Update rest of documentation for changes to EvalFactor and memorize_passes_needed().

chrish42 · chrish42 · commit 55330c23693e · 2015-05-14T20:25:47.000-04:00
diff --git a/doc/expert-model-specification.rst b/doc/expert-model-specification.rst
@@ -78,12 +78,11 @@ things are.)
    from patsy import (ModelDesc, EvalEnvironment, Term, EvalFactor,
                       LookupFactor, demo_data, dmatrix)
    data = demo_data("a", "x")
-   env = EvalEnvironment.capture()
 
    # LookupFactor takes a dictionary key:
    a_lookup = LookupFactor("a")
    # EvalFactor takes arbitrary Python code:
-   x_transform = EvalFactor("np.log(x ** 2)", env)
+   x_transform = EvalFactor("np.log(x ** 2)")
    # First argument is empty list for dmatrix; we would need to put
    # something there if we were calling dmatrices.
    desc = ModelDesc([],
@@ -157,7 +156,7 @@ The full interface looks like this:
        :term:`hashable`. These methods will determine which factors
        Patsy considers equal for purposes of redundancy elimination.
 
-    .. method:: memorize_passes_needed(state)
+    .. method:: memorize_passes_needed(state, eval_env)
 
        Return the number of passes through the data that this factor
        will need in order to set up any :ref:`stateful-transforms`.
@@ -171,6 +170,9 @@ The full interface looks like this:
        will be passed back in to all memorization and evaluation
        methods.
 
+       `eval_env` is an :class:`EvalEnvironment` object, describing
+       the Python environment where the factor is being evaluated.
+
     .. method:: memorize_chunk(state, which_pass, data)
 
        Called repeatedly with each 'chunk' of data produced by the
diff --git a/doc/formulas.rst b/doc/formulas.rst
@@ -58,13 +58,12 @@ To make this more concrete, here's how you could manually construct
 the same objects that Patsy will construct if given the above
 formula::
 
-  from patsy import EvalEnvironment, ModelDesc
-  env = EvalEnvironment.capture()
-  ModelDesc([Term([EvalFactor("y", env)])],
+  from patsy import ModelDesc
+  ModelDesc([Term([EvalFactor("y")])],
             [Term([]),
-             Term([EvalFactor("a", env)]),
-             Term([EvalFactor("a", env), EvalFactor("b", env)]),
-             Term([EvalFactor("np.log(x)", env)])])
+             Term([EvalFactor("a")]),
+             Term([EvalFactor("a"), EvalFactor("b")]),
+             Term([EvalFactor("np.log(x)")])])
 
 Compare to what you get from parsing the above formula::
 
