summaryrefslogtreecommitdiff
path: root/jsonschema/_utils.py
blob: 97b7f8f5a28293f9c4593f6e60cbe8bc2f746611 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
from collections.abc import Mapping, MutableMapping, Sequence
from urllib.parse import urlsplit
import itertools
import json
import pkgutil
import re


class URIDict(MutableMapping):
    """
    Dictionary which uses normalized URIs as keys.
    """

    def normalize(self, uri):
        return urlsplit(uri).geturl()

    def __init__(self, *args, **kwargs):
        self.store = dict()
        self.store.update(*args, **kwargs)

    def __getitem__(self, uri):
        return self.store[self.normalize(uri)]

    def __setitem__(self, uri, value):
        self.store[self.normalize(uri)] = value

    def __delitem__(self, uri):
        del self.store[self.normalize(uri)]

    def __iter__(self):
        return iter(self.store)

    def __len__(self):
        return len(self.store)

    def __repr__(self):
        return repr(self.store)


class Unset(object):
    """
    An as-of-yet unset attribute or unprovided default parameter.
    """

    def __repr__(self):
        return "<unset>"


def load_schema(name):
    """
    Load a schema from ./schemas/``name``.json and return it.
    """

    data = pkgutil.get_data("jsonschema", "schemas/{0}.json".format(name))
    return json.loads(data.decode("utf-8"))


def format_as_index(indices):
    """
    Construct a single string containing indexing operations for the indices.

    For example, [1, 2, "foo"] -> [1][2]["foo"]

    Arguments:

        indices (sequence):

            The indices to format.
    """

    if not indices:
        return ""
    return "[%s]" % "][".join(repr(index) for index in indices)


def find_additional_properties(instance, schema):
    """
    Return the set of additional properties for the given ``instance``.

    Weeds out properties that should have been validated by ``properties`` and
    / or ``patternProperties``.

    Assumes ``instance`` is dict-like already.
    """

    properties = schema.get("properties", {})
    patterns = "|".join(schema.get("patternProperties", {}))
    for property in instance:
        if property not in properties:
            if patterns and re.search(patterns, property):
                continue
            yield property


def extras_msg(extras):
    """
    Create an error message for extra items or properties.
    """

    if len(extras) == 1:
        verb = "was"
    else:
        verb = "were"
    return ", ".join(repr(extra) for extra in extras), verb


def types_msg(instance, types):
    """
    Create an error message for a failure to match the given types.

    If the ``instance`` is an object and contains a ``name`` property, it will
    be considered to be a description of that object and used as its type.

    Otherwise the message is simply the reprs of the given ``types``.
    """

    reprs = []
    for type in types:
        try:
            reprs.append(repr(type["name"]))
        except Exception:
            reprs.append(repr(type))
    return "%r is not of type %s" % (instance, ", ".join(reprs))


def flatten(suitable_for_isinstance):
    """
    isinstance() can accept a bunch of really annoying different types:

        * a single type
        * a tuple of types
        * an arbitrary nested tree of tuples

    Return a flattened tuple of the given argument.
    """

    types = set()

    if not isinstance(suitable_for_isinstance, tuple):
        suitable_for_isinstance = (suitable_for_isinstance,)
    for thing in suitable_for_isinstance:
        if isinstance(thing, tuple):
            types.update(flatten(thing))
        else:
            types.add(thing)
    return tuple(types)


def ensure_list(thing):
    """
    Wrap ``thing`` in a list if it's a single str.

    Otherwise, return it unchanged.
    """

    if isinstance(thing, str):
        return [thing]
    return thing


def _mapping_equal(one, two):
    """
    Check if two mappings are equal using the semantics of `equal`.
    """
    if len(one) != len(two):
        return False
    return all(
        key in two and equal(value, two[key])
        for key, value in one.items()
    )


def _sequence_equal(one, two):
    """
    Check if two sequences are equal using the semantics of `equal`.
    """
    if len(one) != len(two):
        return False
    return all(equal(i, j) for i, j in zip(one, two))


def equal(one, two):
    """
    Check if two things are equal evading some Python type hierarchy semantics.

    Specifically in JSON Schema, evade `bool` inheriting from `int`,
    recursing into sequences to do the same.
    """
    if isinstance(one, str) or isinstance(two, str):
        return one == two
    if isinstance(one, Sequence) and isinstance(two, Sequence):
        return _sequence_equal(one, two)
    if isinstance(one, Mapping) and isinstance(two, Mapping):
        return _mapping_equal(one, two)
    return unbool(one) == unbool(two)


def unbool(element, true=object(), false=object()):
    """
    A hack to make True and 1 and False and 0 unique for ``uniq``.
    """

    if element is True:
        return true
    elif element is False:
        return false
    return element


def uniq(container):
    """
    Check if all of a container's elements are unique.

    Tries to rely on the container being recursively sortable, or otherwise
    falls back on (slow) brute force.
    """
    try:
        sort = sorted(unbool(i) for i in container)
        sliced = itertools.islice(sort, 1, None)

        for i, j in zip(sort, sliced):
            return not _sequence_equal(i, j)

    except (NotImplementedError, TypeError):
        seen = []
        for e in container:
            e = unbool(e)

            for i in seen:
                if equal(i, e):
                    return False

            seen.append(e)
    return True