File: parameters.rst

package info (click to toggle)
python-cyclopts 3.12.0-3
  • links: PTS, VCS
  • area: main
  • in suites: sid
  • size: 3,288 kB
  • sloc: python: 11,445; makefile: 24
file content (392 lines) | stat: -rw-r--r-- 15,237 bytes parent folder | download | duplicates (2)
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
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
==========
Parameters
==========

Typically, Cyclopts gets all the information it needs from object names, type hints, and the function docstring:

.. code-block:: python

   from cyclopts import App

   app = App(help="This is help for the root application.")

   @app.command
   def foo(value: int):  # Cyclopts uses the ``value`` name and ``int`` type hint
       """Cyclopts uses this short description for help.

       Parameters
       ----------
       value: int
           Cyclopts uses this description for ``value``'s help.
       """

   app()

Running the example:

.. code-block:: console

   $ my-script --help
   Usage: my-script COMMAND

   This is help for the root application.

   ╭─ Commands ──────────────────────────────────────────────────────────╮
   │ foo        Cyclopts uses this short description for help.           │
   │ --help,-h  Display this message and exit.                           │
   │ --version  Display application version.                             │
   ╰─────────────────────────────────────────────────────────────────────╯

   $ my-script foo --help
   Usage: my-script [ARGS] [OPTIONS]

   Cyclopts uses this short description for help.

   ╭─ Parameters ─────────────────────────────────────────────────────────────────────────╮
   │ *  VALUE --value  Cyclopts uses this description for value's help. [required]        │
   ╰──────────────────────────────────────────────────────────────────────────────────────╯

This keeps the code as clean and terse as possible.
However, if more control is required, we can provide additional information by `annotating <https://docs.python.org/3/library/typing.html#typing.Annotated>`_ type hints with :class:`.Parameter`.

.. code-block:: python

   from cyclopts import App, Parameter
   from typing import Annotated

   app = App()

   @app.command
   def foo(bar: Annotated[int, Parameter(...)]):
       pass

   app()

:class:`.Parameter` gives complete control on how Cyclopts processes the annotated parameter.
See the :ref:`API` page for all configurable options.
This page will investigate some of the more common use-cases.

.. note::
   :class:`.Parameter` can also be used as a decorator.
   This is :ref:`particularly useful for class definitions <Namespace Flattening>`.

------
Naming
------
Like :ref:`command names <Command Changing Name>`, CLI parameter names are derived from their python counterparts.
However, sometimes customization is needed.

.. _Parameters - Naming - Manual Naming:

^^^^^^^^^^^^^
Manual Naming
^^^^^^^^^^^^^
Parameter names (and their short forms) can be manually specified:

.. code-block:: python

   from cyclopts import App, Parameter
   from typing import Annotated

   app = App()

   @app.default
   def main(
       *,
       foo: Annotated[str, Parameter(name=["--foo", "-f"])],  # Adding a short-form
       bar: Annotated[str, Parameter(name="--something-else")],
   ):
       pass

   app()

.. code-block:: console

   $ my-script --help

   Usage: main COMMAND [OPTIONS]
   ╭─ Commands ──────────────────────────────────────────────╮
   │ --help -h  Display this message and exit.               │
   │ --version  Display application version.                 │
   ╰─────────────────────────────────────────────────────────╯
   ╭─ Parameters ────────────────────────────────────────────╮
   │ *  --foo             -f  [required]                     │
   │ *  --something-else      [required]                     │
   ╰─────────────────────────────────────────────────────────╯

Manually set names via :attr:`Parameter.name <cyclopts.Parameter.name>` are not subject to :attr:`Parameter.name_transform <cyclopts.Parameter.name_transform>`.


^^^^^^^^^^^^^^
Name Transform
^^^^^^^^^^^^^^
The name transform function that converts the python variable name to it's CLI counterpart can be configured by setting :attr:`Parameter.name_transform <cyclopts.Parameter.name_transform>` (defaults to :func:`.default_name_transform`).

.. code-block:: python

   from cyclopts import App, Parameter
   from typing import Annotated

   app = App()

   def name_transform(s: str) -> str:
       return s.upper()

   @app.default
   def main(
       *,
       foo: Annotated[str, Parameter(name_transform=name_transform)],
       bar: Annotated[str, Parameter(name_transform=name_transform)],
   ):
       pass

   app()

.. code-block:: console

   $ my-script --help
   Usage: main COMMAND [OPTIONS]

   ╭─ Commands ──────────────────────────────────────────────╮
   │ --help -h  Display this message and exit.               │
   │ --version  Display application version.                 │
   ╰─────────────────────────────────────────────────────────╯
   ╭─ Parameters ────────────────────────────────────────────╮
   │ *  --FOO  [required]                                    │
   │ *  --BAR  [required]                                    │
   ╰─────────────────────────────────────────────────────────╯

Notice how the parameter is now ``--FOO`` instead of the standard ``--foo``.

.. note::
   The returned string is **before** the standard ``--`` is prepended.

Generally, it is not very useful to set the name transform on **individual** parameters; it would be easier/clearer :ref:`to manually specify the name <Parameters - Naming - Manual Naming>`.
However, we can change the default name transform for the **entire app** by configuring the app's :ref:`default_parameter <Default Parameter>`.

To change the :attr:`~cyclopts.Parameter.name_transform` across your entire app, add the following to your :class:`~cyclopts.App` configuration:

.. code-block:: python

   app = App(
       default_parameter=Parameter(name_transform=my_custom_name_transform),
   )

----
Help
----
It is recommended to use docstrings for your parameter help, but if necessary, you can explicitly set a help string:

.. code-block:: python

   @app.command
   def foo(value: Annotated[int, Parameter(help="THIS IS USED.")]):
       """
       Parameters
       ----------
       value: int
           This description is not used; got overridden.
       """

.. code-block:: console

   $ my-script foo --help
   ╭─ Parameters ──────────────────────────────────────────────────╮
   │ *  VALUE,--value  THIS IS USED. [required]                    │
   ╰───────────────────────────────────────────────────────────────╯

.. _Converters:

----------
Converters
----------

Cyclopts has a powerful coercion engine that automatically converts CLI string tokens to the types hinted in a function signature.
However, sometimes a custom :attr:`~.Parameter.converter` is required.

Lets consider a case where we want the user to specify a file size, and we want to allows suffixes like `"MB"`.

.. code-block:: python

   from cyclopts import App, Parameter, Token
   from typing import Annotated, Sequence
   from pathlib import Path

   app = App()

   mapping = {
       "kb": 1024,
       "mb": 1024 * 1024,
       "gb": 1024 * 1024 * 1024,
   }

   def byte_units(type_, tokens: Sequence[Token]) -> int:
       # type_ is ``int``,
       value = tokens[0].value.lower()
       try:
           return type_(value)  # If this works, it didn't have a suffix.
       except ValueError:
           pass
       number, suffix = value[:-2], value[-2:]
       return int(number) * mapping[suffix]

   @app.command
   def zero(file: Path, size: Annotated[int, Parameter(converter=byte_units)]):
       """Creates a file of all-zeros."""
       print(f"Writing {size} zeros to {file}.")
       file.write_bytes(bytes(size))

   app()

.. code-block:: console

   $ my-script zero out.bin 100
   Writing 100 zeros to out.bin.

   $ my-script zero out.bin 1kb
   Writing 1024 zeros to out.bin.

   $ my-script zero out.bin 3mb
   Writing 3145728 zeros to out.bin.

The converter function gets the annotated type, and the :class:`.Token` s parsed for this argument.
Tokens are Cyclopt's way of bookkeeping user inputs; in the last command the ``tokens`` object would look like:

.. code-block:: python

    # tokens is a length-1 tuple. The variable "size" only takes in 1 token:
    tuple(
      Token(
         keyword=None,  # "3mb" was provided positionally, not by keyword
         value='3mb',   # The string from the command line
         source='cli',  # The value came from the command line, as opposed to other Cyclopts mechanisms.
         index=0,       # For the variable "size", this is the first (0th) token.
      ),
   )

----------------
Validating Input
----------------
Just because data is of the correct type, doesn't mean it's valid.
If we had a program that accepts integer user age as an input, ``-1`` is an integer, but not a valid age.

.. code-block:: python

   from cyclopts import App, Parameter
   from typing import Annotated

   app = App()

   def validate_age(type_, value):
       if value < 0:
           raise ValueError("Negative ages not allowed.")
       if value > 150:
           raise ValueError("You are too old to be using this application.")

   @app.default
   def allowed_to_buy_alcohol(age: Annotated[int, Parameter(validator=validate_age)]):
       print("Under 21: prohibited." if age < 21 else "Good to go!")

   app()

.. code-block:: console

   $ my-script 30
   Good to go!

   $ my-script 10
   Under 21: prohibited.

   $ my-script -1
   ╭─ Error ──────────────────────────────────────────────────────────────────────╮
   │ Invalid value "-1" for "AGE". Negative ages not allowed.                     │
   ╰──────────────────────────────────────────────────────────────────────────────╯

   $ my-script 200
   ╭─ Error ──────────────────────────────────────────────────────────────────────╮
   │ Invalid value "200" for "AGE". You are too old to be using this application. │
   ╰──────────────────────────────────────────────────────────────────────────────╯

Certain builtin error types (:exc:`ValueError`, :exc:`TypeError`, :exc:`AssertionError`) will be re-interpreted by Cyclopts and formatted into a prettier message for the application user.

Cyclopts has some :ref:`builtin validators <Parameter Validators>` for common situations
We can create a similar app as above:

.. code-block:: python

   from cyclopts import App, Parameter, validators
   from typing import Annotated

   app = App()

   @app.default
   def allowed_to_buy_alcohol(age: Annotated[int, Parameter(validator=validators.Number(gte=0, lte=150))]):
       # gte - greater than or equal to
       # lte - less than or equal to
       print("Under 21: prohibited." if age < 21 else "Good to go!")

   app()

Taking this one step further, Cyclopts has some :ref:`builtin convenience types <Annotated Types>`. If we didn't care about the upper age bound, we could simplify the application to:

.. code-block:: python

   from cyclopts import App
   from cyclopts.types import NonNegativeInt

   app = App()

   @app.default
   def allowed_to_buy_alcohol(age: NonNegativeInt):
       print("Under 21: prohibited." if age < 21 else "Good to go!")

   app()

--------------------
Parameter Resolution
--------------------
Cyclopts can combine multiple :class:`.Parameter` annotations together.
Say you want to define a new :obj:`int` type that uses the :ref:`byte-centric converter from above<Converters>`.

We can define the type:

.. code-block:: python

   ByteSize = Annotated[int, Parameter(converter=byte_units)]

We can then either directly annotate a function parameter with this:

.. code-block:: python

   @app.command
   def zero(size: ByteSize):
       pass

or even stack annotations to add additional features, like a validator:

.. code-block:: python

   def must_be_multiple_of_4096(type_, value):
       assert value % 4096 == 0, "Size must be a multiple of 4096"


   @app.command
   def zero(size: Annotated[ByteSize, Parameter(validator=must_be_multiple_of_4096)]):
       pass

Python automatically flattens out annotations, so this is interpreted as:

.. code-block:: python

   Annotated[ByteSize, Parameter(converter=byte_units), Parameter(validator=must_be_multiple_of_4096)]

Cyclopts will search **right-to-left** for **set** parameter attributes until one is found. I.e. right-most parameter attributes have the highest priority.

.. code-block:: console

   $ my-script 1234
   ╭─ Error ──────────────────────────────────────────────────────────────────────╮
   │ Invalid value "1234" for "SIZE". Size must be a multiple of 4096             │
   ╰──────────────────────────────────────────────────────────────────────────────╯

See :ref:`Parameter Resolution Order<Parameter Resolution Order>` for more details.