root / classifier / doc / storage.txt

=======
storage
=======

Storage is a backend used to store data for the bayesian network.

Let's work in a special database::

    >>> db_file = 'test.db'
    >>> import settings
    >>> settings.SQLURI = 'sqlite:///%s' % db_file
    >>> import os
    >>> if os.path.exists(db_file):
    ...     os.remove(db_file)


It works with languages, words, categories, and words within categories::

    >>> from storage import SQLStorage

Each storage is affiliated with a user, so each user can have it's own
bayesian data::

    >>> storage = SQLStorage('tester')

Next we can store languages, since each word is in a given language::

    >>> storage.list_languages()
    []
    >>> storage.add_language('fr')
    >>> storage.add_language('en')
    >>> storage.list_languages()
    [u'fr', u'en']
    >>> storage.del_language('fr')
    >>> storage.list_languages()
    [u'en']

We can work with categories::

    >>> storage.list_categories()
    []
    >>> storage.add_category('one')
    >>> storage.add_category('two')
    >>> storage.list_categories()
    [u'one', u'two']
    >>> storage.del_category('one')
    >>> storage.list_categories()
    [u'two']


And words, that are specific to a language. If `list_words` does
not have the language as a parameter, it will return all words of
all languages::

    >>> res = storage.list_words()
    >>> res
    <generator object at 0x...>

Since we get a generator (for optimisation purpose) we need to act upon::

    >>> list(res)
    []
    >>> storage.add_word('coucou', 'fr')
    >>> storage.add_word('hello', 'en')
    >>> list(storage.list_words())
    [(u'coucou', u'fr'), (u'hello', u'en')]
    >>> list(storage.list_words('fr'))
    [(u'coucou', u'fr')]

We can also add words within categories::

    >>> storage.add_word('coucou', 'fr', ('cat1', 'cat2'))

This will add categories automatically::

    >>> storage.list_categories()
    [u'two', u'cat1', u'cat2']

And get for each word its category::

    >>> storage.word_categories('coucou', 'fr')
    [u'cat1', u'cat2']

This classification is updatable::

    >>> storage.add_word('coucou', 'fr', ('cat1',))
    >>> storage.word_categories('coucou', 'fr')
    [u'cat1']

We can count words:

    >>> storage.word_count()
    2

We can also remove words::

    >>> storage.del_word('coucou')
    >>> list(storage.list_words())
    [(u'hello', u'en')]


Let's remove the test db::

    >>> import os
    >>> os.remove(db_file)