mockthink

A pure-python in-memory mock of rethinkdb


License
Other
Install
pip install mockthink==0.9.5

Documentation

MockThink

MockThink is an in-process Python clone of RethinkDB's API. For testing.

It's not done yet. Don't use it. But if you want to know how you would use it, keep reading.

MockThink provides a stub connection object which can be passed to normal ReQL queries. Instead of being serialized and sent to the server, the ReQL AST is run through an interpreter in the same process. "Tables" and "databases" are based on data given to the MockThink constructor.

Avoiding network calls (for tests themselves as well as setup/teardown) makes testing queries with MockThink orders of magnitude faster.

Usage

Basic

    from pprint import pprint
    from mockthink import MockThink
    import rethinkdb as r

    db = MockThink({
        'dbs': {
            'tara': {
                'tables': {
                    'people': [
                        {'id': 'john-id', 'name': 'John'},
                        {'id': 'sam-id', 'name': 'Sam'}
                    ]
                }
            }
        }
    })

    with db.connect() as conn:
        result = r.db('tara').table('people').map(
            lambda doc: doc.merge({'also_name': doc['name']})
        ).run(conn)
        pprint(list(result))

        # [
        #    {'also_name': 'John', 'id': 'john-id', 'name': 'John'},
        #    {'also_name': 'Sam', 'id': 'sam-id', 'name': 'Sam'}
        # ]

        r.db('tara').table('people').update(
            {'likes_fonz': True}
        ).run(conn)

        result = r.db('tara').table('people').run(conn)
        pprint(list(result))

        # [
        #    {'id': 'john-id', 'likes_fonz': True, 'name': 'John'},
        #    {'id': 'sam-id', 'likes_fonz': True, 'name': 'Sam'}
        # ]

    # data is reset at exit of context manager above

    with db.connect() as conn:
        result = r.db('tara').table('people').run(conn)
        pprint(list(result))
        # [
        #    {'id': 'john-id', 'name': 'John'},
        #    {'id': 'sam-id', 'name': 'Sam'}
        # ]

Full support for secondary indexes

    from pprint import pprint
    from mockthink import MockThink
    import rethinkdb as r

    db = MockThink({
        'dbs': {
            'tara': {
                'tables': {
                    'people': [
                        {'id': 'john-id', 'first_name': 'John', 'last_name': 'Generic'},
                        {'id': 'sam-id', 'first_name': 'Sam', 'last_name': 'Dull'},
                        {'id': 'adam-id', 'first_name': 'Adam', 'last_name': 'Average'}
                    ]
                }
            }
        }
    })

    with db.connect() as conn:

        r.db('tara').table('people').index_create(
            'full_name',
            lambda doc: doc['last_name'] + doc['first_name']
        ).run(conn)

        r.db('tara').table('people').index_wait().run(conn)

        result = r.db('tara').table('people').get_all(
            'GenericJohn', 'AverageAdam', index='full_name'
        ).run(conn)
        pprint(list(result))
        # {'id': 'john-id', 'first_name': 'John', 'last_name': 'Generic'},
        # {'id': 'adam-id', 'first_name': 'Adam', 'last_name': 'Average'}

Testing

The most confusing test failures are those caused by errors in test frameworks and harnesses themselves. This means they need to be tested very thoroughly.

The main testing is a suite of functional tests which are targeted at the individual query level, e.g. testing all the ways in which r.merge might be called. These are all complete ReQL queries, but avoid complexity beyond the target query to make failures easier to diagnose.

The integration tests cover more complicated queries, e.g. eq_join->map->eq_join->map.

Both the functional and integration tests have two modes of execution, mockthink and rethink. The second mode runs the same tests against a running RethinkDB instance, and is much slower due to the network calls. mockthink mode is for testing MockThink's behavior against our expectations; rethink mode is for testing our expectations against reality.

Roadmap

The vast majority of query terms are already implemented. The following terms remain prior to initial release:

  • index_status, info, http, coerce_to, expr, default, for_each, binary, args, match
  • ReQL time functions: to/from iso8601, to/from epoch time.

Will be implemented after initial release:

  • Domain-specific assertions, e.g. mockthink.assert_document_exists(db, table, doc_to_match). Consider these very imporant, just not essential.
  • Geospatial queries: I plan on implementing these, but so far I haven't needed any of Rethink's geospatial features. I do use all the other features, and need to test the code relying on those features sooner rather than later. That makes geospatial queries a lower priority.
  • changes stream: Probably useful but orthogonal to most query types. Missing this term won't affect testability of other queries.

May never be implemented:

  • js: this query evaluates arbitrary javascript code on the server. It's not impossible to handle -- I imagine there's a Python interface to V8 that could do most of the heavy lifting. But I haven't ever needed to use this, and doubt that I ever will. For now I don't plan on implementing it.

License

The MIT License (MIT)

Copyright (c) 2014 Scott Ivey

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.