Scrapy Autounit

Documentation

• Overview
• Installation
• Usage
• Caveats
• Settings
• Command Line Interface
• Internals
   

Overview

Scrapy-Autounit is an automatic test generation tool for your Scrapy spiders.

It generates test fixtures and tests cases as you run your spiders.

The fixtures are generated from the items and requests that your spider returns, then the test cases evaluate those fixtures against your spiders' callbacks.

Scrapy Autounit generates fixtures and tests per spider and callback under the Scrapy project root directory.
Here is an example of the directory tree of your project once the fixtures are created:

my_project
├── autounit
│   ├── __init__.py
│   └── tests
│       ├── __init__.py
│       └── my_spider
│           ├── __init__.py
│           └── my_callback
│               ├── __init__.py
│               ├── fixture1.bin
│               ├── fixture2.bin
│               ├── test_fixtures.py
├── my_project
│   ├── __init__.py
│   ├── items.py
│   ├── middlewares.py
│   ├── pipelines.py
│   ├── settings.py
│   └── spiders
│       ├── __init__.py
│       └── my_spider.py
└── scrapy.cfg

Installation

pip install scrapy_autounit

Usage

Add the spider middleware to your SPIDER_MIDDLEWARES setting (no specific order required):

SPIDER_MIDDLEWARES = {
    'scrapy_autounit.AutounitMiddleware': 950
}

Generating tests

Make sure you enable Scrapy Autounit in your settings:

AUTOUNIT_ENABLED = True

To generate your fixtures and tests just run your spiders as usual, Scrapy Autounit will generate them for you.

$ scrapy crawl my_spider

When the spider finishes, a directory autounit is created in your project root dir, containing all the generated tests/fixtures for the spider you just ran (see the directory tree example above).

If you want to update your tests and fixtures you only need to run your spiders again or use the autounit update command line tool.

Running tests

To run your tests you can use unittest regular commands.

Test all

$ python -m unittest discover autounit/tests/

Test a specific spider

$ python -m unittest discover autounit/tests/my_spider/

Test a specific callback

$ python -m unittest discover autounit/tests/my_spider/my_callback/

Caveats

• Keep in mind that as long as AUTOUNIT_ENABLED is on, each time you run a spider tests/fixtures are going to be generated for its callbacks.
  This means that if you have your tests/fixtures ready to go, this setting should be off to prevent undesired overwrites.
  Each time you want to regenerate your tests (e.g.: due to changes in your spiders), you can turn this on again and run your spiders as usual.
  For example, this setting should be off when running your spiders in Scrapy Cloud.

• Autounit uses an internal _autounit_cassette key in requests' meta dictionaries. Avoid using/overriding this key in your spiders when adding data to meta to prevent unexpected behaviours.
   

Settings

General

• AUTOUNIT_ENABLED
  Set this to True or False to enable or disable unit test generation.

• AUTOUNIT_MAX_FIXTURES_PER_CALLBACK
  Sets the maximum number of fixtures to store per callback.
  Minimum: 10
Default: 10

• AUTOUNIT_EXTRA_PATH
  This is an extra string element to add to the test path and name between the spider name and callback name. You can use this to separate tests from the same spider with different configurations.
  Default: None

• AUTOUNIT_FIXTURE_NAMING_ATTR
  Allows you to specify a spider attribute to be used in your fixtures names.
  The fixture name will consist of your specified attribute value and the internal callback counter.
  fixture_{your_spider_attr}_{internal_counter}.bin
  If this settings is not specified, the default fixtureN.bin naming will be used.
  Default: None

Output

• AUTOUNIT_DONT_TEST_OUTPUT_FIELDS
  Sets a list of fields to be skipped from testing your callbacks' items. It's useful to bypass fields that return a different value on each run.
  For example if you have a field that is always set to datetime.now() in your spider, you probably want to add that field to this list to be skipped on tests. Otherwise you'll get a different value when you're generating your fixtures than when you're running your tests, making your tests fail.
  Default: []

Requests

• AUTOUNIT_DONT_TEST_REQUEST_ATTRS
  Sets a list of request attributes to be skipped when running your tests.
  Default: []

• AUTOUNIT_DONT_RECORD_HEADERS
  Sets a list of headers to exclude from requests recording.
  For security reasons, Autounit already excludes Authorization and Proxy-Authorization headers by default, if you want to record them in your fixtures see AUTOUNIT_RECORD_AUTH_HEADERS.
  Default: []

• AUTOUNIT_RECORD_AUTH_HEADERS
  If you want to include Authorization or Proxy-Authorization headers in your fixtures, add one or both of them to this list.
  Default: []

• AUTOUNIT_DONT_RECORD_META
  Sets a list of jmespath-like keys to exclude from requests meta recording.
  These keys will not be recorded in autounit fixtures.
  Keys syntax supported so far: single_key, key.subkey, key.list[], key.list[].subkey
Default: []

• AUTOUNIT_DONT_TEST_META
  Same as AUTOUNIT_DONT_RECORD_META but this only skips the specified meta keys at the testing stage.
  The keys will be recorded in autounit fixtures.
  Default: []

Spider attributes

• AUTOUNIT_DONT_RECORD_SPIDER_ATTRS
  Sets a list of spider attributes that won't be recorded into your fixtures.
  Default: []

• AUTOUNIT_DONT_TEST_SPIDER_ATTRS
  Sets a list of spider attributes to be skipped from testing your callbacks. These attributes will still be recorded.
  Default: []

Settings

• AUTOUNIT_RECORD_SETTINGS
  Sets a list of settings names to be recorded in the generated test case.
  Default: []

Note: Remember that you can always apply any of these settings per spider including them in your spider's custom_settings class attribute - see https://docs.scrapy.org/en/latest/topics/settings.html#settings-per-spider.
 

Command line interface

• autounit inspect: inspects fixtures returning a JSON object
• autounit update: updates fixtures to callback changes

autounit inspect

To inspect a fixture's data, you need to pass the spider, callback and fixture name to the command:

$ autounit inspect my_spider my_callback fixture3

The fixture can also be passed as a number indicating which fixture number to inspect like this:

$ autounit inspect my_spider my_callback 3

Note: It's highly recommended that your fixtures were generated with scrapy-autounit 0.0.22 or higher to inspect data.

Extracted Data

This command returns a JSON object that can be parsed with tools like jq to inspect specific blocks of data.

The top-level keys of the JSON output are:

spider_name
The name of the spider.

request
The original request that triggered the callback.

response
The response obtained from the original request and passed to the callback.

output_data
The callback's output such as items and requests.
Same as result prior to v0.0.28.

middlewares
The relevant middlewares to replicate when running the tests.

settings
The settings explicitly recorded by the AUTOUNIT_INCLUDED_SETTINGS setting.

init_attrs
The spider's attributes right after its __init__ call.

input_attrs
The spider's attributes right before running the callback.
Same as spider_args or spider_args_in prior to v0.0.28.

output_attrs
The spider's attributes right after running the callback.
Same as spider_args_out prior to v0.0.28.

Then for example, to inspect a fixture's specific request we can do the following:

$ autounit inspect my_spider my_callback 4 | jq '.request'

autounit update

This command updates your fixtures to match your latest changes, avoiding to run the whole spider again.
You can update the whole project, an entire spider, just a callback or a single fixture.

Update the whole project

$ autounit update
WARNING: this will update all the existing fixtures from the current project
Do you want to continue? (y/n)

Update every callback in a spider

$ autounit update -s my_spider

Update every fixture in a callback

$ autounit update -s my_spider -c my_callback

Update a single fixture

# Update fixture number 5
$ autounit update -s my_spider -c my_callback -f 5

Internals

The AutounitMiddleware uses a Recorder to record Cassettes in binary fixtures.

Then, the tests use a Player to playback those Cassettes and compare its output against your current callbacks.

The fixtures contain a pickled and compressed Cassette instance that you can get programmatically by doing:

from scrapy_autounit.cassette import Cassette

cassette = Cassette.from_fixture(path_to_your_fixture)
# cassette.request
# cassette.response
# cassette.output_data
# ...

If you know what you're doing, you can modify that cassette and re-record it by using:

from scrapy_autounit.recorder import Recorder

Recorder.update_fixture(cassette, path)