如何使用JSON模块进行漂亮打印时实现自定义缩进？

Question

所以我使用的是Python 2.7,使用该json模块对以下数据结构进行编码:

'layer1': {
    'layer2': {
        'layer3_1': [ long_list_of_stuff ],
        'layer3_2': 'string'
    }
}

我的问题是我使用漂亮的打印方式打印出来,如下所示:

json.dumps(data_structure, indent=2)

哪个好,除了我要缩进所有内容,除了内容"layer3_1"- 这是一个列出坐标的大量字典,因此,在每个上面设置一个值使得漂亮的打印创建一个包含数千行的文件,示例如下:

{
  "layer1": {
    "layer2": {
      "layer3_1": [
        {
          "x": 1,
          "y": 7
        },
        {
          "x": 0,
          "y": 4
        },
        {
          "x": 5,
          "y": 3
        },
        {
          "x": 6,
          "y": 9
        }
      ],
      "layer3_2": "string"
    }
  }
}

我真正想要的是类似于以下内容:

{
  "layer1": {
    "layer2": {
      "layer3_1": [{"x":1,"y":7},{"x":0,"y":4},{"x":5,"y":3},{"x":6,"y":9}],
      "layer3_2": "string"
    }
  }
}

我听说可以扩展json模块:是否可以将其设置为仅在"layer3_1"对象内部时关闭缩进？如果是这样,有人请告诉我怎么样？

Answer 1

更新

以下是我原来答案的一个版本,已经多次修订.与原版不同,我发布的只是为了展示如何在JFSebastian的工作答案中获得第一个想法,并且像他一样,返回了对象的非缩进字符串表示.最新更新版本返回单独格式化的Python对象JSON.

根据dictOP的注释之一,每个坐标的键将按排序顺序显示,但sort_keys=True前提是在json.dumps()驱动该过程的初始调用中指定了关键字参数,并且它不再将对象的类型更改为字符串.换句话说,现在保持"包裹"对象的实际类型.

我认为不理解我的帖子的原始意图导致一些人低估它 - 所以,主要是因为这个原因,我已经"修复"并多次改进我的答案.当前版本是我原来的答案与@Erik Allik在他的答案中使用的一些想法的混合,以及来自其他用户的有用反馈,显示在此答案的下方评论中.

以下代码在Python 2.7.14和3.6.5中似乎都没有改变.

from _ctypes import PyObj_FromPtr
import json
import re

class NoIndent(object):
    """ Value wrapper. """
    def __init__(self, value):
        self.value = value


class MyEncoder(json.JSONEncoder):
    FORMAT_SPEC = '@@{}@@'
    regex = re.compile(FORMAT_SPEC.format(r'(\d+)'))

    def __init__(self, **kwargs):
        # Save copy of any keyword argument values needed for use here.
        self.__sort_keys = kwargs.get('sort_keys', None)
        super(MyEncoder, self).__init__(**kwargs)

    def default(self, obj):
        return (self.FORMAT_SPEC.format(id(obj)) if isinstance(obj, NoIndent)
                else super(MyEncoder, self).default(obj))

    def encode(self, obj):
        format_spec = self.FORMAT_SPEC  # Local var to expedite access.
        json_repr = super(MyEncoder, self).encode(obj)  # Default JSON.

        # Replace any marked-up object ids in the JSON repr with the
        # value returned from the json.dumps() of the corresponding
        # wrapped Python object.
        for match in self.regex.finditer(json_repr):
            # see https://stackoverflow.com/a/15012814/355230
            id = int(match.group(1))
            no_indent = PyObj_FromPtr(id)
            json_obj_repr = json.dumps(no_indent.value, sort_keys=self.__sort_keys)

            # Replace the matched id string with json formatted representation
            # of the corresponding Python object.
            json_repr = json_repr.replace(
                            '"{}"'.format(format_spec.format(id)), json_obj_repr)

        return json_repr


if __name__ == '__main__':
    from string import ascii_lowercase as letters

    data_structure = {
        'layer1': {
            'layer2': {
                'layer3_1': NoIndent([{"x":1,"y":7}, {"x":0,"y":4}, {"x":5,"y":3},
                                      {"x":6,"y":9},
                                      {k: v for v, k in enumerate(letters)}]),
                'layer3_2': 'string',
                'layer3_3': NoIndent([{"x":2,"y":8,"z":3}, {"x":1,"y":5,"z":4},
                                      {"x":6,"y":9,"z":8}]),
                'layer3_4': NoIndent(list(range(20))),
            }
        }
    }

    print(json.dumps(data_structure, cls=MyEncoder, sort_keys=True, indent=2))

输出:

{
  "layer1": {
    "layer2": {
      "layer3_1": [{"x": 1, "y": 7}, {"x": 0, "y": 4}, {"x": 5, "y": 3}, {"x": 6, "y": 9}, {"a": 0, "b": 1, "c": 2, "d": 3, "e": 4, "f": 5, "g": 6, "h": 7, "i": 8, "j": 9, "k": 10, "l": 11, "m": 12, "n": 13, "o": 14, "p": 15, "q": 16, "r": 17, "s": 18, "t": 19, "u": 20, "v": 21, "w": 22, "x": 23, "y": 24, "z": 25}],
      "layer3_2": "string",
      "layer3_3": [{"x": 2, "y": 8, "z": 3}, {"x": 1, "y": 5, "z": 4}, {"x": 6, "y": 9, "z": 8}],
      "layer3_4": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
    }
  }
}

Answer 2

一个bodge,但是一旦你有来自dumps()的字符串,你可以在它上面执行正则表达式替换,如果你确定它的内容的格式.有点像:

s = json.dumps(data_structure, indent=2)
s = re.sub('\s*{\s*"(.)": (\d+),\s*"(.)": (\d+)\s*}(,?)\s*', r'{"\1":\2,"\3":\4}\5', s)

Answer 3

以下解决方案似乎在Python 2.7.x上正常工作.它使用从Python 2.7中的Custom JSON编码器获取的解决方法来插入纯JavaScript代码,以避免使用基于UUID的替换方案在输出中最终作为JSON字符串的自定义编码对象.

class NoIndent(object):
    def __init__(self, value):
        self.value = value


class NoIndentEncoder(json.JSONEncoder):
    def __init__(self, *args, **kwargs):
        super(NoIndentEncoder, self).__init__(*args, **kwargs)
        self.kwargs = dict(kwargs)
        del self.kwargs['indent']
        self._replacement_map = {}

    def default(self, o):
        if isinstance(o, NoIndent):
            key = uuid.uuid4().hex
            self._replacement_map[key] = json.dumps(o.value, **self.kwargs)
            return "@@%s@@" % (key,)
        else:
            return super(NoIndentEncoder, self).default(o)

    def encode(self, o):
        result = super(NoIndentEncoder, self).encode(o)
        for k, v in self._replacement_map.iteritems():
            result = result.replace('"@@%s@@"' % (k,), v)
        return result

然后这个

obj = {
  "layer1": {
    "layer2": {
      "layer3_2": "string", 
      "layer3_1": NoIndent([{"y": 7, "x": 1}, {"y": 4, "x": 0}, {"y": 3, "x": 5}, {"y": 9, "x": 6}])
    }
  }
}
print json.dumps(obj, indent=2, cls=NoIndentEncoder)

产生以下输出:

{
  "layer1": {
    "layer2": {
      "layer3_2": "string", 
      "layer3_1": [{"y": 7, "x": 1}, {"y": 4, "x": 0}, {"y": 3, "x": 5}, {"y": 9, "x": 6}]
    }
  }
}

它还正确地传递所有选项(例外indent),例如sort_keys=True向下传递到嵌套json.dumps调用.

obj = {
    "layer1": {
        "layer2": {
            "layer3_1": NoIndent([{"y": 7, "x": 1, }, {"y": 4, "x": 0}, {"y": 3, "x": 5, }, {"y": 9, "x": 6}]),
            "layer3_2": "string",
        }
    }
}    
print json.dumps(obj, indent=2, sort_keys=True, cls=NoIndentEncoder)

正确输出:

{
  "layer1": {
    "layer2": {
      "layer3_1": [{"x": 1, "y": 7}, {"x": 0, "y": 4}, {"x": 5, "y": 3}, {"x": 6, "y": 9}], 
      "layer3_2": "string"
    }
  }
}

它也可以与例如collections.OrderedDict:

obj = {
    "layer1": {
        "layer2": {
            "layer3_2": "string",
            "layer3_3": NoIndent(OrderedDict([("b", 1), ("a", 2)]))
        }
    }
}
print json.dumps(obj, indent=2, cls=NoIndentEncoder)

产出:

{
  "layer1": {
    "layer2": {
      "layer3_3": {"b": 1, "a": 2}, 
      "layer3_2": "string"
    }
  }
}

Answer 4

这产生了OP的预期结果：

import json

class MyJSONEncoder(json.JSONEncoder):

  def iterencode(self, o, _one_shot=False):
    list_lvl = 0
    for s in super(MyJSONEncoder, self).iterencode(o, _one_shot=_one_shot):
      if s.startswith('['):
        list_lvl += 1
        s = s.replace('\n', '').rstrip()
      elif 0 < list_lvl:
        s = s.replace('\n', '').rstrip()
        if s and s[-1] == ',':
          s = s[:-1] + self.item_separator
        elif s and s[-1] == ':':
          s = s[:-1] + self.key_separator
      if s.endswith(']'):
        list_lvl -= 1
      yield s

o = {
  "layer1":{
    "layer2":{
      "layer3_1":[{"y":7,"x":1},{"y":4,"x":0},{"y":3,"x":5},{"y":9,"x":6}],
      "layer3_2":"string",
      "layer3_3":["aaa\nbbb","ccc\nddd",{"aaa\nbbb":"ccc\nddd"}],
      "layer3_4":"aaa\nbbb",
    }
  }
}

jsonstr = json.dumps(o, indent=2, separators=(',', ':'), sort_keys=True,
    cls=MyJSONEncoder)
print(jsonstr)
o2 = json.loads(jsonstr)
print('identical objects: {}'.format((o == o2)))