User Guide

Wheezy Core comes with extensions to the following features:

• benchmark
• collections
• config
• datetime
• db
• descriptors
• feistel
• gzip
• i18n
• introspection
• json
• luhn
• mail
• pooling
• retry
• url
• uuid

benchmark

The benchmark module contains a single class Benchmark that measures execution time of your code.

Here is an example:

class BenchmarkTestCase(PublicTestCase):

    def runTest(self):
        p = Benchmark((
            self.test_home,
            self.test_about
            ), 1000)
        p.report('public', baselines={
                'test_home': 1.0,
                'test_about': 0.926
        })

Sample output:

public: 2 x 1000
baseline throughput change target
  100.0%     839rps  +0.0% test_home
   96.2%     807rps  +3.9% test_about

Each of the test cases has been run 1000 times. The shows productivity relative to the first test case (which serves as a baseline for other tests), throughput in requests per second, change from baselines argument passed to report method and target being benchmarked.

Reports are being printed as soon as results are available.

collections

The collections module contains types and functions that define various collections, iterators and algorithms.

Classes:

• ItemAdapter - adapts defaultdict(list).__getitem__ accessor to return item at index from the list. If key is not found return None.
• attrdict - a dictionary with attribute-style access. Maps attribute access to dictionary.
• defaultattrdict - a dictionary with attribute-style access. Maps attribute access to dictionary. Extends defaultdict.

Functions:

• first_item_adapter() - adapts defaultdict(list).__getitem__ accessor to return the first item from the list.
• last_item_adapter() - adapts defaultdict(list).__getitem__ accessor to return the last item from the list.
• distinct() - returns generator for unique items in seq with preserved order.
• gzip_iterator() - iterates over items and returns generator of gzipped items. Argument compress_level sets compression level.

config

Config - promotes options dict to attributes. If an option can not be found in options, tries to get it from master. master must have a requested option otherwise an error is raised:

m = {'DEBUG': False}
c = Config(options={'DEBUG': True}, master=m)
assert True == c.DEBUG

master - object with dictionary or attribute style of access.

datetime

Represents an instant in time, typically expressed as a date and time of day.

Classes:

• utc - defines UTC timezone. There are two instances of the class: GMT and UTC.

Functions:

• format_http_datetime() - formats datetime to a string following rfc1123 pattern:

  >>> from wheezy.core.datetime import UTC
  >>> now = datetime(2011, 9, 19, 10, 45, 30, 0, UTC)
  >>> format_http_datetime(now)
  'Mon, 19 Sep 2011 10:45:30 GMT'
  

• parse_http_datetime() - parses a string in rfc1123 format to datetime:

  >>> parse_http_datetime('Mon, 19 Sep 2011 10:45:30 GMT')
  datetime.datetime(2011, 9, 19, 10, 45, 30)
  

• total_seconds() - returns a total number of seconds for the given time delta (datetime.timedelta or int):

  >>> total_seconds(timedelta(hours=2))
  7200
  

db

TODO

descriptors

TODO

feistel

TODO

gzip

One-shot compression and decompression is provided through the compress() and decompress() functions.

httpclient

HTTPClient sends HTTP requests to a server in order to accomplish an application specific use cases, e.g. remote web server API, etc:

>>> from wheezy.core.httpclient import HTTPClient
>>> c = HTTPClient('http://buildbot.buildbot.net/json/')
>>> c.get('project')
200
>>> project = c.json
>>>> str(project.title)
Buildbot

Here is another example that demonstarates etag handling (the second time we request events the server responds with HTTP status code 304, not modified):

>>> c = HTTPClient('https://api.github.com/repos/python/cpython/')
>>> c.get('events')
200
>>> c.headers['content-encoding']
['gzip']
>>> c.get('events')
304

Supports: HTTP(S) GET/HEAD/POST verbs, follows redirects, handles cookies and etags between requests, gzip content encoding.

i18n

Internationalisation is the process of adapting an application to different languages, regional differences and technical requirements. Internationalization of software is designing an application so that it can be adapted to various languages and regions without engineering changes.

gettext is an internationalization and localization (i18n) system commonly used for writing multilingual programs on Unix-like operating systems.

TranslationsManager - manages several languages and translation domains. You can use method load() to load all available languages and domains from the given directory (typically it is i18n directory within our application root directory).

Translations directory structure must follow gettext requirements (this is how it looks up data below the i18n directory):

{localedir}/{lang}/LC_MESSAGES/{domain}.mo

In order to generate a .mo file from a .po file:

$ msgfmt domain.po

TranslationsManager supports the following arguments in initialization:

• directories - a list of directories that holds translations.
• default_lang - a default language in translations. Defaults to en.

TranslationsManager supports a fallback mechanism. You can use add_fallback() to add fallback languages.

>>> from wheezy.core.i18n import TranslationsManager
>>> tm = TranslationsManager(['i18n'], default_lang='en')
>>> tm.add_fallback(('uk', 'ru'))
>>> tm.fallbacks
{'uk': ('uk', 'ru', 'en')}

The default language is always appended to the fallback list.

TranslationsManager supports dictionary access that accepts a language code as a key. So the following represents all translations related to en language code:

lang = tm['en']

lang is an instance of defaultattrdict where attributes correspond to translation file (translation domain), if it is not available there is fallback to an instance of gettext.NullTranslations:

assert 'Hello' == lang.messages.gettext('hello')

Seamless integration with gettext module simplifies your application internationalization and localization.

introspection

Type introspection is a capability to determine the type of an object at runtime.

import_name() - dynamically imports an object by its full name. The following two imports are equivalent:

from datetime import timedelta
import_name('datetime.timedelta')

import_name() lets you introduce lazy imports into your application.

json

Extends the standard json module from Python2.6 and simplejson for Python2.5 with support for date, datetime, time and Decimal types.

• json_encode() encodes obj as a JSON formatted string. Correctly escapes forward slashes to be able to embed javascript code. Decimal objects are converted to string (same applies when used with simplejson).
• json_decode() decodes a JSON document to a Python object. Float is parsed as Decimal.

luhn

TODO

mail

TODO

pooling

TODO

retry

TODO

url

Every URL consists of the following: the scheme name (or protocol), followed by a colon and two slashes, then, a domain name (alternatively, IP address), a port number (optionally), the path of the resource to be fetched, a query string, and an optional fragment identifier. Here is the syntax:

scheme://domain:port/path?query_string#fragment_id

The url module provides integration with urlparse module.

UrlParts - concrete class for urlparse.urlsplit() results, where argument parts is a tupple of length 6. There are the following methods:

• geturl() - returns the re-combined version of the original URL as a string.
• join(other) - joins with another UrlParts instance by taking none-empty values from other. Returns new UrlParts instance.

There is a factory function urlparts() for UrlParts, that let you create an instance of UrlParts with partial content.

uuid

A universally unique identifier (UUID) is an identifier that enable distributed systems to uniquely identify information without significant central coordination. A UUID is a 16-byte (128-bit) number.

The following functions available:

• shrink_uuid() - returns base64 representation of a uuid:

  >>> shrink_uuid(UUID('a4af2f54-e988-4f5c-bfd6-351c79299b74'))
  'pK8vVOmIT1y_1jUceSmbdA'
  

• parse_uuid() - decodes a base64 string to uuid:

  >>> parse_uuid('pK8vVOmIT1y_1jUceSmbdA')
  UUID('a4af2f54-e988-4f5c-bfd6-351c79299b74')
  

There is also a module attribute UUID_EMPTY defined, that is just an instance of UUID '00000000-0000-0000-0000-000000000000'.