Question

我在列表list indices must be integers, not str的迭代过程中收到错误meters_info。为什么我会收到这样的错误？谢谢大家指出我，原始应用程序在 django ，我正在将其转换为烧瓶应用程序。在django，他们是datastructures.SortedDict。我如何在烧瓶中使用相同的功能 See this link

from django.utils import datastructures
meters_info = datastructures.SortedDict([
            ("instance", {
                'label': '',
                'description': _("Existence of instance"),
            }),
            ("instance:<type>", {
                'label': '',
                'description': _("Existence of instance <type> "
                                 "(openstack types)"),
            }),
            ("memory", {
                'label': '',
                'description': _("Volume of RAM"),
            }),])

Flask App

def _get_nova_meters_info(self):
        meters_info= [
            ("instance", {
                'label': '',
                'description': "Existence of instance",
            }),
            ("instance:<type>", {
                'label': '',
                'description': "Existence of instance <type> (openstack types)",
            }),
            ("memory", {
                'label': '',
                'description': "Volume of RAM",
            }),]

在控制台中收到消息

File "/home/vagrant/api/ceilometer.py", line 137, in _get_nova_meters_info

meters_info[name]=dict(meters_info["instance:<type>"])
TypeError: list indices must be integers, not str

请参阅以下我收到错误的方法

for flavor in self.get_flavor_names():
            name='instance:%s' %flavor
            meters_info[name]=dict(meters_info["instance:<type>"])

            meters_info[name]['description']= (
                'Duration of instance type %s (openstack flavor)' % flavor)
        return meters_info


def get_flavor_names(self):
        return ['m1.tiny', 'm1.small', 'm1.medium', 'm1.large', 'm1.nano','m1.xlarge', 'm1.micro']

Answer 1

您有一个包含词典的元组列表。无法使用字符串访问，您需要使用整数。最初的Django应用程序使用了一种名为SortedDict的特殊类型的Django字典，幸好Python自2.7以来就有一个等价的OrderedDict。你只需要像这样导入它

from collections import OrderedDict

然后调整语法以使用OrderedDict并删除值周围的_()：

meters_info = OrderedDict([
            ("instance", {
                'label': '',
                'description': "Existence of instance",
            }),
            ("instance:<type>", {
                'label': '',
                'description': "Existence of instance <type> "
                                 "(openstack types)",
            }),
            ("memory", {
                'label': '',
                'description': "Volume of RAM",
            }),])

Answer 2

要以这种方式使用数据，您需要将meter_info重建为字典：

        meters_info= {
            "instance" : {
                'label': '',
                'description': "Existence of instance",
            },
            "instance:<type>" : {
                'label': '',
                'description': "Existence of instance <type> (openstack types)",
            }, 
            "memory" : {
                'label': '',
                'description': "Volume of RAM",
            },}

如果你有相同的结构来使用类，它也会更好。例如

        class Meters_info_param_stucture(object):
            def __init__(self, label = None, description = None):
                self.label = label
                self.description = description

        meters_info= {
            "instance" : Meters_info_param_stucture(description = "Existence of instance"),
            "instance:<type>" : Meters_info_param_stucture(description = "Existence of instance <type> (openstack types)"),
            "memory" : Meters_info_param_stucture(description = "Volume of RAM"),}

        >>>meters_info["memory"].description
        'Volume of RAM'

Answer 3

问题是您的数组是list tuple个，而tuple有str作为第一个元素，而dict是meters_info[0][0][<some_key>]第二。所以你应该像这样访问它：

meters_info[0][0]["instance"]

例如：

meters_info

但是我不明白你保持这种结构和原因。您应该考虑将bitwise_and重组为字典，例如Dmitry.Samborskyi suggests。

TypeError：列表索引必须是整数，而不是Flask App中的str

3 个答案: