collective.elephantvocabulary 0.2.4

type of zope vocabularies that dont "forget", like elephants

Introduction

Like elephants don’t forget anything, so does not collective.elephantvocabulary. It provides a wrapper around for existing zope.schema vocabularies and make them not forget anything.

Example usecase would be a vocabulary (source) of users which from certain point in time wants to hide / deactivate some users for form or listing. But at the same time you want keep old references to user term working. This is where collective.elephantvocabulary comes into the picture. With it you wrap existing vocabulary of users and provide set of hidden list of users (term values).

Contents

• Introduction

• Usage

• Credits

• Todo

• History

  • 0.2.4 (2013-10-26)

  • 0.2.3 (2013-10-25)

  • 0.2.2 (2010-10-12)

  • 0.2.1 (2010-10-11)

  • 0.2 (2010-10-11)

  • 0.1.3 (2010-10-11)

  • 0.1.2 (2010-10-08)

  • 0.1.1 (2010-10-08)

  • 0.1 (2010-10-08)

Usage

Some example content and vocabularies

>>> context = layer.context
>>> example_vocab = layer.example_vocab
>>> example_source = layer.example_source
>>> [i.value for i in example_vocab]
[1, 2, 3, 4]

Below is out wraper method we use to make our existing vocab more elephant-like.

>>> from collective.elephantvocabulary import wrap_vocabulary

In first exampe we pass to our wrap_vocabulary a vocabulary of [1, 2, 3, 4] and we set terms 2 and 3 to hidden. wrap_vocabulary returns VocabularyFactory which needs to be called with context (you could also register it with as utility).

>>> wrapped_vocab_factory = wrap_vocabulary(example_vocab, hidden_terms=[2, 3])
>>> print wrapped_vocab_factory
<collective.elephantvocabulary.vocabulary.VocabularyFactory object at ...>

>>> wrapped_vocab = wrapped_vocab_factory(context)
>>> [i.value for i in wrapped_vocab]
[1, 4]

>>> len(wrapped_vocab) == len(example_vocab)
True

>>> 2 in wrapped_vocab
True

>>> 5 in wrapped_vocab
False

>>> wrapped_vocab.getTerm(3).value
3

Similar we can limit items shown only to the set we want (via visible_terms)

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 visible_terms=[2, 3])(context)
>>> [i.value for i in wrapped_vocab]
[2, 3]

>>> len(wrapped_vocab) == len(example_vocab)
True

>>> 2 in wrapped_vocab
True

>>> 5 in wrapped_vocab
False

>>> wrapped_vocab.getTerm(1).value
1

Above we see what collective.elephantvocabulary is all about. When listing vocabulary hidden terms are not listed. But when item is requested with its term value then term is also returned. Also length of vocabulary is unchanged. It still shows original lenght of vocabulary.

We can also call vocabulary by name it was register with ZCA machinery..

>>> wrapped_vocab = wrap_vocabulary('example-vocab',
...                                  hidden_terms=[2, 3])(context)
>>> [i.value for i in wrapped_vocab]
[1, 4]

hidden_terms or visible_terms parameter (second argument we pass to wrap_vocabulary) can also be callable which expects 2 parameters, context and original vocabulary.

>>> def hidden_terms(context, vocab):
...     return [1, 4]

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 hidden_terms=hidden_terms)(context)
>>> [i.value for i in wrapped_vocab]
[2, 3]

>>> def visible_terms(context, vocab):
...     return [1, 4]

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 visible_terms=hidden_terms)(context)
>>> [i.value for i in wrapped_vocab]
[1, 4]

collective.elephantvocabulary also works with sources.

>>> [i.value for i in example_source]
[1, 2, 3, 4]

>>> [i.value for i in example_source.search()]
[1, 2]

>>> wrapped_source = wrap_vocabulary(example_source, hidden_terms=[1, 4])(context)
>>> [i.value for i in wrapped_source.search()]
[2]

>>> wrapped_source = wrap_vocabulary(example_source, visible_terms=[1, 4])(context)
>>> [i.value for i in wrapped_source.search()]
[1]

If vocabulary already provides set of hidden terms they are passed to wrapped vocabulary.

>>> example_vocab.hidden_terms = [1, 2]
>>> wrapped_vocab = wrap_vocabulary(example_vocab)(context)
>>> [i.value for i in wrapped_vocab]
[3, 4]

>>> del example_vocab.hidden_terms

>>> example_vocab.visible_terms= [1, 2]
>>> wrapped_vocab = wrap_vocabulary(example_vocab)(context)
>>> [i.value for i in wrapped_vocab]
[1, 2]

>>> del example_vocab.visible_terms

Vocabulary will ass to the list of passed visible_terms or hidden_terms.

>>> example_vocab.hidden_terms = [1, 2]
>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 hidden_terms=[2, 3])(context)
>>> [i.value for i in wrapped_vocab]
[4]

>>> del example_vocab.hidden_terms

>>> example_vocab.visible_terms= [1]
>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 visible_terms=[1, 2, 3])(context)
>>> [i.value for i in wrapped_vocab]
[1, 2, 3]

>>> del example_vocab.visible_terms

hidden_terms and visible_terms can also work together.

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...                                 visible_terms=[1, 2, 3],
...                                 hidden_terms=[2])(context)
>>> [i.value for i in wrapped_vocab]
[1, 3]

We could also store hidden_terms and visible_terms in plone.registry. Instead of creating our own methods which reads from plone.registry collective.elephantvocabulary provides helper parameters: hidden_terms_from_registry and visible_terms_from_registry.

>>> from zope.component import getUtility
>>> from plone.registry import field
>>> from plone.registry import Record
>>> from plone.registry.interfaces import IRegistry

>>> example_registry_record = Record(
...         field.List(title=u"Test", min_length=0, max_length=10,
...                    value_type=field.Int(title=u"Value")))
>>> example_registry_record.value = [1, 2]

>>> registry = getUtility(IRegistry)
>>> registry.records['example.hidden_terms'] = example_registry_record
>>> registry.records['example.visible_terms'] = example_registry_record

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...         visible_terms_from_registry='example.visible_terms')(context)
>>> [i.value for i in wrapped_vocab]
[1, 2]

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...         hidden_terms_from_registry='example.hidden_terms')(context)
>>> [i.value for i in wrapped_vocab]
[3, 4]

Or we can use them in combination.

>>> example_registry_record2 = Record(
...         field.List(title=u"Test", min_length=0, max_length=10,
...                    value_type=field.Int(title=u"Value")))
>>> example_registry_record2.value = [1, 2, 3]
>>> registry.records['example.visible_terms'] = example_registry_record2

>>> wrapped_vocab = wrap_vocabulary(example_vocab,
...         visible_terms_from_registry='example.visible_terms',
...         hidden_terms_from_registry='example.hidden_terms')(context)
>>> [i.value for i in wrapped_vocab]
[3]

And if we don’t pass anything to wrap_vocabulary then it should ack as normal vocabulary.

>>> wrapped_vocab5 = wrap_vocabulary(example_vocab)(context)
>>> [i.value for i in wrapped_vocab5]
[1, 2, 3, 4]

Credits

Generously sponsored by 4teamwork.

• Rok Garbas, author

Todo

• provide test / documentation for custom wrapper class

• coverage should show 100%, but its failing on method and import lines, weird.

History

0.2.4 (2013-10-26)

• fixing 0.2.3 release [garbas]

0.2.3 (2013-10-25)

• Don’t memoize the persistent plone.registry utility. This avoids issues with object access across different ZODB connections (see issue #2). [lgraf]

• Avoid nesting internal term lists (see issue #3). [lgraf]

• Avoid indefinite growth of internal term lists (see issue #4). [lgraf]

• Split the README.rst up into several files. Put the testing part in tests.rst in the main directory so this test file can also be found when we are distributed on PyPI. [maurits]

0.2.2 (2010-10-12)

• support for other type of vocabs (IVocabulary, IIterableSource) [garbas]

• BUG(Fixed): registry should be not be loaded at __init__ time [garbas]

0.2.1 (2010-10-11)

• new parameters visible_terms_from_registry and hidden_terms_from_registry which reads values directly from plone.registry. [garbas]

0.2 (2010-10-11)

• visible_terms parameter added to wrap_vocabulary, by default visible_terms and hidden_terms work “together” (via WrapperBase) [garbas]

0.1.3 (2010-10-11)

• marking wrapper vocabularies with IElephantVocabulary interface [garbas]

0.1.2 (2010-10-08)

• misspelled dependency, feeling silly [garbas]

0.1.1 (2010-10-08)

• add dependencies from where we import (using mr.igor) [garbas]

• add link to zope.schema which was breaking formating for rst formatting [garbas]

• initial release was broken (missing README.rst) [garbas]

0.1 (2010-10-08)

• initial release [garbas]