Skip to content

Latest commit

 

History

History
481 lines (366 loc) · 16.8 KB

AUTHORING_GUIDE.md

File metadata and controls

481 lines (366 loc) · 16.8 KB
Raw

Python Sample Authoring Guide

We're happy you want to write a Python sample! Like a lot of Pythonistas, we're opinationed and fussy. This guide intends to be a reference for the format and style expected of samples that live in python-docs-samples.

Canonical sample

The Cloud Storage Sample is a great example of what we expect from samples. It's a great sample to copy and start from.

The basics

No matter what, all samples must:

  1. Have a license header.
  2. Pass lint.
  3. Be either a web application or a runnable console application.
  4. Have a requirements.txt containing all of its third-party dependencies.
  5. Work in Python 2.7, 3.5, & 3.6. App Engine Standard is exempt as it only supports 2.7. Our default version is currently Python 3.5.
  6. Have tests.
  7. Declare all dependencies in a requirements.txt. All requirements must be pinned.

Style & linting

We follow pep8 and the external Google Python Style Guide we verify this with flake8. In general:

  1. 4 spaces.
  2. CamelCase only for classes, snake_case elsewhere.
  3. _ for private variables, members, functions, and methods only. Samples should generally have very few private items.
  4. CAP_WITH_UNDERSCORES for constants.
  5. mixedCase is only acceptable when interface with code that doesn't follow our style guide.
  6. 79-character line limit.
  7. Imports should be in three sections separated by a blank line - standard library, third-party, and package-local. Sample will have very few package-local imports.

See Automated tools for information on how to run the lint checker.

Beyond PEP8, there are several idioms and style nits we prefer.

  1. Use single quotes (') except for docstrings (which use """).

  2. Typically import modules over members, for example from gcloud import datastore instead of from gcloud.datastore import Client. Although you should use your best judgment, for example from oauth2client.contrib.flask_util import UserOAuth2 and from oauth2client.client import GoogleCredentials are both totally fine.

  3. Never alias imports unless there is a name collision.

  4. Use .format() over % for string interpolation.

  5. Generally put a blank line above control statements that start new indented blocks, for example:

    # Good
do_stuff()

if other_stuff():
    more_stuff()

# Not so good
do_stuff()
if other_stuff():
    more_stuff()

    This rule can be relaxed for counter or accumulation variables used in loops.

  6. Don't use parenthesis on multiple return values (return one, two) or in destructuring assignment (one, two = some_function()).

  7. Prefer not to do hanging indents if possible. If you break at the first logical grouping, it shouldn't be necessary. For example:

    # Good
do_some_stuff_please(
    a_parameter, another_parameter, wow_so_many_parameters,
    much_parameter, very_function)

# Not so good
do_some_stuff_please(a_parameter, another_parameter, wow_so_many_parameters,
                     much_parameter, very_function)

    Similarly with strings and other such things:

    long_string = (
    'Alice was beginning to get very tired of sitting by her sister on '
    'the bank, and of having nothing to do: once or twice she had peeped '
    'into the book her sister was reading, but it had no pictures or ...'
)

  8. Use descriptive variables names in comprehensions, for example:

    # Good
blogs = [blog for blog in bob_laws_law_blog]

# Not so good
blogs = [x for x in bob_laws_law_blog]

The sample format

In general our sample format follows ideas borrowed from Literate Programming. Notably, your sample program should self-contained, readable from top to bottom, and should be fairly self-documenting. Prefer descriptive names. Use comments and docstrings only as needed to further explain. Always introduce functions and variables before they are used. Prefer less indirection. Prefer imperative programming as it is easier to understand.

Shebang

If, and only if, your sample application is a command-line application then include a shebang as the first line. Separate the shebang from the rest of the application with a blank line. The shebang should always be:

#!/usr/bin/env python

Don't include shebangs in web applications or test files.

License header

All samples should start with the following (modulo shebang line):

# Copyright 2017 Google, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

Module-level docstring

All samples should contain a module-level docstring. For command-line applications, this docstring will be used as the summary when -h is passed. The docstring should be succinct and should avoid repeating information available in readmes or documentation.

Here's a simple docstring for a command-line application with straightforward usage:

This application demonstrates how to perform basic operations on blobs
(objects) in a Google Cloud Storage bucket.

For more information, see the README.md under /storage and the documentation
at https://cloud.google.com/storage/docs.

Here's a docstring from a command-line application that requires a little bit more explanation:

"""This application demonstrates how to upload and download encrypted blobs
(objects) in Google Cloud Storage.

Use `generate-encryption-key` to generate an example key:

    python encryption.py generate-encryption-key

Then use the key to upload and download files encrypted with a custom key.

For more information, see the README.md under /storage and the documentation
at https://cloud.google.com/storage/docs/encryption.
"""

Finally, here's a docstring from a web application:

"""Google Cloud Endpoints sample application.

Demonstrates how to create a simple echo API as well as how to use the
various authentication methods available.

For more information, see the README.md under /appengine/flexible and the
documentation at https://cloud.google.com/appengine/docs/flexible.
"""

Functions & classes

Very few samples will require authoring classes. Prefer functions whenever possible. See this video for some insight into why classes aren't as necessary as you might think in Python. Classes also introduce cognitive load. If you do write a class in a sample be prepared to justify its existence during code review.

Always prefer descriptive function names even if they are long. For example upload_file, upload_encrypted_file, and list_resource_records. Similarly, prefer long and descriptive parameter names. For example source_file_name, dns_zone_name, and base64_encryption_key.

Here's an example of a top-level function in a command-line application:

def list_blobs(bucket_name):
    """Lists all the blobs in the bucket."""
    storage_client = storage.Client()
    bucket = storage_client.get_bucket(bucket_name)

    blobs = bucket.list_blobs()

    for blob in blobs:
        print(blob.name)

Notice the simple docstring and descriptive argument name (bucket_name implying a string instead of just bucket which could imply a class instance).

This particular function is intended to be the "top of the stack" - the function executed when the command-line sample is run by the user. As such, notice that it prints the blobs instead of returning. In general top of the stack functions in command-line applications should print, but use your best judgment.

Here's an example of a more complicated top-level function in a command-line application:

def download_encrypted_blob(
        bucket_name, source_blob_name, destination_file_name,
        base64_encryption_key):
    """Downloads a previously-encrypted blob from Google Cloud Storage.

    The encryption key provided must be the same key provided when uploading
    the blob.
    """
    storage_client = storage.Client()
    bucket = storage_client.get_bucket(bucket_name)
    blob = bucket.blob(source_blob_name)

    # Encryption key must be an AES256 key represented as a bytestring with
    # 32 bytes. Since it's passed in as a base64 encoded string, it needs
    # to be decoded.
    encryption_key = base64.b64decode(base64_encryption_key)

    blob.download_to_filename(
        destination_file_name, encryption_key=encryption_key)

    print('Blob {} downloaded to {}.'.format(
        source_blob_name,
        destination_file_name))

Note the verbose parameter names and the extended description that helps the user form context. If there were more parameters or if the parameters had complex context, then it might make sense to expand the docstring to include an Args section such as:

Args:
    bucket_name: The name of the cloud storage bucket.
    source_blob_name: The name of the blob in the bucket to download.
    destination_file_name: The blob will be downloaded to this path.
    base64_encryption_key: A base64-encoded RSA256 encryption key. Must be the
        same key used to encrypt the file.

Generally, however, it's rarely necessary to exhaustively document the parameters this way. Lean towards unsurprising arguments with descriptive names, as having to resort to this kind of docstring might be extremely accurate but it comes at the cost of high redundancy, signal-to-noise ratio, and increased cognitive load.

Finally, if absolutely necessary feel free to document the type for the parameters, for example:

Args:
    credentials (google.oauth2.credentials.Credentials): Credentials authorized
      for the current user.

If documenting primitive types, be sure to note if they have a particular set of constraints, for example A base64-encoded string or Must be between 0 and 10.

Request handlers

In general these follow the same rules as top-level functions. Here's a sample function from a web application:

@app.route('/pubsub/push', methods=['POST'])
def pubsub_push():
    """Receives push notifications from Cloud Pub/Sub."""
    # Verify the token - if it's not the same token used when creating the
    # notification channel then this request did not come from Pub/Sub.
    if (request.args.get('token', '') !=
            current_app.config['PUBSUB_VERIFICATION_TOKEN']):
        return 'Invalid request', 400

    envelope = json.loads(request.data.decode('utf-8'))
    payload = base64.b64decode(envelope['message']['data'])

    MESSAGES.append(payload)

    # Returning any 2xx status indicates successful receipt of the message.
    return 'OK', 200

Note the name of the function matches the URL route. The docstring is kept simple because reading the function body reveals how the parameters are used.

Use print or logging.info in request handlers to print useful information as needed.

Argparse section

For command-line samples, you'll need an argparse section to handle parsing command-line arguments and executing the sample functions. This section lives within the if __name__ == '__main__' clause:

if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description=__doc__,
        formatter_class=argparse.RawDescriptionHelpFormatter)

Note the use of __doc__ (the module-level docstring) as the description for the parser. This helps you not repeat yourself and gives users useful information when invoking the program. We also use RawDescriptionHelpFormatter to prevent argparse from re-formatting the docstring.

Command-line arguments should generally have a 1-to-1 match to function arguments. For example:

parser.add_argument('source_file_name')
parser.add_argument('destination_blob_name')

Again, descriptive names prevent you from having to exhaustively describe every parameter.

Some samples demonstrate multiple functions. You should use subparsers to handle this, for example:

if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description=__doc__,
        formatter_class=argparse.RawDescriptionHelpFormatter)
    parser.add_argument('bucket_name', help='Your cloud storage bucket.')

    subparsers = parser.add_subparsers(dest='command')

    subparsers.add_parser('list', help=list_blobs.__doc__)

    upload_parser = subparsers.add_parser('upload', help=upload_blob.__doc__)
    upload_parser.add_argument('source_file_name')
    upload_parser.add_argument('destination_blob_name')

    args = parser.parse_args()

    if args.command == 'list':
        list_blobs(args.bucket_name)
    elif args.command == 'upload':
        upload_blob(
            args.bucket_name,
            args.source_file_name,
            args.destination_blob_name)

Local server

For web application samples using Flask that don't run on App Engine Standard, the if __name__ == '__main__' clause should handle starting the development server:

if __name__ == '__main__':
    # This is used when running locally. Gunicorn is used to run the
    # application on Google App Engine. See entrypoint in app.yaml.
    app.run(host='127.0.0.1', port=8080, debug=True)

Writing tests

  • Use pytest-style tests and plain asserts. Don't use unittest-style tests or assertX mthods.
  • All tests in this repository are system tests. This means they hit real services and should use little to no mocking.
  • Tests should avoid doing very strict assertions. The exact output format from an API call can change, but as long as sample still works assertions should pass.
  • Tests will run against Python 2.7 and 3. The only exception is App Engine standard- these samples are only be tested against Python 2.7.
  • Samples that use App Engine Standard should use the App Engine testbed for system testing. See existing App Engine tests for how to use this.

Running tests and automated tools

Using nox

The testing of python-docs-samples is managed by nox. Nox allows us to run a variety of tests, including the linter, Python 2.7, Python 3, App Engine, and automatic README generation.

To use nox, install it globally with pip:

$ pip install nox-automation

Nox automatically discovers all samples in the repository and generates three types of sessions for each sample in this repository:

  1. A test sessions (gae, py27 and py35) for running the system tests against a specific Python version.
  2. lint sessions for running the style linter .
  3. readmegen sessions for regenerating READMEs.

Because nox generates all of these sessions, it's often useful to filter down by just the sample you're working on. For example, if you just want to see which sessions are available for storage samples:

$ nox -k storage -l
* gae(sample='./appengine/standard/storage/api-client')
* gae(sample='./appengine/standard/storage/appengine-client')
* lint(sample='./appengine/flexible/storage')
* lint(sample='./appengine/standard/storage/api-client')
* lint(sample='./appengine/standard/storage/appengine-client')
* lint(sample='./storage/api')
* lint(sample='./storage/cloud-client')
* lint(sample='./storage/transfer_service')
* py27(sample='./appengine/flexible/storage')
* py27(sample='./storage/api')
* py27(sample='./storage/cloud-client')
* py35(sample='./appengine/flexible/storage')
* py35(sample='./storage/api')
* py35(sample='./storage/cloud-client')
* readmegen(sample='./storage/api')
* readmegen(sample='./storage/cloud-client')
* readmegen(sample='./storage/transfer_service')

Now you can use nox to run a specific session, for example, if you want to lint the storage cloud-client samples:

$ nox -s "lint(sample='./storage/cloud-client')"

Test environment setup

Because all the tests here are system tests, you'll need to have a Google Cloud project with billing enabled. Once you have this configured, you'll need to set environment variables for the tests to be able to use your project and its resources. See testing/test-env.tmpl.sh for a list of all environment variables used by all tests. Not every test needs all of these variables.