def __init__(self,
parselet,
selector_handler=None,
strict=False,
debug=False):
"""
Take a parselet and optional selector_handler
and build an abstract representation of the Parsley extraction
logic.
Two helper class methods can be used to instantiate a Parselet
from JSON rules: :meth:`.from_jsonstring`, :meth:`.from_jsonfile`.
:param dict parselet: Parsley script as a Python dict object
:param boolean strict: Set to *True* is you want to
enforce that missing required keys raise an Exception; default is False
(i.e. lenient/non-strict mode)
:param selector_handler: an instance of :class:`selectors.SelectorHandler`
optional selector handler instance;
defaults to an instance of :class:`selectors.DefaultSelectorHandler`
:raises: :class:`.InvalidKeySyntax`
Example:
>>> import parslepy
>>> rules = {
... "heading": "h1#main",
... "news(li.newsitem)": [{
... "title": ".",
... "url": "a/@href"
... }],
... }
>>> p = parslepy.Parselet(rules)
>>> type(p)
<class 'parslepy.base.Parselet'>
Use :meth:`~base.Parselet.extract` or :meth:`~base.Parselet.parse`
to get extracted content from documents.
"""
if debug:
self.DEBUG = True
if strict:
self.STRICT_MODE = True
self.parselet = parselet
if not selector_handler:
self.selector_handler = DefaultSelectorHandler(debug=self.DEBUG)
elif not (isinstance(selector_handler, SelectorHandler)):
raise ValueError("You must provide a SelectorHandler instance")
else:
self.selector_handler = selector_handler
self.compile()
