Question

现在我有了unicode对象

u'[
    [
        [
            ["Hello","w_0","w_0,w_10",1],
            [".","w_1","",0," "],
            ["This","w_2","",0," "],
        ]
        [
            ["\u4f60\u597d","w_10","w_0,w_10",1],
            ["\uff0c","w_11","",0],
            ["\u6211","w_12","",0],
        ],
        "http://www.kekenet.com/menu/201206/187574.shtml",154191
    ]
]'

如何将其转换为列表对象，如

[
    [
        [
            ["Hello","w_0","w_0,w_10",1],
            [".","w_1","",0," "],
            ["This","w_2","",0," "],
        ],
        [
            ["\u4f60\u597d","w_10","w_0,w_10",1],
            ["\uff0c","w_11","",0],
            ["\u6211","w_12","",0],
        ],
        "http://www.kekenet.com/menu/201206/187574.shtml",154191
    ],
]

Answer 1

您可以使用ast.literal_eval安全地评估包含Python文字的字符串。

from ast import literal_eval

a = '["Hello", "World!", 2]'
b = literal_eval(a)
# ["Hello", "World!", 2]

请注意，字符串只能被破坏：字符串，字节，数字，元组，列表，dicts，集合，布尔值和None（取自文档here）

Answer 2

在这种情况下，你很幸运，因为你的字符串似乎是有效的JSON，所以你可以这样做：

from simplejson import json

value = u'[
    [
        [
            ["Hello","w_0","w_0,w_10",1],
            [".","w_1","",0," "],
            ["This","w_2","",0," "],
        ]
        [
            ["\u4f60\u597d","w_10","w_0,w_10",1],
            ["\uff0c","w_11","",0],
            ["\u6211","w_12","",0],
        ],
        "http://www.kekenet.com/menu/201206/187574.shtml",154191
    ]
]'

array_value = json.loads(value)

为此，它依赖于使用双引号字符串的字符串（根据json规范），即["Hello","w_0","w_0,w_10",1]而不是['Hello','w_0','w_0,w_10',1]

与使用eval的解决方案相比，它更安全...如果此值来自用户输入或其他不受信任的代码，则使用eval非常危险，而json是安全的。

修改
正如已经指出的那样，张贴的字符串是无效的json ......虽然我不知道它是不是一个错字？

问题是：

value = u'[ [ [ ["Hello","w_0","w_0,w_10",1], [".","w_1","",0," "], ["This","w_2","",0," "], ] <-- should be a comma here [ ["\u4f60\u597d","w_10","w_0,w_10",1],

如何将包含list的字符串转换为真实列表

2 个答案: