886_062_622_070_312_
double_lookup_always_add_many2() lowest time====10.547_432_184_219_360_351_562_
double_lookup_many() highest time================9.843_203_306_198_120_117_188_
double_lookup_many() average time================9.723_606_467_247_009_277_344_
double_lookup_many() lowest time=================9.604_009_628_295_898_437_500_

以上是手动排序的（CPU 设置为 800Mhz，通过 sudo cpupower frequency-set --related --governor powersave --min 800MHz --max 800MHz），但输出来自以下代码：

#!/usr/bin/python3



import time as t
WORST="() highest time"
AVG="() average time"
BEST="() lowest time"

stats: dict[str,float]=dict({})

def TimeTakenDecorator(func):
    #src: https://stackoverflow.com/a/70954147/19999437
    def wraper(*args,**kwargs):
        global stats
        tmp_avg=stats.get(func.__name__+AVG)
        #too_fast:bool
        if (None == tmp_avg) or (tmp_avg > 0.5):
            print(f'Calling "{func.__name__}()"') #,end="")
            #too_fast=False
        #else:
            #too_fast=True
        start = t.time()
        func(*args,**kwargs)
        end = t.time()
        diff=end - start
        if None == stats.get(func.__name__+WORST):
            stats[func.__name__+WORST]=diff
        if None == tmp_avg:
            stats[func.__name__+AVG]=diff
            tmp_avg=diff
        if None == stats.get(func.__name__+BEST):
            stats[func.__name__+BEST]=diff
        if diff > stats[func.__name__+WORST]:
            stats[func.__name__+WORST]=diff
        if diff < stats[func.__name__+BEST]:
            stats[func.__name__+BEST]=diff
        stats[func.__name__+AVG]=(tmp_avg + diff) /2
        if diff > 0.5:
            print(f'Elapsed time for function call "{func.__name__}": {diff:.20f}')
            #print(" "+str(diff))
        #if not too_fast:
        #    print()
    return wraper

something=5_234_567
REPEATS=1_000_000

#init_set() highest time==========================5.374_412_298_202_514_648_438_
#init_set() average time==========================5.374_412_298_202_514_648_438_
#init_set() lowest time===========================5.374_412_298_202_514_648_438_
@TimeTakenDecorator
def init_set():
    #print("Initializing the set:")
    global g1_set
    g1_set = set()
    for i in range(10_000_000):
        g1_set.add(i)

#re_set() highest time============================1.283_961_534_500_122_070_312_
#re_set() average time============================1.233_813_919_126_987_457_275_
#re_set() lowest time=============================1.027_379_512_786_865_234_375_
@TimeTakenDecorator
def re_set():
    #print("Resetting the set:")
    global g2_set
    global g1_set
    g2_set=g1_set.copy()

#double_lookup_many() highest time================9.843_203_306_198_120_117_188_
#double_lookup_many() average time================9.723_606_467_247_009_277_344_
#double_lookup_many() lowest time=================9.604_009_628_295_898_437_500_
@TimeTakenDecorator
def double_lookup_many():
    #print("Using double lookup:")
    for i in range(REPEATS):
        double_lookup_never_add()

#double_lookup_always_add_many() highest time====10.278_126_478_195_190_429_688_
#double_lookup_always_add_many() average time====10.234_028_577_804_565_429_688_
#double_lookup_always_add_many() lowest time=====10.189_930_677_413_940_429_688_
@TimeTakenDecorator
def double_lookup_always_add_many():
    #print("Using double lookup:")
    for i in range(REPEATS):
        double_lookup_always_add()

#using_len_many() highest time===================10.652_944_326_400_756_835_938_
#using_len_many() average time===================10.642_948_746_681_213_378_906_
#using_len_many() lowest time====================10.632_953_166_961_669_921_875_
@TimeTakenDecorator
def using_len_many():
    #print("Using len():")
    for i in range(REPEATS):
        use_len_and_always_add()

#double_lookup_never_add() highest time===========0.000_107_288_360_595_703_125_
#double_lookup_never_add() average time===========0.000_001_333_327_164_114_879_
#double_lookup_never_add() lowest time============0.000_000_953_674_316_406_250_
@TimeTakenDecorator
def double_lookup_never_add():
    global g2_set
    if something not in g2_set:
        g2_set.add(something)
        #pass
        #print(f"The element({something}) was added therefore it was not already in the set.")
    #else:
        #g2_set.add(something)
        #pass
        #print(f"The element({something}) was not added because it was already in the set.")

#double_lookup_always_add() highest time==========0.000_087_261_199_951_171_875_
#double_lookup_always_add() average time==========0.000_001_681_037_948_399_603_
#double_lookup_always_add() lowest time===========0.000_001_192_092_895_507_812_
@TimeTakenDecorator
def double_lookup_always_add():
    global g2_set
    if something not in g2_set:
        g2_set.add(something)
    else:
        g2_set.add(something)

#use_len_and_always_add() highest time============0.000_089_168_548_583_984_375_
#use_len_and_always_add() average time============0.000_002_022_699_288_719_765_
#use_len_and_always_add() lowest time=============0.000_001_668_930_053_710_938_
@TimeTakenDecorator
def use_len_and_always_add():
    global g2_set
    pre_len = len(g2_set)
    g2_set.add(something)
    #pass
    if pre_len != len(g2_set):
        pass
        #print(f"The element({something}) was added therefore it was not already in the set.")
    #else:
    #    pass
        #print(f"The element({something}) was not added because it was already in the set.")

#double_lookup_never_add2() highest time==========0.000_120_401_382_446_289_062_
#double_lookup_never_add2() average time==========0.000_001_423_196_238_256_303_
#double_lookup_never_add2() lowest time===========0.000_001_192_092_895_507_812_
@TimeTakenDecorator
def double_lookup_never_add2():
    global g2_set
    if something not in g2_set:
        g2_set.add(something)

#double_lookup_always_add_many2() highest time===10.584_211_587_905_883_789_062_
#double_lookup_always_add_many2() average time===10.565_821_886_062_622_070_312_
#double_lookup_always_add_many2() lowest time====10.547_432_184_219_360_351_562_
@TimeTakenDecorator
def double_lookup_always_add_many2():
    global g2_set
    for i in range(REPEATS):
        g2_set.clear()
        double_lookup_never_add2()


def main():
    init_set()
    re_set()
    using_len_many()
    re_set()
    double_lookup_many()
    re_set()
    double_lookup_always_add_many()
    re_set()
    double_lookup_always_add_many2()
    print("Once more in reverse order:")
    re_set()
    double_lookup_always_add_many2()
    re_set()
    double_lookup_always_add_many()
    re_set()
    double_lookup_many()
    re_set()
    using_len_many()
    import json
    #from json import encoder
    #encoder.FLOAT_REPR = lambda o: f'{o:.20f}' #.format(o) #format(o, '.20f')
    #src: https://stackoverflow.com/a/69056325/19999437
    class RoundingFloat(float):
        __repr__ = staticmethod(lambda x: format(x, '.30f'))

    json.encoder.c_make_encoder = None
    #if hasattr(json.encoder, 'FLOAT_REPR'):
    #    # Python 2
    #    json.encoder.FLOAT_REPR = RoundingFloat.__repr__
    #else:
        # Python 3
    json.encoder.float = RoundingFloat
    print(json.dumps(stats, sort_keys=False, indent=2)) #, parse_float=lambda x: f"{x:.20f}"))
    import re
    for k in stats:
        time_form=re.sub(r'([0-9_]+\.)?([0-9]{3})', '\\1\\2_', f"{stats[k]:_.21f}")
        #print(f"{k:-<45} {stats[k]:.20f}")
        print(f"{k:=<45}={time_form:=>33}")


main();

回答于 2025-04-17 由 Python大师

分享举报

正如之前的回答所提到的，Python 中集合的 add 方法是不会返回任何东西的。顺便说一下，这个问题曾在 Python 的邮件列表上讨论过：http://mail.python.org/pipermail/python-ideas/2009-February/002877.html。

回答于 2025-04-17 由 Python大师

分享举报

在Python中，set.add()这个方法不会返回任何东西。如果你想检查某个元素是否在集合中，可以使用not in这个操作符。

z = set()
if y not in z: # If the object is not in the list yet...
    print something
z.add(y)

如果你真的想知道在你添加这个元素之前，它是否已经在集合里，你可以先把这个结果存储为一个布尔值（真或假）。

z = set()
was_here = y not in z
z.add(y)
if was_here: # If the object was not in the list yet...
    print something

不过，我觉得你可能并不需要这样做。

这是Python的一种习惯：当一个方法更新某个对象时，它通常会返回None。你可以选择忽略这个习惯；实际上，有些方法并不遵循这个规则。不过，这种习惯是比较常见和被认可的：我建议你遵循这个习惯，并记住这一点。

回答于 2025-04-17 由 Python大师

分享举报

add to set" 在Java中返回布尔值 - 那Python呢？

7 个回答

撰写回答