summaryrefslogtreecommitdiffstats
path: root/README.md
blob: 2b8f227f90dc492d179675706fed4155152ee9bf (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
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
[![Build Status](https://dev.azure.com/asottile/asottile/_apis/build/status/asottile.cfgv?branchName=master)](https://dev.azure.com/asottile/asottile/_build/latest?definitionId=24&branchName=master)
[![Azure DevOps coverage](https://img.shields.io/azure-devops/coverage/asottile/asottile/24/master.svg)](https://dev.azure.com/asottile/asottile/_build/latest?definitionId=24&branchName=master)

cfgv
====

Validate configuration and produce human readable error messages.

## Installation

`pip install cfgv`

## Sample error messages

These are easier to see by example.  Here's an example where I typo'd `true`
in a [pre-commit](https://pre-commit.com) configuration.

```
pre_commit.clientlib.InvalidConfigError:
==> File /home/asottile/workspace/pre-commit/.pre-commit-config.yaml
==> At Config()
==> At key: repos
==> At Repository(repo='https://github.com/pre-commit/pre-commit-hooks')
==> At key: hooks
==> At Hook(id='flake8')
==> At key: always_run
=====> Expected bool got str
```

## API

### `cfgv.validate(value, schema)`

Perform validation on the schema:
- raises `ValidationError` on failure
- returns the value on success (for convenience)

### `cfgv.apply_defaults(value, schema)`

Returns a new value which sets all missing optional values to their defaults.

### `cfgv.remove_defaults(value, schema)`

Returns a new value which removes all optional values that are set to their
defaults.

### `cfgv.load_from_filename(filename, schema, load_strategy, exc_tp=ValidationError)`

Load a file given the `load_strategy`.  Reraise any errors as `exc_tp`.  All
defaults will be populated in the resulting value.

Most useful when used with `functools.partial` as follows:

```python
load_my_cfg = functools.partial(
    cfgv.load_from_filename,
    schema=MY_SCHEMA,
    load_strategy=json.loads,
    exc_tp=MyError,
)
```

## Making a schema

A schema validates a container -- `cfgv` provides `Map` and `Array` for
most normal cases.

### writing your own schema container

If the built-in containers below don't quite satisfy your usecase, you can
always write your own.  Containers use the following interface:

```python
class Container(object):
    def check(self, v):
        """check the passed in value (do not modify `v`)"""

    def apply_defaults(self, v):
        """return a new value with defaults applied (do not modify `v`)"""

    def remove_defaults(self, v):
        """return a new value with defaults removed (do not modify `v`)"""
```

### `Map(object_name, id_key, *items)`

The most basic building block for creating a schema is a `Map`

- `object_name`: will be displayed in error messages
- `id_key`: will be used to identify the object in error messages.  Set to
  `None` if there is no identifying key for the object.
- `items`: validator objects such as `Required` or `Optional`

Consider the following schema:

```python
Map(
    'Repo', 'url',
    Required('url', check_any),
)
```

In an error message, the map may be displayed as:

- `Repo(url='https://github.com/pre-commit/pre-commit')`
- `Repo(url=MISSING)` (if the key is not present)

### `Array(of, allow_empty=True)`

Used to nest maps inside of arrays.  For arrays of scalars, see `check_array`.

- `of`: A `Map` / `Array` or other sub-schema.
- `allow_empty`: when `False`, `Array` will ensure at least one element.

When validated, this will check that each element adheres to the sub-schema.

## Validator objects

Validator objects are used to validate key-value-pairs of a `Map`.

### writing your own validator

If the built-in validators below don't quite satisfy your usecase, you can
always write your own.  Validators use the following interface:

```python
class Validator(object):
    def check(self, dct):
        """check that your specific key has the appropriate value in `dct`"""

    def apply_default(self, dct):
        """modify `dct` and set the default value if it is missing"""

    def remove_default(self, dct):
        """modify `dct` and remove the default value if it is present"""
```

It may make sense to _borrow_ functions from the built in validators.  They
additionally use the following interface(s):

- `self.key`: the key to check
- `self.check_fn`: the [check function](#check-functions)
- `self.default`: a default value to set.

### `Required(key, check_fn)`

Ensure that a key is present in a `Map` and adheres to the
[check function](#check-functions).

### `RequiredRecurse(key, schema)`

Similar to `Required`, but uses a [schema](#making-a-schema).

### `Optional(key, check_fn, default)`

If a key is present, check that it adheres to the
[check function](#check-functions).

- `apply_defaults` will set the `default` if it is not present.
- `remove_defaults` will remove the value if it is equal to `default`.

### `OptionalRecurse(key, schema, default)`

Similar to `Optional` but uses a [schema](#making-a-schema).

- `apply_defaults` will set the `default` if it is not present and then
  validate it with the schema.
- `remove_defaults` will remove defaults using the schema, and then remove the
  value it if it is equal to `default`.

### `OptionalNoDefault(key, check_fn)`

Like `Optional`, but does not `apply_defaults` or `remove_defaults`.

### `Conditional(key, check_fn, condition_key, condition_value, ensure_absent=False)`

- If `condition_key` is equal to the `condition_value`, the specific `key`
will be checked using the [check function](#check-functions).
- If `ensure_absent` is `True` and the condition check fails, the `key` will
be checked for absense.

Note that the `condition_value` is checked for equality, so any object
implementing `__eq__` may be used.  A few are provided out of the box
for this purpose, see [equality helpers](#equality-helpers).

### `ConditionalOptional(key, check_fn, default, condition_key, condition_value, ensure_absent=False)`

Similar to ``Conditional`` and ``Optional``.

### `ConditionalRecurse(key, schema, condition_key, condition_value, ensure_absent=True)`

Similar to `Conditional`, but uses a [schema](#making-a-schema).

### `NoAdditionalKeys(keys)`

Use in a mapping to ensure that only the `keys` specified are present.

## Equality helpers

Equality helpers at the very least implement `__eq__` for their behaviour.

They may also implement `def describe_opposite(self):` for use in the
`ensure_absent=True` error message (otherwise, the `__repr__` will be used).

### `Not(val)`

Returns `True` if the value is not equal to `val`.

### `In(*values)`

Returns `True` if the value is contained in `values`.

### `NotIn(*values)`

Returns `True` if the value is not contained in `values`.

## Check functions

A number of check functions are provided out of the box.

A check function takes a single parameter, the `value`, and either raises a
`ValidationError` or returns nothing.

### `check_any(_)`

A noop check function.

### `check_type(tp, typename=None)`

Returns a check function to check for a specific type.  Setting `typename`
will replace the type's name in the error message.

For example:

```python
Required('key', check_type(int))
# 'Expected bytes' in both python2 and python3.
Required('key', check_type(bytes, typename='bytes'))
```

Several type checking functions are provided out of the box:

- `check_bool`
- `check_bytes`
- `check_int`
- `check_string`
- `check_text`

### `check_one_of(possible)`

Returns a function that checks that the value is contained in `possible`.

For example:

```python
Required('language', check_one_of(('javascript', 'python', 'ruby')))
```

### `check_regex(v)`

Ensures that `v` is a valid python regular expression.

### `check_array(inner_check)`

Returns a function that checks that a value is a sequence and that each
value in that sequence adheres to the `inner_check`.

For example:

```python
Required('args', check_array(check_string))
```

### `check_and(*fns)`

Returns a function that performs multiple checks on a value.

For example:

```python
Required('language', check_and(check_string, my_check_language))
```