Skip to main content

Utility for comparing results between data sources

Project description

Comparator

COMPARATOR

pypi versions CircleCI Coverage Status

Comparator is a utility for comparing the results of queries run against two databases. Future development will include support for APIs, static files, and more.

Installation

pip install comparator

Usage

Overview

from spackl import db

import comparator as cpt

conf = db.Config()
l = db.Postgres(**conf.default)
r = db.Postgres(**conf.other_db)
query = 'SELECT * FROM my_table ORDER BY 1'

c = cpt.Comparator(l, query, r)
c.run_comparisons()
[('basic_comp', True)]

Included Comparisons

There are some basic comparisons included, and they can be imported and passed using constants.

from comparator.comps import BASIC_COMP, LEN_COMP

c = cpt.Comparator(l, query, r, comps=[BASIC_COMP, LEN_COMP])
c.run_comparisons()
[('basic_comp', True), ('len_comp', True)]

Queries and Exceptions

It’s possible to run different queries against each database. You can raise exceptions if that’s your speed.

lq = 'SELECT * FROM my_table ORDER BY 1'
rq = 'SELECT id, uuid, name FROM reporting.my_table ORDER BY 1'
comparisons = [BASIC_COMP, LEN_COMP]

c = cpt.Comparator(l, lq, r, rq, comps=comparisons)

for result in c.compare():
    if not result:
        raise Exception('{} check failed!'.format(result.name))

Custom Comparisons

You’ll probably want to define your own comparison checks. You can do so by defining functions that accept left and right args, which correspond to the results of the queries against your “left” and “right” data source, respectively. Perform whatever magic you like, and return a boolean (or not… your choice).

def left_is_longer(left, right):
    # Return True if left contains more rows than right
    return len(left) > len(right)


def totals_are_equal(left, right):
    # Return True if sum(left) == sum(right)
    sl, sr = 0, 0
    for row in left:
        sl += int(row[1])
    for row in right:
        sr += int(row[1])
    return sl == sr


c = cpt.Comparator(l, query, r, comps=[left_is_longer, totals_are_equal])
c.run_comparisons()
[('left_is_longer', False), ('totals_are_equal', True)]

Access Comparator and Query Results

The results of both queries and comparisons can be checked using standard operators, as well as for “truthiness” (ex: failures = [result.name for result in c.compare() if result is False]).

Comparisons do not always need to return a boolean. Accessing the resulting value of such a comparison is simple.

def len_diff(left, right):
    return len(left) - len(right)


c = cpt.Comparator(l, query, r, comps=len_diff)
res = c.run_comparisons()[0]
if res == 0:
    print('They match')
elif res < 0:
    print('Left is shorter by {}'.format(res.result))
else:
    print('Left is longer by {}'.format(res.result))

It’s recommended that you use the spackl package for instantiating your “left” and “right” data source objects (pip install spackl). This package was originally part of comparator, and provides the following functionality:

Query results are contained in the QueryResult class, which provides simple yet powerful ways to look up and access the output of the query. Data can be retrieved as a dict, list, json string, or pandas DataFrame. Rows/columns can be accesed by index, attribute, or key. Iterating on the QueryResult returns a QueryResultRow, which has the same lookup functionality, as well as standard operators (<, >, =, etc).

from spackl import db

conf = db.Config()
pg = db.Postgres(**conf.default)
res = pg.query(query_string)

res          # [{'a': 1, 'b': 2, 'c': 3}, {'a': 4, 'b': 5, 'c': 6}, {'a': 7, 'b': 8, 'c': 9}]

res.a        # (1, 4, 7)
res['a']     # (1, 4, 7)
res[0]       # QueryResultRow : (1, 2, 3)

res[0].a     # 1
res[0]['a']  # 1
res[0][0]    # 1

res.dict()   # {'a': (1, 4, 7), 'b': (2, 5, 8), 'c': (3, 6, 9)}
res.list()   # [(1, 2, 3), (4, 5, 6), (7, 8, 9)]
res.first()  # QueryResultRow : (1, 2, 3)

These result sets can be used to great effect in comparison callables. For example, accessing the result of a query as a pandas DataFrame allows for an endless variety of checks/manipulations do be done on a single query output.

Support is being added to spackl to allow for querying from files and APIs using the same methods, allowing for easy comparison between many disparate data sources. Stay tuned.

CHANGELOG

0.4.0 (2019-03-09)

  • BREAKING - All source modules and methods have been stripped out
  • Functionality has been moved to the spackl package (pip install spackl)
  • The comparator package will expect spackl to be used for all left and right data sources

0.4.0rc3 (2018-12-05)

  • Adds better transaction handling in the PostgresDb class
  • Cleans up calls to connect() in the Db classes

0.4.0rc2 (2018-12-05)

  • BREAKING - QueryPair arguments order has changed (QueryPair(left, lquery, right, rquery))
  • QueryPair, Comparator, and ComparatorSet no longer require a “right” Db

0.4.0rc1 (2018-11-07)

  • DEPRECATED - the from_list method on ComparatorSet
  • adds the QueryPair class
  • BREAKING - Comparator and ComparatorSet are instantiated with QueryPair objects
  • BREAKING - ComparatorSet.from_dict() requires the dict as the first argument
  • BREAKING - QueryResult.keys() and QueryResult.values() both return generators
  • the rquery passed to a QueryPair can be formatted with the lquery query result
  • adds the QueryResultCol class
  • adds the append, pop, extend, and filter methods on QueryResult
  • downgrades pandas version requirement to >=0.22.0
  • improves docstrings on QueryResult methods
  • adds slice handling to QueryResult
  • adds empty property to QueryResult

0.3.2 (2018-10-04)

  • adds MANIFEST.in for readme and changes

0.3.1 (2018-10-03)

  • adds creds_file to possible BigQueryDb init kwargs

0.3.0 (2018-10-03)

  • DEPRECATED - the query_df method on BaseDb and subclasses
  • DEPRECATED - the output kwarg for Comparator results
  • adds the execute method on BaseDb and subclasses
  • adds the QueryResult and QueryResultRow classes
  • adds the ComparatorSet class
  • adds list_tables and delete_table methods to BigQueryDb
  • cleans up some python 2/3 compatability using six

0.2.1 (2018-09-19)

  • officially support Python 2.7, 3.6, and 3.7

0.2.0 (2018-09-18)

  • adds query_df methods for returning pandas DataFrames
  • adds output kwarg to Comparator to allow calling the query_df method

0.1.0 (2018-09-12)

  • initial release

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for comparator, version 0.4.0
Filename, size File type Python version Upload date Hashes
Filename, size comparator-0.4.0-py2.7.egg (26.2 kB) File type Egg Python version 2.7 Upload date Hashes View
Filename, size comparator-0.4.0-py3.6.egg (26.3 kB) File type Egg Python version 3.6 Upload date Hashes View
Filename, size comparator-0.4.0-py3.7.egg (26.4 kB) File type Egg Python version 3.7 Upload date Hashes View
Filename, size comparator-0.4.0.tar.gz (13.7 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page