Merge branch 'tlz2' into tlz

eriknw · eriknw · commit 0ff1118fe6df · 2016-11-03T10:52:07.000-05:00
oops, I pushed the wrong thing to github.
diff --git a/AUTHORS.md b/AUTHORS.md
@@ -29,3 +29,5 @@ Joe Jevnik                                      [@llllllllll](https://github.com
 Rory Kirchner                                      [@roryk](https://github.com/roryk)
 
 [Steven Cutting](http://steven-cutting.github.io) [@steven_cutting](https://github.com/steven-cutting)
+
+Aric Coady                                      [@coady](https://github.com/coady)
diff --git a/MANIFEST.in b/MANIFEST.in
@@ -1 +1,3 @@
+include LICENSE.txt
+
 include toolz/tests/*.py
diff --git a/README.rst b/README.rst
@@ -5,7 +5,7 @@ Toolz
 
 A set of utility functions for iterators, functions, and dictionaries.
 
-See the PyToolz documentation at http://toolz.readthedocs.org
+See the PyToolz documentation at https://toolz.readthedocs.io
 
 LICENSE
 -------
@@ -21,12 +21,6 @@ Install
 
     pip install toolz
 
-or
-
-::
-
-    easy_install toolz
-
 Structure and Heritage
 ----------------------
 
@@ -54,7 +48,7 @@ These functions come from the legacy of functional languages for list
 processing. They interoperate well to accomplish common complex tasks.
 
 Read our `API
-Documentation <http://toolz.readthedocs.org/en/latest/api.html>`__ for
+Documentation <https://toolz.readthedocs.io/en/latest/api.html>`__ for
 more details.
 
 Example
diff --git a/conda.recipe/meta.yaml b/conda.recipe/meta.yaml
@@ -22,5 +22,5 @@ test:
       - py.test -x --doctest-modules --pyargs toolz
 
 about:
-    home: http://toolz.readthedocs.org/
+    home: https://toolz.readthedocs.io/
     license: BSD
diff --git a/doc/source/control.rst b/doc/source/control.rst
@@ -8,9 +8,8 @@ need for simultaneous thought is restricted to only a few elements at a time.
 
 All modern languages provide mechanisms to build data into data structures and
 to build functions out of other functions.  The third element of programming,
-besides data and functions is control flow.  Control flow is the third
-element of programming, after data and functions.  Building complex control
-flow out of simple control flow presents deeper challenges.
+besides data and functions, is control flow.  Building complex control flow
+out of simple control flow presents deeper challenges.
 
 
 What?
diff --git a/doc/source/streaming-analytics.rst b/doc/source/streaming-analytics.rst
@@ -34,7 +34,7 @@ These functions correspond to the SQL commands ``SELECT`` and ``WHERE``.
 .. code::
 
    >>> from toolz.curried import pipe, map, filter, get
-   >>> pipe(accounts, filter(lambda (id, name, balance, gender): balance > 150),
+   >>> pipe(accounts, filter(lambda acc: acc[2] > 150),
    ...                map(get([1, 2])),
    ...                list)
 
@@ -150,9 +150,9 @@ that adds up the balances for each group:
 
 .. code::
 
-   >>> binop = lambda total, (id, name, bal, gend): total + bal
+   >>> binop = lambda total, account: total + account[2]
 
-   >>> reduceby(get(3), binop, accounts)
+   >>> reduceby(get(3), binop, accounts, 0)
    {'F': 400, 'M': 400}
 
 
@@ -307,9 +307,9 @@ composition, ...) users interested in data analytics might be better served by
 using projects specific to data analytics like Pandas_ or SQLAlchemy.
 
 
-.. _groupby: http://toolz.readthedocs.org/en/latest/api.html#toolz.itertoolz.groupby
-.. _join: http://toolz.readthedocs.org/en/latest/api.html#toolz.itertoolz.join
-.. _reduceby: http://toolz.readthedocs.org/en/latest/api.html#toolz.itertoolz.reduceby
-.. _valmap: http://toolz.readthedocs.org/en/latest/api.html#toolz.itertoolz.valmap
+.. _groupby: https://toolz.readthedocs.io/en/latest/api.html#toolz.itertoolz.groupby
+.. _join: https://toolz.readthedocs.io/en/latest/api.html#toolz.itertoolz.join
+.. _reduceby: https://toolz.readthedocs.io/en/latest/api.html#toolz.itertoolz.reduceby
+.. _valmap: https://toolz.readthedocs.io/en/latest/api.html#toolz.dicttoolz.valmap
 .. _Pandas: http://pandas.pydata.org/pandas-docs/stable/groupby.html
-.. _curried: http://toolz.readthedocs.org/en/latest/curry.html
+.. _curried: https://toolz.readthedocs.io/en/latest/curry.html
diff --git a/setup.py b/setup.py
@@ -20,4 +20,16 @@
       package_data={'toolz': ['tests/*.py']},
       long_description=(open('README.rst').read() if exists('README.rst')
                         else ''),
-      zip_safe=False)
+      zip_safe=False,
+      classifiers=[
+          "Development Status :: 5 - Production/Stable",
+          "License :: OSI Approved :: BSD License",
+          "Programming Language :: Python",
+          "Programming Language :: Python :: 2.6",
+          "Programming Language :: Python :: 2.7",
+          "Programming Language :: Python :: 3",
+          "Programming Language :: Python :: 3.3",
+          "Programming Language :: Python :: 3.4",
+          "Programming Language :: Python :: 3.5",
+          "Programming Language :: Python :: Implementation :: CPython",
+          "Programming Language :: Python :: Implementation :: PyPy"])
diff --git a/tlz/_build_tlz.py b/tlz/_build_tlz.py
@@ -1,12 +1,9 @@
 import sys
 import types
-#try:
-#    from importlib import import_module
-#except ImportError:
-#    raise ImportError('"tlz" package is not available in Python 2.6')
 import toolz
 from toolz.compatibility import import_module
 
+
 class TlzLoader(object):
     """ Finds and loads ``tlz`` modules when added to sys.meta_path"""
     def __init__(self):
diff --git a/toolz/_signatures.py b/toolz/_signatures.py
@@ -17,7 +17,7 @@
 import itertools
 import operator
 
-from .compatibility import PY3
+from .compatibility import PY3, import_module
 from .functoolz import (is_partial_args, is_arity, has_varargs,
                         has_keywords, num_required_args)
 
@@ -706,10 +706,7 @@ def expand_sig(sig):
 def create_signature_registry(module_info=module_info, signatures=signatures):
     for module, info in module_info.items():
         if isinstance(module, str):
-            modnames = module.split('.')
-            module = __import__(module)
-            for attr in modnames[1:]:
-                module = getattr(module, attr)
+            module = import_module(module)
         for name, sigs in info.items():
             if hasattr(module, name):
                 new_sigs = tuple(expand_sig(sig) for sig in sigs)
diff --git a/toolz/compatibility.py b/toolz/compatibility.py
@@ -7,7 +7,7 @@
 
 __all__ = ('map', 'filter', 'range', 'zip', 'reduce', 'zip_longest',
            'iteritems', 'iterkeys', 'itervalues', 'filterfalse',
-           'PY3', 'PY34', 'PYPY')
+           'PY3', 'PY34', 'PYPY', 'import_module')
 
 if PY3:
     map = map
diff --git a/toolz/functoolz.py b/toolz/functoolz.py
@@ -4,7 +4,7 @@
 from operator import attrgetter
 from textwrap import dedent
 
-from .compatibility import PY3, PY33, PY34, PYPY
+from .compatibility import PY3, PY33, PY34, PYPY, import_module
 from .utils import no_default
 
 
@@ -169,7 +169,7 @@ class curry(object):
 
     See Also:
         toolz.curried - namespace of curried functions
-                        http://toolz.readthedocs.org/en/latest/curry.html
+                        https://toolz.readthedocs.io/en/latest/curry.html
     """
     def __init__(self, *args, **kwargs):
         if not args:
@@ -321,17 +321,33 @@ def __get__(self, instance, owner):
             return self
         return curry(self, instance)
 
-    # pickle protocol because functools.partial objects can't be pickled
-    def __getstate__(self):
-        # dictoolz.keyfilter, I miss you!
+    def __reduce__(self):
+        func = self.func
+        modname = getattr(func, '__module__', None)
+        funcname = getattr(func, '__name__', None)
+        if modname and funcname:
+            module = import_module(modname)
+            obj = getattr(module, funcname, None)
+            if obj is self:
+                return funcname
+            elif isinstance(obj, curry) and obj.func is func:
+                func = '%s.%s' % (modname, funcname)
+
+        # functools.partial objects can't be pickled
         userdict = tuple((k, v) for k, v in self.__dict__.items()
                          if k != '_partial')
-        return self.func, self.args, self.keywords, userdict
-
-    def __setstate__(self, state):
-        func, args, kwargs, userdict = state
-        self.__init__(func, *args, **(kwargs or {}))
-        self.__dict__.update(userdict)
+        state = (type(self), func, self.args, self.keywords, userdict)
+        return (_restore_curry, state)
+
+
+def _restore_curry(cls, func, args, kwargs, userdict):
+    if isinstance(func, str):
+        modname, funcname = func.rsplit('.', 1)
+        module = import_module(modname)
+        func = getattr(module, funcname).func
+    obj = cls(func, *args, **(kwargs or {}))
+    obj.__dict__.update(userdict)
+    return obj
 
 
 @curry
diff --git a/toolz/itertoolz.py b/toolz/itertoolz.py
@@ -214,7 +214,7 @@ def _merge_sorted_binary_key(seqs, key):
         yield val1
 
 
-def interleave(seqs, pass_exceptions=()):
+def interleave(seqs):
     """ Interleave a sequence of sequences
 
     >>> list(interleave([[1, 2], [3, 4]]))
@@ -227,16 +227,15 @@ def interleave(seqs, pass_exceptions=()):
 
     Returns a lazy iterator
     """
-    iters = map(iter, seqs)
-    while iters:
-        newiters = []
-        for itr in iters:
-            try:
+    iters = itertools.cycle(map(iter, seqs))
+    while True:
+        try:
+            for itr in iters:
                 yield next(itr)
-                newiters.append(itr)
-            except (StopIteration,) + tuple(pass_exceptions):
-                pass
-        iters = newiters
+            return
+        except StopIteration:
+            predicate = partial(operator.is_not, itr)
+            iters = itertools.cycle(itertools.takewhile(predicate, iters))
 
 
 def unique(seq, key=None):
@@ -666,16 +665,8 @@ def sliding_window(n, seq):
     >>> list(map(mean, sliding_window(2, [1, 2, 3, 4])))
     [1.5, 2.5, 3.5]
     """
-    it = iter(seq)
-    # An efficient FIFO data structure with maximum length
-    d = collections.deque(itertools.islice(it, n), n)
-    if len(d) != n:
-        raise StopIteration()
-    d_append = d.append
-    for item in it:
-        yield tuple(d)
-        d_append(item)
-    yield tuple(d)
+    return zip(*(collections.deque(itertools.islice(it, i), 0) or it
+               for i, it in enumerate(itertools.tee(seq, n))))
 
 
 no_pad = '__no__pad__'
@@ -722,7 +713,10 @@ def partition_all(n, seq):
     """
     args = [iter(seq)] * n
     it = zip_longest(*args, fillvalue=no_pad)
-    prev = next(it)
+    try:
+        prev = next(it)
+    except StopIteration:
+        return
     for item in it:
         yield prev
         prev = item
diff --git a/toolz/tests/test_itertoolz.py b/toolz/tests/test_itertoolz.py
@@ -335,11 +335,10 @@ def test_pluck():
 
     data = [{'id': 1, 'name': 'cheese'}, {'id': 2, 'name': 'pies', 'price': 1}]
     assert list(pluck('id', data)) == [1, 2]
-    assert list(pluck('price', data, None)) == [None, 1]
+    assert list(pluck('price', data, 0)) == [0, 1]
     assert list(pluck(['id', 'name'], data)) == [(1, 'cheese'), (2, 'pies')]
     assert list(pluck(['name'], data)) == [('cheese',), ('pies',)]
-    assert list(pluck(['price', 'other'], data, None)) == [(None, None),
-                                                           (1, None)]
+    assert list(pluck(['price', 'other'], data, 0)) == [(0, 0), (1, 0)]
 
     assert raises(IndexError, lambda: list(pluck(1, [[0]])))
     assert raises(KeyError, lambda: list(pluck('name', [{'id': 1}])))
diff --git a/toolz/tests/test_serialization.py b/toolz/tests/test_serialization.py
@@ -40,3 +40,18 @@ def test_instanceproperty():
     assert p2.__get__(None) is None
     assert p2.__get__(0) is False
     assert p2.__get__(1) is True
+
+
+def f(x, y):
+    return x, y
+
+
+def test_flip():
+    flip = pickle.loads(pickle.dumps(toolz.functoolz.flip))
+    assert flip is toolz.functoolz.flip
+    g1 = flip(f)
+    g2 = pickle.loads(pickle.dumps(g1))
+    assert g1(1, 2) == g2(1, 2) == f(2, 1)
+    g1 = flip(f)(1)
+    g2 = pickle.loads(pickle.dumps(g1))
+    assert g1(2) == g2(2) == f(2, 1)

Original file line number	Diff line number	Diff line change
`@@ -1 +1,3 @@`
	`1`	`+include LICENSE.txt`
	`2`	`+`
`1`	`3`	`include toolz/tests/*.py`