@mouad的答案很好，如果你假设两个字典都只包含简单的值。然而，如果你有包含字典的字典，你会得到一个异常，因为字典是不可哈希的。

在我的脑海中，这样做可能有用:

def compare_dictionaries(dict1, dict2):
     if dict1 is None or dict2 is None:
        print('Nones')
        return False

     if (not isinstance(dict1, dict)) or (not isinstance(dict2, dict)):
        print('Not dict')
        return False

     shared_keys = set(dict1.keys()) & set(dict2.keys())

     if not ( len(shared_keys) == len(dict1.keys()) and len(shared_keys) == len(dict2.keys())):
        print('Not all keys are shared')
        return False


     dicts_are_equal = True
     for key in dict1.keys():
         if isinstance(dict1[key], dict) or isinstance(dict2[key], dict):
             dicts_are_equal = dicts_are_equal and compare_dictionaries(dict1[key], dict2[key])
         else:
             dicts_are_equal = dicts_are_equal and all(atleast_1d(dict1[key] == dict2[key]))

     return dicts_are_equal

由于似乎没有人提到deepdiff，为了完整起见，我将在这里添加它。我发现它非常方便获得(嵌套)对象的差异:

安装

pip install deepdiff

示例代码

import deepdiff
import json

dict_1 = {
    "a": 1,
    "nested": {
        "b": 1,
    }
}

dict_2 = {
    "a": 2,
    "nested": {
        "b": 2,
    }
}

diff = deepdiff.DeepDiff(dict_1, dict_2)
print(json.dumps(diff, indent=4))

输出

{
    "values_changed": {
        "root['a']": {
            "new_value": 2,
            "old_value": 1
        },
        "root['nested']['b']": {
            "new_value": 2,
            "old_value": 1
        }
    }
}

注意关于检查结果的漂亮打印:如果两个字典具有相同的属性键(可能与示例中的属性值不同)，上面的代码就可以工作。但是，如果存在一个“extra”属性是字典之一，json.dumps()将失败

TypeError: Object of type PrettyOrderedSet is not JSON serializable

解决方案:使用diff.to_json()和json.loads() / json.dumps()来漂亮地打印:

import deepdiff
import json

dict_1 = {
    "a": 1,
    "nested": {
        "b": 1,
    },
    "extra": 3
}

dict_2 = {
    "a": 2,
    "nested": {
        "b": 2,
    }
}

diff = deepdiff.DeepDiff(dict_1, dict_2)
print(json.dumps(json.loads(diff.to_json()), indent=4))  

输出:

{
    "dictionary_item_removed": [
        "root['extra']"
    ],
    "values_changed": {
        "root['a']": {
            "new_value": 2,
            "old_value": 1
        },
        "root['nested']['b']": {
            "new_value": 2,
            "old_value": 1
        }
    }
}

替代方案:使用pprint，结果是不同的格式:

import pprint

# same code as above

pprint.pprint(diff, indent=4)

输出:

{   'dictionary_item_removed': [root['extra']],
    'values_changed': {   "root['a']": {   'new_value': 2,
                                           'old_value': 1},
                          "root['nested']['b']": {   'new_value': 2,
                                                     'old_value': 1}}}

功能很好IMO，清晰直观。但为了给你(另一个)答案，我是这么说的:

def compare_dict(dict1, dict2):
    for x1 in dict1.keys():
        z = dict1.get(x1) == dict2.get(x1)
        if not z:
            print('key', x1)
            print('value A', dict1.get(x1), '\nvalue B', dict2.get(x1))
            print('-----\n')

可能对你或任何人都有用。

编辑:

我已经创建了一个递归版本的上面..在其他答案中没有看到吗

def compare_dict(a, b):
    # Compared two dictionaries..
    # Posts things that are not equal..
    res_compare = []
    for k in set(list(a.keys()) + list(b.keys())):
        if isinstance(a[k], dict):
            z0 = compare_dict(a[k], b[k])
        else:
            z0 = a[k] == b[k]

        z0_bool = np.all(z0)
        res_compare.append(z0_bool)
        if not z0_bool:
            print(k, a[k], b[k])
    return np.all(res_compare)

我正在使用这个解决方案，在Python 3中完美地为我工作

import logging
log = logging.getLogger(__name__)

...

    def deep_compare(self,left, right, level=0):
        if type(left) != type(right):
            log.info("Exit 1 - Different types")
            return False

        elif type(left) is dict:
            # Dict comparison
            for key in left:
                if key not in right:
                    log.info("Exit 2 - missing {} in right".format(key))
                    return False
                else:
                    if not deep_compare(left[str(key)], right[str(key)], level +1 ):
                        log.info("Exit 3 - different children")
                        return False
            return True
        elif type(left) is list:
            # List comparison
            for key in left:
                if key not in right:
                    log.info("Exit 4 - missing {} in right".format(key))
                    return False
                else:
                    if not deep_compare(left[left.index(key)], right[right.index(key)], level +1 ):
                        log.info("Exit 5 - different children")
                        return False
            return True
        else:
            # Other comparison
            return left == right

        return False

它比较dict、list和其他单独实现“==”操作符的类型。 如果你需要比较其他不同的东西，你需要在“If树”中添加一个新的分支。

希望这能有所帮助。

对两个字典进行深度比较的最简单的方法(也是最健壮的方法之一)是将它们序列化为JSON格式，对键进行排序，并比较字符串结果:

import json
if json.dumps(x, sort_keys=True) == json.dumps(y, sort_keys=True):
   ... Do something ...

在PyUnit中有一个比较字典的方法。我使用以下两个字典对它进行了测试，它完全符合您的要求。

d1 = {1: "value1",
      2: [{"subKey1":"subValue1",
           "subKey2":"subValue2"}]}
d2 = {1: "value1",
      2: [{"subKey2":"subValue2",
           "subKey1": "subValue1"}]
      }


def assertDictEqual(self, d1, d2, msg=None):
        self.assertIsInstance(d1, dict, 'First argument is not a dictionary')
        self.assertIsInstance(d2, dict, 'Second argument is not a dictionary')

        if d1 != d2:
            standardMsg = '%s != %s' % (safe_repr(d1, True), safe_repr(d2, True))
            diff = ('\n' + '\n'.join(difflib.ndiff(
                           pprint.pformat(d1).splitlines(),
                           pprint.pformat(d2).splitlines())))
            standardMsg = self._truncateMessage(standardMsg, diff)
            self.fail(self._formatMessage(msg, standardMsg))

我不建议在生产代码中导入unittest。我的想法是，PyUnit中的源代码可以重新配置，以在生产环境中运行。它使用pprint来“漂亮地打印”字典。调整这段代码以使其“适合生产”似乎很容易。