pipeline/nltk/decorators.py

"""

Decorator module by Michele Simionato <[email protected]>

Copyright Michele Simionato, distributed under the terms of the BSD License (see below).

http://www.phyast.pitt.edu/~micheles/python/documentation.html



Included in NLTK for its support of a nice memoization decorator.

"""

__docformat__ = "restructuredtext en"

## The basic trick is to generate the source code for the decorated function
## with the right signature and to evaluate it.
## Uncomment the statement 'print >> sys.stderr, func_src'  in _decorator
## to understand what is going on.

__all__ = ["decorator", "new_wrapper", "getinfo"]

import sys

# Hack to keep NLTK's "tokenize" module from colliding with the "tokenize" in
# the Python standard library.
OLD_SYS_PATH = sys.path[:]
sys.path = [p for p in sys.path if p and "nltk" not in str(p)]
import inspect

sys.path = OLD_SYS_PATH


def __legacysignature(signature):
    """

    For retrocompatibility reasons, we don't use a standard Signature.

    Instead, we use the string generated by this method.

    Basically, from a Signature we create a string and remove the default values.

    """
    listsignature = str(signature)[1:-1].split(",")
    for counter, param in enumerate(listsignature):
        if param.count("=") > 0:
            listsignature[counter] = param[0 : param.index("=")].strip()
        else:
            listsignature[counter] = param.strip()
    return ", ".join(listsignature)


def getinfo(func):
    """

    Returns an info dictionary containing:

    - name (the name of the function : str)

    - argnames (the names of the arguments : list)

    - defaults (the values of the default arguments : tuple)

    - signature (the signature : str)

    - fullsignature (the full signature : Signature)

    - doc (the docstring : str)

    - module (the module name : str)

    - dict (the function __dict__ : str)



    >>> def f(self, x=1, y=2, *args, **kw): pass



    >>> info = getinfo(f)



    >>> info["name"]

    'f'

    >>> info["argnames"]

    ['self', 'x', 'y', 'args', 'kw']



    >>> info["defaults"]

    (1, 2)



    >>> info["signature"]

    'self, x, y, *args, **kw'



    >>> info["fullsignature"]

    <Signature (self, x=1, y=2, *args, **kw)>

    """
    assert inspect.ismethod(func) or inspect.isfunction(func)
    argspec = inspect.getfullargspec(func)
    regargs, varargs, varkwargs = argspec[:3]
    argnames = list(regargs)
    if varargs:
        argnames.append(varargs)
    if varkwargs:
        argnames.append(varkwargs)
    fullsignature = inspect.signature(func)
    # Convert Signature to str
    signature = __legacysignature(fullsignature)

    # pypy compatibility
    if hasattr(func, "__closure__"):
        _closure = func.__closure__
        _globals = func.__globals__
    else:
        _closure = func.func_closure
        _globals = func.func_globals

    return dict(
        name=func.__name__,
        argnames=argnames,
        signature=signature,
        fullsignature=fullsignature,
        defaults=func.__defaults__,
        doc=func.__doc__,
        module=func.__module__,
        dict=func.__dict__,
        globals=_globals,
        closure=_closure,
    )


def update_wrapper(wrapper, model, infodict=None):
    "akin to functools.update_wrapper"
    infodict = infodict or getinfo(model)
    wrapper.__name__ = infodict["name"]
    wrapper.__doc__ = infodict["doc"]
    wrapper.__module__ = infodict["module"]
    wrapper.__dict__.update(infodict["dict"])
    wrapper.__defaults__ = infodict["defaults"]
    wrapper.undecorated = model
    return wrapper


def new_wrapper(wrapper, model):
    """

    An improvement over functools.update_wrapper. The wrapper is a generic

    callable object. It works by generating a copy of the wrapper with the

    right signature and by updating the copy, not the original.

    Moreovoer, 'model' can be a dictionary with keys 'name', 'doc', 'module',

    'dict', 'defaults'.

    """
    if isinstance(model, dict):
        infodict = model
    else:  # assume model is a function
        infodict = getinfo(model)
    assert (
        not "_wrapper_" in infodict["argnames"]
    ), '"_wrapper_" is a reserved argument name!'
    src = "lambda %(signature)s: _wrapper_(%(signature)s)" % infodict
    funcopy = eval(src, dict(_wrapper_=wrapper))
    return update_wrapper(funcopy, model, infodict)


# helper used in decorator_factory
def __call__(self, func):
    return new_wrapper(lambda *a, **k: self.call(func, *a, **k), func)


def decorator_factory(cls):
    """

    Take a class with a ``.caller`` method and return a callable decorator

    object. It works by adding a suitable __call__ method to the class;

    it raises a TypeError if the class already has a nontrivial __call__

    method.

    """
    attrs = set(dir(cls))
    if "__call__" in attrs:
        raise TypeError(
            "You cannot decorate a class with a nontrivial " "__call__ method"
        )
    if "call" not in attrs:
        raise TypeError("You cannot decorate a class without a " ".call method")
    cls.__call__ = __call__
    return cls


def decorator(caller):
    """

    General purpose decorator factory: takes a caller function as

    input and returns a decorator with the same attributes.

    A caller function is any function like this::



     def caller(func, *args, **kw):

         # do something

         return func(*args, **kw)



    Here is an example of usage:



    >>> @decorator

    ... def chatty(f, *args, **kw):

    ...     print("Calling %r" % f.__name__)

    ...     return f(*args, **kw)



    >>> chatty.__name__

    'chatty'



    >>> @chatty

    ... def f(): pass

    ...

    >>> f()

    Calling 'f'



    decorator can also take in input a class with a .caller method; in this

    case it converts the class into a factory of callable decorator objects.

    See the documentation for an example.

    """
    if inspect.isclass(caller):
        return decorator_factory(caller)

    def _decorator(func):  # the real meat is here
        infodict = getinfo(func)
        argnames = infodict["argnames"]
        assert not (
            "_call_" in argnames or "_func_" in argnames
        ), "You cannot use _call_ or _func_ as argument names!"
        src = "lambda %(signature)s: _call_(_func_, %(signature)s)" % infodict
        # import sys; print >> sys.stderr, src # for debugging purposes
        dec_func = eval(src, dict(_func_=func, _call_=caller))
        return update_wrapper(dec_func, func, infodict)

    return update_wrapper(_decorator, caller)


def getattr_(obj, name, default_thunk):
    "Similar to .setdefault in dictionaries."
    try:
        return getattr(obj, name)
    except AttributeError:
        default = default_thunk()
        setattr(obj, name, default)
        return default


@decorator
def memoize(func, *args):
    dic = getattr_(func, "memoize_dic", dict)
    # memoize_dic is created at the first call
    if args in dic:
        return dic[args]
    result = func(*args)
    dic[args] = result
    return result


##########################     LEGALESE    ###############################

##   Redistributions of source code must retain the above copyright
##   notice, this list of conditions and the following disclaimer.
##   Redistributions in bytecode form must reproduce the above copyright
##   notice, this list of conditions and the following disclaimer in
##   the documentation and/or other materials provided with the
##   distribution.

##   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
##   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
##   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
##   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
##   HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
##   INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
##   BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
##   OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
##   ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
##   TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
##   USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
##   DAMAGE.