I am done

rl/Lib/site-packages/sympy/codegen/__init__.py

@@ -0,0 +1,24 @@
+""" The ``sympy.codegen`` module contains classes and functions for building
+abstract syntax trees of algorithms. These trees may then be printed by the
+code-printers in ``sympy.printing``.
+
+There are several submodules available:
+- ``sympy.codegen.ast``: AST nodes useful across multiple languages.
+- ``sympy.codegen.cnodes``: AST nodes useful for the C family of languages.
+- ``sympy.codegen.fnodes``: AST nodes useful for Fortran.
+- ``sympy.codegen.cfunctions``: functions specific to C (C99 math functions)
+- ``sympy.codegen.ffunctions``: functions specific to Fortran (e.g. ``kind``).
+
+
+
+"""
+from .ast import (
+    Assignment, aug_assign, CodeBlock, For, Attribute, Variable, Declaration,
+    While, Scope, Print, FunctionPrototype, FunctionDefinition, FunctionCall
+)
+
+__all__ = [
+    'Assignment', 'aug_assign', 'CodeBlock', 'For', 'Attribute', 'Variable',
+    'Declaration', 'While', 'Scope', 'Print', 'FunctionPrototype',
+    'FunctionDefinition', 'FunctionCall',
+]

rl/Lib/site-packages/sympy/codegen/abstract_nodes.py

@@ -0,0 +1,18 @@
+"""This module provides containers for python objects that are valid
+printing targets but are not a subclass of SymPy's Printable.
+"""
+
+
+from sympy.core.containers import Tuple
+
+
+class List(Tuple):
+    """Represents a (frozen) (Python) list (for code printing purposes)."""
+    def __eq__(self, other):
+        if isinstance(other, list):
+            return self == List(*other)
+        else:
+            return self.args == other
+
+    def __hash__(self):
+        return super().__hash__()

rl/Lib/site-packages/sympy/codegen/algorithms.py

@@ -0,0 +1,180 @@
+from sympy.core.containers import Tuple
+from sympy.core.numbers import oo
+from sympy.core.relational import (Gt, Lt)
+from sympy.core.symbol import (Dummy, Symbol)
+from sympy.functions.elementary.complexes import Abs
+from sympy.functions.elementary.miscellaneous import Min, Max
+from sympy.logic.boolalg import And
+from sympy.codegen.ast import (
+    Assignment, AddAugmentedAssignment, break_, CodeBlock, Declaration, FunctionDefinition,
+    Print, Return, Scope, While, Variable, Pointer, real
+)
+from sympy.codegen.cfunctions import isnan
+
+""" This module collects functions for constructing ASTs representing algorithms. """
+
+def newtons_method(expr, wrt, atol=1e-12, delta=None, *, rtol=4e-16, debug=False,
+                   itermax=None, counter=None, delta_fn=lambda e, x: -e/e.diff(x),
+                   cse=False, handle_nan=None,
+                   bounds=None):
+    """ Generates an AST for Newton-Raphson method (a root-finding algorithm).
+
+    Explanation
+    ===========
+
+    Returns an abstract syntax tree (AST) based on ``sympy.codegen.ast`` for Netwon's
+    method of root-finding.
+
+    Parameters
+    ==========
+
+    expr : expression
+    wrt : Symbol
+        With respect to, i.e. what is the variable.
+    atol : number or expression
+        Absolute tolerance (stopping criterion)
+    rtol : number or expression
+        Relative tolerance (stopping criterion)
+    delta : Symbol
+        Will be a ``Dummy`` if ``None``.
+    debug : bool
+        Whether to print convergence information during iterations
+    itermax : number or expr
+        Maximum number of iterations.
+    counter : Symbol
+        Will be a ``Dummy`` if ``None``.
+    delta_fn: Callable[[Expr, Symbol], Expr]
+        computes the step, default is newtons method. For e.g. Halley's method
+        use delta_fn=lambda e, x: -2*e*e.diff(x)/(2*e.diff(x)**2 - e*e.diff(x, 2))
+    cse: bool
+        Perform common sub-expression elimination on delta expression
+    handle_nan: Token
+        How to handle occurrence of not-a-number (NaN).
+    bounds: Optional[tuple[Expr, Expr]]
+        Perform optimization within bounds
+
+    Examples
+    ========
+
+    >>> from sympy import symbols, cos
+    >>> from sympy.codegen.ast import Assignment
+    >>> from sympy.codegen.algorithms import newtons_method
+    >>> x, dx, atol = symbols('x dx atol')
+    >>> expr = cos(x) - x**3
+    >>> algo = newtons_method(expr, x, atol=atol, delta=dx)
+    >>> algo.has(Assignment(dx, -expr/expr.diff(x)))
+    True
+
+    References
+    ==========
+
+    .. [1] https://en.wikipedia.org/wiki/Newton%27s_method
+
+    """
+
+    if delta is None:
+        delta = Dummy()
+        Wrapper = Scope
+        name_d = 'delta'
+    else:
+        Wrapper = lambda x: x
+        name_d = delta.name
+
+    delta_expr = delta_fn(expr, wrt)
+    if cse:
+        from sympy.simplify.cse_main import cse
+        cses, (red,) = cse([delta_expr.factor()])
+        whl_bdy = [Assignment(dum, sub_e) for dum, sub_e in cses]
+        whl_bdy += [Assignment(delta, red)]
+    else:
+        whl_bdy = [Assignment(delta, delta_expr)]
+    if handle_nan is not None:
+        whl_bdy += [While(isnan(delta), CodeBlock(handle_nan, break_))]
+    whl_bdy += [AddAugmentedAssignment(wrt, delta)]
+    if bounds is not None:
+        whl_bdy += [Assignment(wrt, Min(Max(wrt, bounds[0]), bounds[1]))]
+    if debug:
+        prnt = Print([wrt, delta], r"{}=%12.5g {}=%12.5g\n".format(wrt.name, name_d))
+        whl_bdy += [prnt]
+    req = Gt(Abs(delta), atol + rtol*Abs(wrt))
+    declars = [Declaration(Variable(delta, type=real, value=oo))]
+    if itermax is not None:
+        counter = counter or Dummy(integer=True)
+        v_counter = Variable.deduced(counter, 0)
+        declars.append(Declaration(v_counter))
+        whl_bdy.append(AddAugmentedAssignment(counter, 1))
+        req = And(req, Lt(counter, itermax))
+    whl = While(req, CodeBlock(*whl_bdy))
+    blck = declars
+    if debug:
+        blck.append(Print([wrt], r"{}=%12.5g\n".format(wrt.name)))
+    blck += [whl]
+    return Wrapper(CodeBlock(*blck))
+
+
+def _symbol_of(arg):
+    if isinstance(arg, Declaration):
+        arg = arg.variable.symbol
+    elif isinstance(arg, Variable):
+        arg = arg.symbol
+    return arg
+
+
+def newtons_method_function(expr, wrt, params=None, func_name="newton", attrs=Tuple(), *, delta=None, **kwargs):
+    """ Generates an AST for a function implementing the Newton-Raphson method.
+
+    Parameters
+    ==========
+
+    expr : expression
+    wrt : Symbol
+        With respect to, i.e. what is the variable
+    params : iterable of symbols
+        Symbols appearing in expr that are taken as constants during the iterations
+        (these will be accepted as parameters to the generated function).
+    func_name : str
+        Name of the generated function.
+    attrs : Tuple
+        Attribute instances passed as ``attrs`` to ``FunctionDefinition``.
+    \\*\\*kwargs :
+        Keyword arguments passed to :func:`sympy.codegen.algorithms.newtons_method`.
+
+    Examples
+    ========
+
+    >>> from sympy import symbols, cos
+    >>> from sympy.codegen.algorithms import newtons_method_function
+    >>> from sympy.codegen.pyutils import render_as_module
+    >>> x = symbols('x')
+    >>> expr = cos(x) - x**3
+    >>> func = newtons_method_function(expr, x)
+    >>> py_mod = render_as_module(func)  # source code as string
+    >>> namespace = {}
+    >>> exec(py_mod, namespace, namespace)
+    >>> res = eval('newton(0.5)', namespace)
+    >>> abs(res - 0.865474033102) < 1e-12
+    True
+
+    See Also
+    ========
+
+    sympy.codegen.algorithms.newtons_method
+
+    """
+    if params is None:
+        params = (wrt,)
+    pointer_subs = {p.symbol: Symbol('(*%s)' % p.symbol.name)
+                    for p in params if isinstance(p, Pointer)}
+    if delta is None:
+        delta = Symbol('d_' + wrt.name)
+        if expr.has(delta):
+            delta = None  # will use Dummy
+    algo = newtons_method(expr, wrt, delta=delta, **kwargs).xreplace(pointer_subs)
+    if isinstance(algo, Scope):
+        algo = algo.body
+    not_in_params = expr.free_symbols.difference({_symbol_of(p) for p in params})
+    if not_in_params:
+        raise ValueError("Missing symbols in params: %s" % ', '.join(map(str, not_in_params)))
+    declars = tuple(Variable(p, real) for p in params)
+    body = CodeBlock(algo, Return(wrt))
+    return FunctionDefinition(real, func_name, declars, body, attrs=attrs)

rl/Lib/site-packages/sympy/codegen/approximations.py

@@ -0,0 +1,187 @@
+import math
+from sympy.sets.sets import Interval
+from sympy.calculus.singularities import is_increasing, is_decreasing
+from sympy.codegen.rewriting import Optimization
+from sympy.core.function import UndefinedFunction
+
+"""
+This module collects classes useful for approimate rewriting of expressions.
+This can be beneficial when generating numeric code for which performance is
+of greater importance than precision (e.g. for preconditioners used in iterative
+methods).
+"""
+
+class SumApprox(Optimization):
+    """
+    Approximates sum by neglecting small terms.
+
+    Explanation
+    ===========
+
+    If terms are expressions which can be determined to be monotonic, then
+    bounds for those expressions are added.
+
+    Parameters
+    ==========
+
+    bounds : dict
+        Mapping expressions to length 2 tuple of bounds (low, high).
+    reltol : number
+        Threshold for when to ignore a term. Taken relative to the largest
+        lower bound among bounds.
+
+    Examples
+    ========
+
+    >>> from sympy import exp
+    >>> from sympy.abc import x, y, z
+    >>> from sympy.codegen.rewriting import optimize
+    >>> from sympy.codegen.approximations import SumApprox
+    >>> bounds = {x: (-1, 1), y: (1000, 2000), z: (-10, 3)}
+    >>> sum_approx3 = SumApprox(bounds, reltol=1e-3)
+    >>> sum_approx2 = SumApprox(bounds, reltol=1e-2)
+    >>> sum_approx1 = SumApprox(bounds, reltol=1e-1)
+    >>> expr = 3*(x + y + exp(z))
+    >>> optimize(expr, [sum_approx3])
+    3*(x + y + exp(z))
+    >>> optimize(expr, [sum_approx2])
+    3*y + 3*exp(z)
+    >>> optimize(expr, [sum_approx1])
+    3*y
+
+    """
+
+    def __init__(self, bounds, reltol, **kwargs):
+        super().__init__(**kwargs)
+        self.bounds = bounds
+        self.reltol = reltol
+
+    def __call__(self, expr):
+        return expr.factor().replace(self.query, lambda arg: self.value(arg))
+
+    def query(self, expr):
+        return expr.is_Add
+
+    def value(self, add):
+        for term in add.args:
+            if term.is_number or term in self.bounds or len(term.free_symbols) != 1:
+                continue
+            fs, = term.free_symbols
+            if fs not in self.bounds:
+                continue
+            intrvl = Interval(*self.bounds[fs])
+            if is_increasing(term, intrvl, fs):
+                self.bounds[term] = (
+                    term.subs({fs: self.bounds[fs][0]}),
+                    term.subs({fs: self.bounds[fs][1]})
+                )
+            elif is_decreasing(term, intrvl, fs):
+                self.bounds[term] = (
+                    term.subs({fs: self.bounds[fs][1]}),
+                    term.subs({fs: self.bounds[fs][0]})
+                )
+            else:
+                return add
+
+        if all(term.is_number or term in self.bounds for term in add.args):
+            bounds = [(term, term) if term.is_number else self.bounds[term] for term in add.args]
+            largest_abs_guarantee = 0
+            for lo, hi in bounds:
+                if lo <= 0 <= hi:
+                    continue
+                largest_abs_guarantee = max(largest_abs_guarantee,
+                                            min(abs(lo), abs(hi)))
+            new_terms = []
+            for term, (lo, hi) in zip(add.args, bounds):
+                if max(abs(lo), abs(hi)) >= largest_abs_guarantee*self.reltol:
+                    new_terms.append(term)
+            return add.func(*new_terms)
+        else:
+            return add
+
+
+class SeriesApprox(Optimization):
+    """ Approximates functions by expanding them as a series.
+
+    Parameters
+    ==========
+
+    bounds : dict
+        Mapping expressions to length 2 tuple of bounds (low, high).
+    reltol : number
+        Threshold for when to ignore a term. Taken relative to the largest
+        lower bound among bounds.
+    max_order : int
+        Largest order to include in series expansion
+    n_point_checks : int (even)
+        The validity of an expansion (with respect to reltol) is checked at
+        discrete points (linearly spaced over the bounds of the variable). The
+        number of points used in this numerical check is given by this number.
+
+    Examples
+    ========
+
+    >>> from sympy import sin, pi
+    >>> from sympy.abc import x, y
+    >>> from sympy.codegen.rewriting import optimize
+    >>> from sympy.codegen.approximations import SeriesApprox
+    >>> bounds = {x: (-.1, .1), y: (pi-1, pi+1)}
+    >>> series_approx2 = SeriesApprox(bounds, reltol=1e-2)
+    >>> series_approx3 = SeriesApprox(bounds, reltol=1e-3)
+    >>> series_approx8 = SeriesApprox(bounds, reltol=1e-8)
+    >>> expr = sin(x)*sin(y)
+    >>> optimize(expr, [series_approx2])
+    x*(-y + (y - pi)**3/6 + pi)
+    >>> optimize(expr, [series_approx3])
+    (-x**3/6 + x)*sin(y)
+    >>> optimize(expr, [series_approx8])
+    sin(x)*sin(y)
+
+    """
+    def __init__(self, bounds, reltol, max_order=4, n_point_checks=4, **kwargs):
+        super().__init__(**kwargs)
+        self.bounds = bounds
+        self.reltol = reltol
+        self.max_order = max_order
+        if n_point_checks % 2 == 1:
+            raise ValueError("Checking the solution at expansion point is not helpful")
+        self.n_point_checks = n_point_checks
+        self._prec = math.ceil(-math.log10(self.reltol))
+
+    def __call__(self, expr):
+        return expr.factor().replace(self.query, lambda arg: self.value(arg))
+
+    def query(self, expr):
+        return (expr.is_Function and not isinstance(expr, UndefinedFunction)
+                and len(expr.args) == 1)
+
+    def value(self, fexpr):
+        free_symbols = fexpr.free_symbols
+        if len(free_symbols) != 1:
+            return fexpr
+        symb, = free_symbols
+        if symb not in self.bounds:
+            return fexpr
+        lo, hi = self.bounds[symb]
+        x0 = (lo + hi)/2
+        cheapest = None
+        for n in range(self.max_order+1, 0, -1):
+            fseri = fexpr.series(symb, x0=x0, n=n).removeO()
+            n_ok = True
+            for idx in range(self.n_point_checks):
+                x = lo + idx*(hi - lo)/(self.n_point_checks - 1)
+                val = fseri.xreplace({symb: x})
+                ref = fexpr.xreplace({symb: x})
+                if abs((1 - val/ref).evalf(self._prec)) > self.reltol:
+                    n_ok = False
+                    break
+
+            if n_ok:
+                cheapest = fseri
+            else:
+                break
+
+        if cheapest is None:
+            return fexpr
+        else:
+            return cheapest

rl/Lib/site-packages/sympy/codegen/cfunctions.py

@@ -0,0 +1,536 @@
+"""
+This module contains SymPy functions mathcin corresponding to special math functions in the
+C standard library (since C99, also available in C++11).
+
+The functions defined in this module allows the user to express functions such as ``expm1``
+as a SymPy function for symbolic manipulation.
+
+"""
+from sympy.core.function import ArgumentIndexError, Function
+from sympy.core.numbers import Rational
+from sympy.core.power import Pow
+from sympy.core.singleton import S
+from sympy.functions.elementary.exponential import exp, log
+from sympy.functions.elementary.miscellaneous import sqrt
+
+
+def _expm1(x):
+    return exp(x) - S.One
+
+
+class expm1(Function):
+    """
+    Represents the exponential function minus one.
+
+    Explanation
+    ===========
+
+    The benefit of using ``expm1(x)`` over ``exp(x) - 1``
+    is that the latter is prone to cancellation under finite precision
+    arithmetic when x is close to zero.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import expm1
+    >>> '%.0e' % expm1(1e-99).evalf()
+    '1e-99'
+    >>> from math import exp
+    >>> exp(1e-99) - 1
+    0.0
+    >>> expm1(x).diff(x)
+    exp(x)
+
+    See Also
+    ========
+
+    log1p
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return exp(*self.args)
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+    def _eval_expand_func(self, **hints):
+        return _expm1(*self.args)
+
+    def _eval_rewrite_as_exp(self, arg, **kwargs):
+        return exp(arg) - S.One
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_exp
+
+    @classmethod
+    def eval(cls, arg):
+        exp_arg = exp.eval(arg)
+        if exp_arg is not None:
+            return exp_arg - S.One
+
+    def _eval_is_real(self):
+        return self.args[0].is_real
+
+    def _eval_is_finite(self):
+        return self.args[0].is_finite
+
+
+def _log1p(x):
+    return log(x + S.One)
+
+
+class log1p(Function):
+    """
+    Represents the natural logarithm of a number plus one.
+
+    Explanation
+    ===========
+
+    The benefit of using ``log1p(x)`` over ``log(x + 1)``
+    is that the latter is prone to cancellation under finite precision
+    arithmetic when x is close to zero.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import log1p
+    >>> from sympy import expand_log
+    >>> '%.0e' % expand_log(log1p(1e-99)).evalf()
+    '1e-99'
+    >>> from math import log
+    >>> log(1 + 1e-99)
+    0.0
+    >>> log1p(x).diff(x)
+    1/(x + 1)
+
+    See Also
+    ========
+
+    expm1
+    """
+    nargs = 1
+
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return S.One/(self.args[0] + S.One)
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    def _eval_expand_func(self, **hints):
+        return _log1p(*self.args)
+
+    def _eval_rewrite_as_log(self, arg, **kwargs):
+        return _log1p(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_log
+
+    @classmethod
+    def eval(cls, arg):
+        if arg.is_Rational:
+            return log(arg + S.One)
+        elif not arg.is_Float:  # not safe to add 1 to Float
+            return log.eval(arg + S.One)
+        elif arg.is_number:
+            return log(Rational(arg) + S.One)
+
+    def _eval_is_real(self):
+        return (self.args[0] + S.One).is_nonnegative
+
+    def _eval_is_finite(self):
+        if (self.args[0] + S.One).is_zero:
+            return False
+        return self.args[0].is_finite
+
+    def _eval_is_positive(self):
+        return self.args[0].is_positive
+
+    def _eval_is_zero(self):
+        return self.args[0].is_zero
+
+    def _eval_is_nonnegative(self):
+        return self.args[0].is_nonnegative
+
+_Two = S(2)
+
+def _exp2(x):
+    return Pow(_Two, x)
+
+class exp2(Function):
+    """
+    Represents the exponential function with base two.
+
+    Explanation
+    ===========
+
+    The benefit of using ``exp2(x)`` over ``2**x``
+    is that the latter is not as efficient under finite precision
+    arithmetic.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import exp2
+    >>> exp2(2).evalf() == 4.0
+    True
+    >>> exp2(x).diff(x)
+    log(2)*exp2(x)
+
+    See Also
+    ========
+
+    log2
+    """
+    nargs = 1
+
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return self*log(_Two)
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+    def _eval_rewrite_as_Pow(self, arg, **kwargs):
+        return _exp2(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_Pow
+
+    def _eval_expand_func(self, **hints):
+        return _exp2(*self.args)
+
+    @classmethod
+    def eval(cls, arg):
+        if arg.is_number:
+            return _exp2(arg)
+
+
+def _log2(x):
+    return log(x)/log(_Two)
+
+
+class log2(Function):
+    """
+    Represents the logarithm function with base two.
+
+    Explanation
+    ===========
+
+    The benefit of using ``log2(x)`` over ``log(x)/log(2)``
+    is that the latter is not as efficient under finite precision
+    arithmetic.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import log2
+    >>> log2(4).evalf() == 2.0
+    True
+    >>> log2(x).diff(x)
+    1/(x*log(2))
+
+    See Also
+    ========
+
+    exp2
+    log10
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return S.One/(log(_Two)*self.args[0])
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    @classmethod
+    def eval(cls, arg):
+        if arg.is_number:
+            result = log.eval(arg, base=_Two)
+            if result.is_Atom:
+                return result
+        elif arg.is_Pow and arg.base == _Two:
+            return arg.exp
+
+    def _eval_evalf(self, *args, **kwargs):
+        return self.rewrite(log).evalf(*args, **kwargs)
+
+    def _eval_expand_func(self, **hints):
+        return _log2(*self.args)
+
+    def _eval_rewrite_as_log(self, arg, **kwargs):
+        return _log2(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_log
+
+
+def _fma(x, y, z):
+    return x*y + z
+
+
+class fma(Function):
+    """
+    Represents "fused multiply add".
+
+    Explanation
+    ===========
+
+    The benefit of using ``fma(x, y, z)`` over ``x*y + z``
+    is that, under finite precision arithmetic, the former is
+    supported by special instructions on some CPUs.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x, y, z
+    >>> from sympy.codegen.cfunctions import fma
+    >>> fma(x, y, z).diff(x)
+    y
+
+    """
+    nargs = 3
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex in (1, 2):
+            return self.args[2 - argindex]
+        elif argindex == 3:
+            return S.One
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    def _eval_expand_func(self, **hints):
+        return _fma(*self.args)
+
+    def _eval_rewrite_as_tractable(self, arg, limitvar=None, **kwargs):
+        return _fma(arg)
+
+
+_Ten = S(10)
+
+
+def _log10(x):
+    return log(x)/log(_Ten)
+
+
+class log10(Function):
+    """
+    Represents the logarithm function with base ten.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import log10
+    >>> log10(100).evalf() == 2.0
+    True
+    >>> log10(x).diff(x)
+    1/(x*log(10))
+
+    See Also
+    ========
+
+    log2
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return S.One/(log(_Ten)*self.args[0])
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    @classmethod
+    def eval(cls, arg):
+        if arg.is_number:
+            result = log.eval(arg, base=_Ten)
+            if result.is_Atom:
+                return result
+        elif arg.is_Pow and arg.base == _Ten:
+            return arg.exp
+
+    def _eval_expand_func(self, **hints):
+        return _log10(*self.args)
+
+    def _eval_rewrite_as_log(self, arg, **kwargs):
+        return _log10(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_log
+
+
+def _Sqrt(x):
+    return Pow(x, S.Half)
+
+
+class Sqrt(Function):  # 'sqrt' already defined in sympy.functions.elementary.miscellaneous
+    """
+    Represents the square root function.
+
+    Explanation
+    ===========
+
+    The reason why one would use ``Sqrt(x)`` over ``sqrt(x)``
+    is that the latter is internally represented as ``Pow(x, S.Half)`` which
+    may not be what one wants when doing code-generation.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import Sqrt
+    >>> Sqrt(x)
+    Sqrt(x)
+    >>> Sqrt(x).diff(x)
+    1/(2*sqrt(x))
+
+    See Also
+    ========
+
+    Cbrt
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return Pow(self.args[0], Rational(-1, 2))/_Two
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+    def _eval_expand_func(self, **hints):
+        return _Sqrt(*self.args)
+
+    def _eval_rewrite_as_Pow(self, arg, **kwargs):
+        return _Sqrt(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_Pow
+
+
+def _Cbrt(x):
+    return Pow(x, Rational(1, 3))
+
+
+class Cbrt(Function):  # 'cbrt' already defined in sympy.functions.elementary.miscellaneous
+    """
+    Represents the cube root function.
+
+    Explanation
+    ===========
+
+    The reason why one would use ``Cbrt(x)`` over ``cbrt(x)``
+    is that the latter is internally represented as ``Pow(x, Rational(1, 3))`` which
+    may not be what one wants when doing code-generation.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cfunctions import Cbrt
+    >>> Cbrt(x)
+    Cbrt(x)
+    >>> Cbrt(x).diff(x)
+    1/(3*x**(2/3))
+
+    See Also
+    ========
+
+    Sqrt
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return Pow(self.args[0], Rational(-_Two/3))/3
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    def _eval_expand_func(self, **hints):
+        return _Cbrt(*self.args)
+
+    def _eval_rewrite_as_Pow(self, arg, **kwargs):
+        return _Cbrt(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_Pow
+
+
+def _hypot(x, y):
+    return sqrt(Pow(x, 2) + Pow(y, 2))
+
+
+class hypot(Function):
+    """
+    Represents the hypotenuse function.
+
+    Explanation
+    ===========
+
+    The hypotenuse function is provided by e.g. the math library
+    in the C99 standard, hence one may want to represent the function
+    symbolically when doing code-generation.
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x, y
+    >>> from sympy.codegen.cfunctions import hypot
+    >>> hypot(3, 4).evalf() == 5.0
+    True
+    >>> hypot(x, y)
+    hypot(x, y)
+    >>> hypot(x, y).diff(x)
+    x/hypot(x, y)
+
+    """
+    nargs = 2
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex in (1, 2):
+            return 2*self.args[argindex-1]/(_Two*self.func(*self.args))
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+
+    def _eval_expand_func(self, **hints):
+        return _hypot(*self.args)
+
+    def _eval_rewrite_as_Pow(self, arg, **kwargs):
+        return _hypot(arg)
+
+    _eval_rewrite_as_tractable = _eval_rewrite_as_Pow
+
+
+class isnan(Function):
+    nargs = 1

rl/Lib/site-packages/sympy/codegen/cnodes.py

@@ -0,0 +1,156 @@
+"""
+AST nodes specific to the C family of languages
+"""
+
+from sympy.codegen.ast import (
+    Attribute, Declaration, Node, String, Token, Type, none,
+    FunctionCall, CodeBlock
+    )
+from sympy.core.basic import Basic
+from sympy.core.containers import Tuple
+from sympy.core.sympify import sympify
+
+void = Type('void')
+
+restrict = Attribute('restrict')  # guarantees no pointer aliasing
+volatile = Attribute('volatile')
+static = Attribute('static')
+
+
+def alignof(arg):
+    """ Generate of FunctionCall instance for calling 'alignof' """
+    return FunctionCall('alignof', [String(arg) if isinstance(arg, str) else arg])
+
+
+def sizeof(arg):
+    """ Generate of FunctionCall instance for calling 'sizeof'
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.ast import real
+    >>> from sympy.codegen.cnodes import sizeof
+    >>> from sympy import ccode
+    >>> ccode(sizeof(real))
+    'sizeof(double)'
+    """
+    return FunctionCall('sizeof', [String(arg) if isinstance(arg, str) else arg])
+
+
+class CommaOperator(Basic):
+    """ Represents the comma operator in C """
+    def __new__(cls, *args):
+        return Basic.__new__(cls, *[sympify(arg) for arg in args])
+
+
+class Label(Node):
+    """ Label for use with e.g. goto statement.
+
+    Examples
+    ========
+
+    >>> from sympy import ccode, Symbol
+    >>> from sympy.codegen.cnodes import Label, PreIncrement
+    >>> print(ccode(Label('foo')))
+    foo:
+    >>> print(ccode(Label('bar', [PreIncrement(Symbol('a'))])))
+    bar:
+    ++(a);
+
+    """
+    __slots__ = _fields = ('name', 'body')
+    defaults = {'body': none}
+    _construct_name = String
+
+    @classmethod
+    def _construct_body(cls, itr):
+        if isinstance(itr, CodeBlock):
+            return itr
+        else:
+            return CodeBlock(*itr)
+
+
+class goto(Token):
+    """ Represents goto in C """
+    __slots__ = _fields = ('label',)
+    _construct_label = Label
+
+
+class PreDecrement(Basic):
+    """ Represents the pre-decrement operator
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cnodes import PreDecrement
+    >>> from sympy import ccode
+    >>> ccode(PreDecrement(x))
+    '--(x)'
+
+    """
+    nargs = 1
+
+
+class PostDecrement(Basic):
+    """ Represents the post-decrement operator
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cnodes import PostDecrement
+    >>> from sympy import ccode
+    >>> ccode(PostDecrement(x))
+    '(x)--'
+
+    """
+    nargs = 1
+
+
+class PreIncrement(Basic):
+    """ Represents the pre-increment operator
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cnodes import PreIncrement
+    >>> from sympy import ccode
+    >>> ccode(PreIncrement(x))
+    '++(x)'
+
+    """
+    nargs = 1
+
+
+class PostIncrement(Basic):
+    """ Represents the post-increment operator
+
+    Examples
+    ========
+
+    >>> from sympy.abc import x
+    >>> from sympy.codegen.cnodes import PostIncrement
+    >>> from sympy import ccode
+    >>> ccode(PostIncrement(x))
+    '(x)++'
+
+    """
+    nargs = 1
+
+
+class struct(Node):
+    """ Represents a struct in C """
+    __slots__ = _fields = ('name', 'declarations')
+    defaults = {'name': none}
+    _construct_name = String
+
+    @classmethod
+    def _construct_declarations(cls, args):
+        return Tuple(*[Declaration(arg) for arg in args])
+
+
+class union(struct):
+    """ Represents a union in C """
+    __slots__ = ()

rl/Lib/site-packages/sympy/codegen/cutils.py

@@ -0,0 +1,8 @@
+from sympy.printing.c import C99CodePrinter
+
+def render_as_source_file(content, Printer=C99CodePrinter, settings=None):
+    """ Renders a C source file (with required #include statements) """
+    printer = Printer(settings or {})
+    code_str = printer.doprint(content)
+    includes = '\n'.join(['#include <%s>' % h for h in printer.headers])
+    return includes + '\n\n' + code_str

rl/Lib/site-packages/sympy/codegen/cxxnodes.py

@@ -0,0 +1,14 @@
+"""
+AST nodes specific to C++.
+"""
+
+from sympy.codegen.ast import Attribute, String, Token, Type, none
+
+class using(Token):
+    """ Represents a 'using' statement in C++ """
+    __slots__ = _fields = ('type', 'alias')
+    defaults = {'alias': none}
+    _construct_type = Type
+    _construct_alias = String
+
+constexpr = Attribute('constexpr')

rl/Lib/site-packages/sympy/codegen/fnodes.py

@@ -0,0 +1,657 @@
+"""
+AST nodes specific to Fortran.
+
+The functions defined in this module allows the user to express functions such as ``dsign``
+as a SymPy function for symbolic manipulation.
+"""
+
+from sympy.codegen.ast import (
+    Attribute, CodeBlock, FunctionCall, Node, none, String,
+    Token, _mk_Tuple, Variable
+)
+from sympy.core.basic import Basic
+from sympy.core.containers import Tuple
+from sympy.core.expr import Expr
+from sympy.core.function import Function
+from sympy.core.numbers import Float, Integer
+from sympy.core.symbol import Str
+from sympy.core.sympify import sympify
+from sympy.logic import true, false
+from sympy.utilities.iterables import iterable
+
+
+
+pure = Attribute('pure')
+elemental = Attribute('elemental')  # (all elemental procedures are also pure)
+
+intent_in = Attribute('intent_in')
+intent_out = Attribute('intent_out')
+intent_inout = Attribute('intent_inout')
+
+allocatable = Attribute('allocatable')
+
+class Program(Token):
+    """ Represents a 'program' block in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.ast import Print
+    >>> from sympy.codegen.fnodes import Program
+    >>> prog = Program('myprogram', [Print([42])])
+    >>> from sympy import fcode
+    >>> print(fcode(prog, source_format='free'))
+    program myprogram
+        print *, 42
+    end program
+
+    """
+    __slots__ = _fields = ('name', 'body')
+    _construct_name = String
+    _construct_body = staticmethod(lambda body: CodeBlock(*body))
+
+
+class use_rename(Token):
+    """ Represents a renaming in a use statement in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import use_rename, use
+    >>> from sympy import fcode
+    >>> ren = use_rename("thingy", "convolution2d")
+    >>> print(fcode(ren, source_format='free'))
+    thingy => convolution2d
+    >>> full = use('signallib', only=['snr', ren])
+    >>> print(fcode(full, source_format='free'))
+    use signallib, only: snr, thingy => convolution2d
+
+    """
+    __slots__ = _fields = ('local', 'original')
+    _construct_local = String
+    _construct_original = String
+
+def _name(arg):
+    if hasattr(arg, 'name'):
+        return arg.name
+    else:
+        return String(arg)
+
+class use(Token):
+    """ Represents a use statement in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import use
+    >>> from sympy import fcode
+    >>> fcode(use('signallib'), source_format='free')
+    'use signallib'
+    >>> fcode(use('signallib', [('metric', 'snr')]), source_format='free')
+    'use signallib, metric => snr'
+    >>> fcode(use('signallib', only=['snr', 'convolution2d']), source_format='free')
+    'use signallib, only: snr, convolution2d'
+
+    """
+    __slots__ = _fields = ('namespace', 'rename', 'only')
+    defaults = {'rename': none, 'only': none}
+    _construct_namespace = staticmethod(_name)
+    _construct_rename = staticmethod(lambda args: Tuple(*[arg if isinstance(arg, use_rename) else use_rename(*arg) for arg in args]))
+    _construct_only = staticmethod(lambda args: Tuple(*[arg if isinstance(arg, use_rename) else _name(arg) for arg in args]))
+
+
+class Module(Token):
+    """ Represents a module in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import Module
+    >>> from sympy import fcode
+    >>> print(fcode(Module('signallib', ['implicit none'], []), source_format='free'))
+    module signallib
+    implicit none
+    <BLANKLINE>
+    contains
+    <BLANKLINE>
+    <BLANKLINE>
+    end module
+
+    """
+    __slots__ = _fields = ('name', 'declarations', 'definitions')
+    defaults = {'declarations': Tuple()}
+    _construct_name = String
+
+    @classmethod
+    def _construct_declarations(cls, args):
+        args = [Str(arg) if isinstance(arg, str) else arg for arg in args]
+        return CodeBlock(*args)
+
+    _construct_definitions = staticmethod(lambda arg: CodeBlock(*arg))
+
+
+class Subroutine(Node):
+    """ Represents a subroutine in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy import fcode, symbols
+    >>> from sympy.codegen.ast import Print
+    >>> from sympy.codegen.fnodes import Subroutine
+    >>> x, y = symbols('x y', real=True)
+    >>> sub = Subroutine('mysub', [x, y], [Print([x**2 + y**2, x*y])])
+    >>> print(fcode(sub, source_format='free', standard=2003))
+    subroutine mysub(x, y)
+    real*8 :: x
+    real*8 :: y
+    print *, x**2 + y**2, x*y
+    end subroutine
+
+    """
+    __slots__ = ('name', 'parameters', 'body')
+    _fields = __slots__ + Node._fields
+    _construct_name = String
+    _construct_parameters = staticmethod(lambda params: Tuple(*map(Variable.deduced, params)))
+
+    @classmethod
+    def _construct_body(cls, itr):
+        if isinstance(itr, CodeBlock):
+            return itr
+        else:
+            return CodeBlock(*itr)
+
+class SubroutineCall(Token):
+    """ Represents a call to a subroutine in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import SubroutineCall
+    >>> from sympy import fcode
+    >>> fcode(SubroutineCall('mysub', 'x y'.split()))
+    '       call mysub(x, y)'
+
+    """
+    __slots__ = _fields = ('name', 'subroutine_args')
+    _construct_name = staticmethod(_name)
+    _construct_subroutine_args = staticmethod(_mk_Tuple)
+
+
+class Do(Token):
+    """ Represents a Do loop in in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy import fcode, symbols
+    >>> from sympy.codegen.ast import aug_assign, Print
+    >>> from sympy.codegen.fnodes import Do
+    >>> i, n = symbols('i n', integer=True)
+    >>> r = symbols('r', real=True)
+    >>> body = [aug_assign(r, '+', 1/i), Print([i, r])]
+    >>> do1 = Do(body, i, 1, n)
+    >>> print(fcode(do1, source_format='free'))
+    do i = 1, n
+        r = r + 1d0/i
+        print *, i, r
+    end do
+    >>> do2 = Do(body, i, 1, n, 2)
+    >>> print(fcode(do2, source_format='free'))
+    do i = 1, n, 2
+        r = r + 1d0/i
+        print *, i, r
+    end do
+
+    """
+
+    __slots__ = _fields = ('body', 'counter', 'first', 'last', 'step', 'concurrent')
+    defaults = {'step': Integer(1), 'concurrent': false}
+    _construct_body = staticmethod(lambda body: CodeBlock(*body))
+    _construct_counter = staticmethod(sympify)
+    _construct_first = staticmethod(sympify)
+    _construct_last = staticmethod(sympify)
+    _construct_step = staticmethod(sympify)
+    _construct_concurrent = staticmethod(lambda arg: true if arg else false)
+
+
+class ArrayConstructor(Token):
+    """ Represents an array constructor.
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.fnodes import ArrayConstructor
+    >>> ac = ArrayConstructor([1, 2, 3])
+    >>> fcode(ac, standard=95, source_format='free')
+    '(/1, 2, 3/)'
+    >>> fcode(ac, standard=2003, source_format='free')
+    '[1, 2, 3]'
+
+    """
+    __slots__ = _fields = ('elements',)
+    _construct_elements = staticmethod(_mk_Tuple)
+
+
+class ImpliedDoLoop(Token):
+    """ Represents an implied do loop in Fortran.
+
+    Examples
+    ========
+
+    >>> from sympy import Symbol, fcode
+    >>> from sympy.codegen.fnodes import ImpliedDoLoop, ArrayConstructor
+    >>> i = Symbol('i', integer=True)
+    >>> idl = ImpliedDoLoop(i**3, i, -3, 3, 2)  # -27, -1, 1, 27
+    >>> ac = ArrayConstructor([-28, idl, 28]) # -28, -27, -1, 1, 27, 28
+    >>> fcode(ac, standard=2003, source_format='free')
+    '[-28, (i**3, i = -3, 3, 2), 28]'
+
+    """
+    __slots__ = _fields = ('expr', 'counter', 'first', 'last', 'step')
+    defaults = {'step': Integer(1)}
+    _construct_expr = staticmethod(sympify)
+    _construct_counter = staticmethod(sympify)
+    _construct_first = staticmethod(sympify)
+    _construct_last = staticmethod(sympify)
+    _construct_step = staticmethod(sympify)
+
+
+class Extent(Basic):
+    """ Represents a dimension extent.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import Extent
+    >>> e = Extent(-3, 3)  # -3, -2, -1, 0, 1, 2, 3
+    >>> from sympy import fcode
+    >>> fcode(e, source_format='free')
+    '-3:3'
+    >>> from sympy.codegen.ast import Variable, real
+    >>> from sympy.codegen.fnodes import dimension, intent_out
+    >>> dim = dimension(e, e)
+    >>> arr = Variable('x', real, attrs=[dim, intent_out])
+    >>> fcode(arr.as_Declaration(), source_format='free', standard=2003)
+    'real*8, dimension(-3:3, -3:3), intent(out) :: x'
+
+    """
+    def __new__(cls, *args):
+        if len(args) == 2:
+            low, high = args
+            return Basic.__new__(cls, sympify(low), sympify(high))
+        elif len(args) == 0 or (len(args) == 1 and args[0] in (':', None)):
+            return Basic.__new__(cls)  # assumed shape
+        else:
+            raise ValueError("Expected 0 or 2 args (or one argument == None or ':')")
+
+    def _sympystr(self, printer):
+        if len(self.args) == 0:
+            return ':'
+        return ":".join(str(arg) for arg in self.args)
+
+assumed_extent = Extent() # or Extent(':'), Extent(None)
+
+
+def dimension(*args):
+    """ Creates a 'dimension' Attribute with (up to 7) extents.
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.fnodes import dimension, intent_in
+    >>> dim = dimension('2', ':')  # 2 rows, runtime determined number of columns
+    >>> from sympy.codegen.ast import Variable, integer
+    >>> arr = Variable('a', integer, attrs=[dim, intent_in])
+    >>> fcode(arr.as_Declaration(), source_format='free', standard=2003)
+    'integer*4, dimension(2, :), intent(in) :: a'
+
+    """
+    if len(args) > 7:
+        raise ValueError("Fortran only supports up to 7 dimensional arrays")
+    parameters = []
+    for arg in args:
+        if isinstance(arg, Extent):
+            parameters.append(arg)
+        elif isinstance(arg, str):
+            if arg == ':':
+                parameters.append(Extent())
+            else:
+                parameters.append(String(arg))
+        elif iterable(arg):
+            parameters.append(Extent(*arg))
+        else:
+            parameters.append(sympify(arg))
+    if len(args) == 0:
+        raise ValueError("Need at least one dimension")
+    return Attribute('dimension', parameters)
+
+
+assumed_size = dimension('*')
+
+def array(symbol, dim, intent=None, *, attrs=(), value=None, type=None):
+    """ Convenience function for creating a Variable instance for a Fortran array.
+
+    Parameters
+    ==========
+
+    symbol : symbol
+    dim : Attribute or iterable
+        If dim is an ``Attribute`` it need to have the name 'dimension'. If it is
+        not an ``Attribute``, then it is passed to :func:`dimension` as ``*dim``
+    intent : str
+        One of: 'in', 'out', 'inout' or None
+    \\*\\*kwargs:
+        Keyword arguments for ``Variable`` ('type' & 'value')
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.ast import integer, real
+    >>> from sympy.codegen.fnodes import array
+    >>> arr = array('a', '*', 'in', type=integer)
+    >>> print(fcode(arr.as_Declaration(), source_format='free', standard=2003))
+    integer*4, dimension(*), intent(in) :: a
+    >>> x = array('x', [3, ':', ':'], intent='out', type=real)
+    >>> print(fcode(x.as_Declaration(value=1), source_format='free', standard=2003))
+    real*8, dimension(3, :, :), intent(out) :: x = 1
+
+    """
+    if isinstance(dim, Attribute):
+        if str(dim.name) != 'dimension':
+            raise ValueError("Got an unexpected Attribute argument as dim: %s" % str(dim))
+    else:
+        dim = dimension(*dim)
+
+    attrs = list(attrs) + [dim]
+    if intent is not None:
+        if intent not in (intent_in, intent_out, intent_inout):
+            intent = {'in': intent_in, 'out': intent_out, 'inout': intent_inout}[intent]
+        attrs.append(intent)
+    if type is None:
+        return Variable.deduced(symbol, value=value, attrs=attrs)
+    else:
+        return Variable(symbol, type, value=value, attrs=attrs)
+
+def _printable(arg):
+    return String(arg) if isinstance(arg, str) else sympify(arg)
+
+
+def allocated(array):
+    """ Creates an AST node for a function call to Fortran's "allocated(...)"
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.fnodes import allocated
+    >>> alloc = allocated('x')
+    >>> fcode(alloc, source_format='free')
+    'allocated(x)'
+
+    """
+    return FunctionCall('allocated', [_printable(array)])
+
+
+def lbound(array, dim=None, kind=None):
+    """ Creates an AST node for a function call to Fortran's "lbound(...)"
+
+    Parameters
+    ==========
+
+    array : Symbol or String
+    dim : expr
+    kind : expr
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.fnodes import lbound
+    >>> lb = lbound('arr', dim=2)
+    >>> fcode(lb, source_format='free')
+    'lbound(arr, 2)'
+
+    """
+    return FunctionCall(
+        'lbound',
+        [_printable(array)] +
+        ([_printable(dim)] if dim else []) +
+        ([_printable(kind)] if kind else [])
+    )
+
+
+def ubound(array, dim=None, kind=None):
+    return FunctionCall(
+        'ubound',
+        [_printable(array)] +
+        ([_printable(dim)] if dim else []) +
+        ([_printable(kind)] if kind else [])
+    )
+
+
+def shape(source, kind=None):
+    """ Creates an AST node for a function call to Fortran's "shape(...)"
+
+    Parameters
+    ==========
+
+    source : Symbol or String
+    kind : expr
+
+    Examples
+    ========
+
+    >>> from sympy import fcode
+    >>> from sympy.codegen.fnodes import shape
+    >>> shp = shape('x')
+    >>> fcode(shp, source_format='free')
+    'shape(x)'
+
+    """
+    return FunctionCall(
+        'shape',
+        [_printable(source)] +
+        ([_printable(kind)] if kind else [])
+    )
+
+
+def size(array, dim=None, kind=None):
+    """ Creates an AST node for a function call to Fortran's "size(...)"
+
+    Examples
+    ========
+
+    >>> from sympy import fcode, Symbol
+    >>> from sympy.codegen.ast import FunctionDefinition, real, Return
+    >>> from sympy.codegen.fnodes import array, sum_, size
+    >>> a = Symbol('a', real=True)
+    >>> body = [Return((sum_(a**2)/size(a))**.5)]
+    >>> arr = array(a, dim=[':'], intent='in')
+    >>> fd = FunctionDefinition(real, 'rms', [arr], body)
+    >>> print(fcode(fd, source_format='free', standard=2003))
+    real*8 function rms(a)
+    real*8, dimension(:), intent(in) :: a
+    rms = sqrt(sum(a**2)*1d0/size(a))
+    end function
+
+    """
+    return FunctionCall(
+        'size',
+        [_printable(array)] +
+        ([_printable(dim)] if dim else []) +
+        ([_printable(kind)] if kind else [])
+    )
+
+
+def reshape(source, shape, pad=None, order=None):
+    """ Creates an AST node for a function call to Fortran's "reshape(...)"
+
+    Parameters
+    ==========
+
+    source : Symbol or String
+    shape : ArrayExpr
+
+    """
+    return FunctionCall(
+        'reshape',
+        [_printable(source), _printable(shape)] +
+        ([_printable(pad)] if pad else []) +
+        ([_printable(order)] if pad else [])
+    )
+
+
+def bind_C(name=None):
+    """ Creates an Attribute ``bind_C`` with a name.
+
+    Parameters
+    ==========
+
+    name : str
+
+    Examples
+    ========
+
+    >>> from sympy import fcode, Symbol
+    >>> from sympy.codegen.ast import FunctionDefinition, real, Return
+    >>> from sympy.codegen.fnodes import array, sum_, bind_C
+    >>> a = Symbol('a', real=True)
+    >>> s = Symbol('s', integer=True)
+    >>> arr = array(a, dim=[s], intent='in')
+    >>> body = [Return((sum_(a**2)/s)**.5)]
+    >>> fd = FunctionDefinition(real, 'rms', [arr, s], body, attrs=[bind_C('rms')])
+    >>> print(fcode(fd, source_format='free', standard=2003))
+    real*8 function rms(a, s) bind(C, name="rms")
+    real*8, dimension(s), intent(in) :: a
+    integer*4 :: s
+    rms = sqrt(sum(a**2)/s)
+    end function
+
+    """
+    return Attribute('bind_C', [String(name)] if name else [])
+
+class GoTo(Token):
+    """ Represents a goto statement in Fortran
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import GoTo
+    >>> go = GoTo([10, 20, 30], 'i')
+    >>> from sympy import fcode
+    >>> fcode(go, source_format='free')
+    'go to (10, 20, 30), i'
+
+    """
+    __slots__ = _fields = ('labels', 'expr')
+    defaults = {'expr': none}
+    _construct_labels = staticmethod(_mk_Tuple)
+    _construct_expr = staticmethod(sympify)
+
+
+class FortranReturn(Token):
+    """ AST node explicitly mapped to a fortran "return".
+
+    Explanation
+    ===========
+
+    Because a return statement in fortran is different from C, and
+    in order to aid reuse of our codegen ASTs the ordinary
+    ``.codegen.ast.Return`` is interpreted as assignment to
+    the result variable of the function. If one for some reason needs
+    to generate a fortran RETURN statement, this node should be used.
+
+    Examples
+    ========
+
+    >>> from sympy.codegen.fnodes import FortranReturn
+    >>> from sympy import fcode
+    >>> fcode(FortranReturn('x'))
+    '       return x'
+
+    """
+    __slots__ = _fields = ('return_value',)
+    defaults = {'return_value': none}
+    _construct_return_value = staticmethod(sympify)
+
+
+class FFunction(Function):
+    _required_standard = 77
+
+    def _fcode(self, printer):
+        name = self.__class__.__name__
+        if printer._settings['standard'] < self._required_standard:
+            raise NotImplementedError("%s requires Fortran %d or newer" %
+                                      (name, self._required_standard))
+        return '{}({})'.format(name, ', '.join(map(printer._print, self.args)))
+
+
+class F95Function(FFunction):
+    _required_standard = 95
+
+
+class isign(FFunction):
+    """ Fortran sign intrinsic for integer arguments. """
+    nargs = 2
+
+
+class dsign(FFunction):
+    """ Fortran sign intrinsic for double precision arguments. """
+    nargs = 2
+
+
+class cmplx(FFunction):
+    """ Fortran complex conversion function. """
+    nargs = 2  # may be extended to (2, 3) at a later point
+
+
+class kind(FFunction):
+    """ Fortran kind function. """
+    nargs = 1
+
+
+class merge(F95Function):
+    """ Fortran merge function """
+    nargs = 3
+
+
+class _literal(Float):
+    _token = None  # type: str
+    _decimals = None  # type: int
+
+    def _fcode(self, printer, *args, **kwargs):
+        mantissa, sgnd_ex = ('%.{}e'.format(self._decimals) % self).split('e')
+        mantissa = mantissa.strip('0').rstrip('.')
+        ex_sgn, ex_num = sgnd_ex[0], sgnd_ex[1:].lstrip('0')
+        ex_sgn = '' if ex_sgn == '+' else ex_sgn
+        return (mantissa or '0') + self._token + ex_sgn + (ex_num or '0')
+
+
+class literal_sp(_literal):
+    """ Fortran single precision real literal """
+    _token = 'e'
+    _decimals = 9
+
+
+class literal_dp(_literal):
+    """ Fortran double precision real literal """
+    _token = 'd'
+    _decimals = 17
+
+
+class sum_(Token, Expr):
+    __slots__ = _fields = ('array', 'dim', 'mask')
+    defaults = {'dim': none, 'mask': none}
+    _construct_array = staticmethod(sympify)
+    _construct_dim = staticmethod(sympify)
+
+
+class product_(Token, Expr):
+    __slots__ = _fields = ('array', 'dim', 'mask')
+    defaults = {'dim': none, 'mask': none}
+    _construct_array = staticmethod(sympify)
+    _construct_dim = staticmethod(sympify)

rl/Lib/site-packages/sympy/codegen/futils.py

@@ -0,0 +1,40 @@
+from itertools import chain
+from sympy.codegen.fnodes import Module
+from sympy.core.symbol import Dummy
+from sympy.printing.fortran import FCodePrinter
+
+""" This module collects utilities for rendering Fortran code. """
+
+
+def render_as_module(definitions, name, declarations=(), printer_settings=None):
+    """ Creates a ``Module`` instance and renders it as a string.
+
+    This generates Fortran source code for a module with the correct ``use`` statements.
+
+    Parameters
+    ==========
+
+    definitions : iterable
+        Passed to :class:`sympy.codegen.fnodes.Module`.
+    name : str
+        Passed to :class:`sympy.codegen.fnodes.Module`.
+    declarations : iterable
+        Passed to :class:`sympy.codegen.fnodes.Module`. It will be extended with
+        use statements, 'implicit none' and public list generated from ``definitions``.
+    printer_settings : dict
+        Passed to ``FCodePrinter`` (default: ``{'standard': 2003, 'source_format': 'free'}``).
+
+    """
+    printer_settings = printer_settings or {'standard': 2003, 'source_format': 'free'}
+    printer = FCodePrinter(printer_settings)
+    dummy = Dummy()
+    if isinstance(definitions, Module):
+        raise ValueError("This function expects to construct a module on its own.")
+    mod = Module(name, chain(declarations, [dummy]), definitions)
+    fstr = printer.doprint(mod)
+    module_use_str = '   %s\n' % '   \n'.join(['use %s, only: %s' % (k, ', '.join(v)) for
+                                                k, v in printer.module_uses.items()])
+    module_use_str += '   implicit none\n'
+    module_use_str += '   private\n'
+    module_use_str += '   public %s\n' % ', '.join([str(node.name) for node in definitions if getattr(node, 'name', None)])
+    return fstr.replace(printer.doprint(dummy), module_use_str)

rl/Lib/site-packages/sympy/codegen/matrix_nodes.py

@@ -0,0 +1,71 @@
+"""
+Additional AST nodes for operations on matrices. The nodes in this module
+are meant to represent optimization of matrix expressions within codegen's
+target languages that cannot be represented by SymPy expressions.
+
+As an example, we can use :meth:`sympy.codegen.rewriting.optimize` and the
+``matin_opt`` optimization provided in :mod:`sympy.codegen.rewriting` to
+transform matrix multiplication under certain assumptions:
+
+    >>> from sympy import symbols, MatrixSymbol
+    >>> n = symbols('n', integer=True)
+    >>> A = MatrixSymbol('A', n, n)
+    >>> x = MatrixSymbol('x', n, 1)
+    >>> expr = A**(-1) * x
+    >>> from sympy import assuming, Q
+    >>> from sympy.codegen.rewriting import matinv_opt, optimize
+    >>> with assuming(Q.fullrank(A)):
+    ...     optimize(expr, [matinv_opt])
+    MatrixSolve(A, vector=x)
+"""
+
+from .ast import Token
+from sympy.matrices import MatrixExpr
+from sympy.core.sympify import sympify
+
+
+class MatrixSolve(Token, MatrixExpr):
+    """Represents an operation to solve a linear matrix equation.
+
+    Parameters
+    ==========
+
+    matrix : MatrixSymbol
+
+      Matrix representing the coefficients of variables in the linear
+      equation. This matrix must be square and full-rank (i.e. all columns must
+      be linearly independent) for the solving operation to be valid.
+
+    vector : MatrixSymbol
+
+      One-column matrix representing the solutions to the equations
+      represented in ``matrix``.
+
+    Examples
+    ========
+
+    >>> from sympy import symbols, MatrixSymbol
+    >>> from sympy.codegen.matrix_nodes import MatrixSolve
+    >>> n = symbols('n', integer=True)
+    >>> A = MatrixSymbol('A', n, n)
+    >>> x = MatrixSymbol('x', n, 1)
+    >>> from sympy.printing.numpy import NumPyPrinter
+    >>> NumPyPrinter().doprint(MatrixSolve(A, x))
+    'numpy.linalg.solve(A, x)'
+    >>> from sympy import octave_code
+    >>> octave_code(MatrixSolve(A, x))
+    'A \\\\ x'
+
+    """
+    __slots__ = _fields = ('matrix', 'vector')
+
+    _construct_matrix = staticmethod(sympify)
+    _construct_vector = staticmethod(sympify)
+
+    @property
+    def shape(self):
+        return self.vector.shape
+
+    def _eval_derivative(self, x):
+        A, b = self.matrix, self.vector
+        return MatrixSolve(A, b.diff(x) - A.diff(x) * MatrixSolve(A, b))

rl/Lib/site-packages/sympy/codegen/numpy_nodes.py

@@ -0,0 +1,110 @@
+from sympy.core.function import Add, ArgumentIndexError, Function
+from sympy.core.power import Pow
+from sympy.core.singleton import S
+from sympy.core.sorting import default_sort_key
+from sympy.functions.elementary.exponential import exp, log
+
+
+def _logaddexp(x1, x2, *, evaluate=True):
+    return log(Add(exp(x1, evaluate=evaluate), exp(x2, evaluate=evaluate), evaluate=evaluate))
+
+
+_two = S.One*2
+_ln2 = log(_two)
+
+
+def _lb(x, *, evaluate=True):
+    return log(x, evaluate=evaluate)/_ln2
+
+
+def _exp2(x, *, evaluate=True):
+    return Pow(_two, x, evaluate=evaluate)
+
+
+def _logaddexp2(x1, x2, *, evaluate=True):
+    return _lb(Add(_exp2(x1, evaluate=evaluate),
+                   _exp2(x2, evaluate=evaluate), evaluate=evaluate))
+
+
+class logaddexp(Function):
+    """ Logarithm of the sum of exponentiations of the inputs.
+
+    Helper class for use with e.g. numpy.logaddexp
+
+    See Also
+    ========
+
+    https://numpy.org/doc/stable/reference/generated/numpy.logaddexp.html
+    """
+    nargs = 2
+
+    def __new__(cls, *args):
+        return Function.__new__(cls, *sorted(args, key=default_sort_key))
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            wrt, other = self.args
+        elif argindex == 2:
+            other, wrt = self.args
+        else:
+            raise ArgumentIndexError(self, argindex)
+        return S.One/(S.One + exp(other-wrt))
+
+    def _eval_rewrite_as_log(self, x1, x2, **kwargs):
+        return _logaddexp(x1, x2)
+
+    def _eval_evalf(self, *args, **kwargs):
+        return self.rewrite(log).evalf(*args, **kwargs)
+
+    def _eval_simplify(self, *args, **kwargs):
+        a, b = (x.simplify(**kwargs) for x in self.args)
+        candidate = _logaddexp(a, b)
+        if candidate != _logaddexp(a, b, evaluate=False):
+            return candidate
+        else:
+            return logaddexp(a, b)
+
+
+class logaddexp2(Function):
+    """ Logarithm of the sum of exponentiations of the inputs in base-2.
+
+    Helper class for use with e.g. numpy.logaddexp2
+
+    See Also
+    ========
+
+    https://numpy.org/doc/stable/reference/generated/numpy.logaddexp2.html
+    """
+    nargs = 2
+
+    def __new__(cls, *args):
+        return Function.__new__(cls, *sorted(args, key=default_sort_key))
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            wrt, other = self.args
+        elif argindex == 2:
+            other, wrt = self.args
+        else:
+            raise ArgumentIndexError(self, argindex)
+        return S.One/(S.One + _exp2(other-wrt))
+
+    def _eval_rewrite_as_log(self, x1, x2, **kwargs):
+        return _logaddexp2(x1, x2)
+
+    def _eval_evalf(self, *args, **kwargs):
+        return self.rewrite(log).evalf(*args, **kwargs)
+
+    def _eval_simplify(self, *args, **kwargs):
+        a, b = (x.simplify(**kwargs).factor() for x in self.args)
+        candidate = _logaddexp2(a, b)
+        if candidate != _logaddexp2(a, b, evaluate=False):
+            return candidate
+        else:
+            return logaddexp2(a, b)

rl/Lib/site-packages/sympy/codegen/pynodes.py

@@ -0,0 +1,11 @@
+from .abstract_nodes import List as AbstractList
+from .ast import Token
+
+
+class List(AbstractList):
+    pass
+
+
+class NumExprEvaluate(Token):
+    """represents a call to :class:`numexpr`s :func:`evaluate`"""
+    __slots__ = _fields = ('expr',)

rl/Lib/site-packages/sympy/codegen/pyutils.py

@@ -0,0 +1,24 @@
+from sympy.printing.pycode import PythonCodePrinter
+
+""" This module collects utilities for rendering Python code. """
+
+
+def render_as_module(content, standard='python3'):
+    """Renders Python code as a module (with the required imports).
+
+    Parameters
+    ==========
+
+    standard :
+        See the parameter ``standard`` in
+        :meth:`sympy.printing.pycode.pycode`
+    """
+
+    printer = PythonCodePrinter({'standard':standard})
+    pystr = printer.doprint(content)
+    if printer._settings['fully_qualified_modules']:
+        module_imports_str = '\n'.join('import %s' % k for k in printer.module_imports)
+    else:
+        module_imports_str = '\n'.join(['from %s import %s' % (k, ', '.join(v)) for
+                                        k, v in printer.module_imports.items()])
+    return module_imports_str + '\n\n' + pystr

rl/Lib/site-packages/sympy/codegen/rewriting.py

@@ -0,0 +1,357 @@
+"""
+Classes and functions useful for rewriting expressions for optimized code
+generation. Some languages (or standards thereof), e.g. C99, offer specialized
+math functions for better performance and/or precision.
+
+Using the ``optimize`` function in this module, together with a collection of
+rules (represented as instances of ``Optimization``), one can rewrite the
+expressions for this purpose::
+
+    >>> from sympy import Symbol, exp, log
+    >>> from sympy.codegen.rewriting import optimize, optims_c99
+    >>> x = Symbol('x')
+    >>> optimize(3*exp(2*x) - 3, optims_c99)
+    3*expm1(2*x)
+    >>> optimize(exp(2*x) - 1 - exp(-33), optims_c99)
+    expm1(2*x) - exp(-33)
+    >>> optimize(log(3*x + 3), optims_c99)
+    log1p(x) + log(3)
+    >>> optimize(log(2*x + 3), optims_c99)
+    log(2*x + 3)
+
+The ``optims_c99`` imported above is tuple containing the following instances
+(which may be imported from ``sympy.codegen.rewriting``):
+
+- ``expm1_opt``
+- ``log1p_opt``
+- ``exp2_opt``
+- ``log2_opt``
+- ``log2const_opt``
+
+
+"""
+from sympy.core.function import expand_log
+from sympy.core.singleton import S
+from sympy.core.symbol import Wild
+from sympy.functions.elementary.complexes import sign
+from sympy.functions.elementary.exponential import (exp, log)
+from sympy.functions.elementary.miscellaneous import (Max, Min)
+from sympy.functions.elementary.trigonometric import (cos, sin, sinc)
+from sympy.assumptions import Q, ask
+from sympy.codegen.cfunctions import log1p, log2, exp2, expm1
+from sympy.codegen.matrix_nodes import MatrixSolve
+from sympy.core.expr import UnevaluatedExpr
+from sympy.core.power import Pow
+from sympy.codegen.numpy_nodes import logaddexp, logaddexp2
+from sympy.codegen.scipy_nodes import cosm1, powm1
+from sympy.core.mul import Mul
+from sympy.matrices.expressions.matexpr import MatrixSymbol
+from sympy.utilities.iterables import sift
+
+
+class Optimization:
+    """ Abstract base class for rewriting optimization.
+
+    Subclasses should implement ``__call__`` taking an expression
+    as argument.
+
+    Parameters
+    ==========
+    cost_function : callable returning number
+    priority : number
+
+    """
+    def __init__(self, cost_function=None, priority=1):
+        self.cost_function = cost_function
+        self.priority=priority
+
+    def cheapest(self, *args):
+        return min(args, key=self.cost_function)
+
+
+class ReplaceOptim(Optimization):
+    """ Rewriting optimization calling replace on expressions.
+
+    Explanation
+    ===========
+
+    The instance can be used as a function on expressions for which
+    it will apply the ``replace`` method (see
+    :meth:`sympy.core.basic.Basic.replace`).
+
+    Parameters
+    ==========
+
+    query :
+        First argument passed to replace.
+    value :
+        Second argument passed to replace.
+
+    Examples
+    ========
+
+    >>> from sympy import Symbol
+    >>> from sympy.codegen.rewriting import ReplaceOptim
+    >>> from sympy.codegen.cfunctions import exp2
+    >>> x = Symbol('x')
+    >>> exp2_opt = ReplaceOptim(lambda p: p.is_Pow and p.base == 2,
+    ...     lambda p: exp2(p.exp))
+    >>> exp2_opt(2**x)
+    exp2(x)
+
+    """
+
+    def __init__(self, query, value, **kwargs):
+        super().__init__(**kwargs)
+        self.query = query
+        self.value = value
+
+    def __call__(self, expr):
+        return expr.replace(self.query, self.value)
+
+
+def optimize(expr, optimizations):
+    """ Apply optimizations to an expression.
+
+    Parameters
+    ==========
+
+    expr : expression
+    optimizations : iterable of ``Optimization`` instances
+        The optimizations will be sorted with respect to ``priority`` (highest first).
+
+    Examples
+    ========
+
+    >>> from sympy import log, Symbol
+    >>> from sympy.codegen.rewriting import optims_c99, optimize
+    >>> x = Symbol('x')
+    >>> optimize(log(x+3)/log(2) + log(x**2 + 1), optims_c99)
+    log1p(x**2) + log2(x + 3)
+
+    """
+
+    for optim in sorted(optimizations, key=lambda opt: opt.priority, reverse=True):
+        new_expr = optim(expr)
+        if optim.cost_function is None:
+            expr = new_expr
+        else:
+            expr = optim.cheapest(expr, new_expr)
+    return expr
+
+
+exp2_opt = ReplaceOptim(
+    lambda p: p.is_Pow and p.base == 2,
+    lambda p: exp2(p.exp)
+)
+
+
+_d = Wild('d', properties=[lambda x: x.is_Dummy])
+_u = Wild('u', properties=[lambda x: not x.is_number and not x.is_Add])
+_v = Wild('v')
+_w = Wild('w')
+_n = Wild('n', properties=[lambda x: x.is_number])
+
+sinc_opt1 = ReplaceOptim(
+    sin(_w)/_w, sinc(_w)
+)
+sinc_opt2 = ReplaceOptim(
+    sin(_n*_w)/_w, _n*sinc(_n*_w)
+)
+sinc_opts = (sinc_opt1, sinc_opt2)
+
+log2_opt = ReplaceOptim(_v*log(_w)/log(2), _v*log2(_w), cost_function=lambda expr: expr.count(
+    lambda e: (  # division & eval of transcendentals are expensive floating point operations...
+        e.is_Pow and e.exp.is_negative  # division
+        or (isinstance(e, (log, log2)) and not e.args[0].is_number))  # transcendental
+    )
+)
+
+log2const_opt = ReplaceOptim(log(2)*log2(_w), log(_w))
+
+logsumexp_2terms_opt = ReplaceOptim(
+    lambda l: (isinstance(l, log)
+               and l.args[0].is_Add
+               and len(l.args[0].args) == 2
+               and all(isinstance(t, exp) for t in l.args[0].args)),
+    lambda l: (
+        Max(*[e.args[0] for e in l.args[0].args]) +
+        log1p(exp(Min(*[e.args[0] for e in l.args[0].args])))
+    )
+)
+
+
+class FuncMinusOneOptim(ReplaceOptim):
+    """Specialization of ReplaceOptim for functions evaluating "f(x) - 1".
+
+    Explanation
+    ===========
+
+    Numerical functions which go toward one as x go toward zero is often best
+    implemented by a dedicated function in order to avoid catastrophic
+    cancellation. One such example is ``expm1(x)`` in the C standard library
+    which evaluates ``exp(x) - 1``. Such functions preserves many more
+    significant digits when its argument is much smaller than one, compared
+    to subtracting one afterwards.
+
+    Parameters
+    ==========
+
+    func :
+        The function which is subtracted by one.
+    func_m_1 :
+        The specialized function evaluating ``func(x) - 1``.
+    opportunistic : bool
+        When ``True``, apply the transformation as long as the magnitude of the
+        remaining number terms decreases. When ``False``, only apply the
+        transformation if it completely eliminates the number term.
+
+    Examples
+    ========
+
+    >>> from sympy import symbols, exp
+    >>> from sympy.codegen.rewriting import FuncMinusOneOptim
+    >>> from sympy.codegen.cfunctions import expm1
+    >>> x, y = symbols('x y')
+    >>> expm1_opt = FuncMinusOneOptim(exp, expm1)
+    >>> expm1_opt(exp(x) + 2*exp(5*y) - 3)
+    expm1(x) + 2*expm1(5*y)
+
+
+    """
+
+    def __init__(self, func, func_m_1, opportunistic=True):
+        weight = 10  # <-- this is an arbitrary number (heuristic)
+        super().__init__(lambda e: e.is_Add, self.replace_in_Add,
+                         cost_function=lambda expr: expr.count_ops() - weight*expr.count(func_m_1))
+        self.func = func
+        self.func_m_1 = func_m_1
+        self.opportunistic = opportunistic
+
+    def _group_Add_terms(self, add):
+        numbers, non_num = sift(add.args, lambda arg: arg.is_number, binary=True)
+        numsum = sum(numbers)
+        terms_with_func, other = sift(non_num, lambda arg: arg.has(self.func), binary=True)
+        return numsum, terms_with_func, other
+
+    def replace_in_Add(self, e):
+        """ passed as second argument to Basic.replace(...) """
+        numsum, terms_with_func, other_non_num_terms = self._group_Add_terms(e)
+        if numsum == 0:
+            return e
+        substituted, untouched = [], []
+        for with_func in terms_with_func:
+            if with_func.is_Mul:
+                func, coeff = sift(with_func.args, lambda arg: arg.func == self.func, binary=True)
+                if len(func) == 1 and len(coeff) == 1:
+                    func, coeff = func[0], coeff[0]
+                else:
+                    coeff = None
+            elif with_func.func == self.func:
+                func, coeff = with_func, S.One
+            else:
+                coeff = None
+
+            if coeff is not None and coeff.is_number and sign(coeff) == -sign(numsum):
+                if self.opportunistic:
+                    do_substitute = abs(coeff+numsum) < abs(numsum)
+                else:
+                    do_substitute = coeff+numsum == 0
+
+                if do_substitute:  # advantageous substitution
+                    numsum += coeff
+                    substituted.append(coeff*self.func_m_1(*func.args))
+                    continue
+            untouched.append(with_func)
+
+        return e.func(numsum, *substituted, *untouched, *other_non_num_terms)
+
+    def __call__(self, expr):
+        alt1 = super().__call__(expr)
+        alt2 = super().__call__(expr.factor())
+        return self.cheapest(alt1, alt2)
+
+
+expm1_opt = FuncMinusOneOptim(exp, expm1)
+cosm1_opt = FuncMinusOneOptim(cos, cosm1)
+powm1_opt = FuncMinusOneOptim(Pow, powm1)
+
+log1p_opt = ReplaceOptim(
+    lambda e: isinstance(e, log),
+    lambda l: expand_log(l.replace(
+        log, lambda arg: log(arg.factor())
+    )).replace(log(_u+1), log1p(_u))
+)
+
+def create_expand_pow_optimization(limit, *, base_req=lambda b: b.is_symbol):
+    """ Creates an instance of :class:`ReplaceOptim` for expanding ``Pow``.
+
+    Explanation
+    ===========
+
+    The requirements for expansions are that the base needs to be a symbol
+    and the exponent needs to be an Integer (and be less than or equal to
+    ``limit``).
+
+    Parameters
+    ==========
+
+    limit : int
+         The highest power which is expanded into multiplication.
+    base_req : function returning bool
+         Requirement on base for expansion to happen, default is to return
+         the ``is_symbol`` attribute of the base.
+
+    Examples
+    ========
+
+    >>> from sympy import Symbol, sin
+    >>> from sympy.codegen.rewriting import create_expand_pow_optimization
+    >>> x = Symbol('x')
+    >>> expand_opt = create_expand_pow_optimization(3)
+    >>> expand_opt(x**5 + x**3)
+    x**5 + x*x*x
+    >>> expand_opt(x**5 + x**3 + sin(x)**3)
+    x**5 + sin(x)**3 + x*x*x
+    >>> opt2 = create_expand_pow_optimization(3, base_req=lambda b: not b.is_Function)
+    >>> opt2((x+1)**2 + sin(x)**2)
+    sin(x)**2 + (x + 1)*(x + 1)
+
+    """
+    return ReplaceOptim(
+        lambda e: e.is_Pow and base_req(e.base) and e.exp.is_Integer and abs(e.exp) <= limit,
+        lambda p: (
+            UnevaluatedExpr(Mul(*([p.base]*+p.exp), evaluate=False)) if p.exp > 0 else
+            1/UnevaluatedExpr(Mul(*([p.base]*-p.exp), evaluate=False))
+        ))
+
+# Optimization procedures for turning A**(-1) * x into MatrixSolve(A, x)
+def _matinv_predicate(expr):
+    # TODO: We should be able to support more than 2 elements
+    if expr.is_MatMul and len(expr.args) == 2:
+        left, right = expr.args
+        if left.is_Inverse and right.shape[1] == 1:
+            inv_arg = left.arg
+            if isinstance(inv_arg, MatrixSymbol):
+                return bool(ask(Q.fullrank(left.arg)))
+
+    return False
+
+def _matinv_transform(expr):
+    left, right = expr.args
+    inv_arg = left.arg
+    return MatrixSolve(inv_arg, right)
+
+
+matinv_opt = ReplaceOptim(_matinv_predicate, _matinv_transform)
+
+
+logaddexp_opt = ReplaceOptim(log(exp(_v)+exp(_w)), logaddexp(_v, _w))
+logaddexp2_opt = ReplaceOptim(log(Pow(2, _v)+Pow(2, _w)), logaddexp2(_v, _w)*log(2))
+
+# Collections of optimizations:
+optims_c99 = (expm1_opt, log1p_opt, exp2_opt, log2_opt, log2const_opt)
+
+optims_numpy = optims_c99 + (logaddexp_opt, logaddexp2_opt,) + sinc_opts
+
+optims_scipy = (cosm1_opt, powm1_opt)

rl/Lib/site-packages/sympy/codegen/scipy_nodes.py

@@ -0,0 +1,79 @@
+from sympy.core.function import Add, ArgumentIndexError, Function
+from sympy.core.power import Pow
+from sympy.core.singleton import S
+from sympy.functions.elementary.exponential import log
+from sympy.functions.elementary.trigonometric import cos, sin
+
+
+def _cosm1(x, *, evaluate=True):
+    return Add(cos(x, evaluate=evaluate), -S.One, evaluate=evaluate)
+
+
+class cosm1(Function):
+    """ Minus one plus cosine of x, i.e. cos(x) - 1. For use when x is close to zero.
+
+    Helper class for use with e.g. scipy.special.cosm1
+    See: https://docs.scipy.org/doc/scipy/reference/generated/scipy.special.cosm1.html
+    """
+    nargs = 1
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return -sin(*self.args)
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+    def _eval_rewrite_as_cos(self, x, **kwargs):
+        return _cosm1(x)
+
+    def _eval_evalf(self, *args, **kwargs):
+        return self.rewrite(cos).evalf(*args, **kwargs)
+
+    def _eval_simplify(self, **kwargs):
+        x, = self.args
+        candidate = _cosm1(x.simplify(**kwargs))
+        if candidate != _cosm1(x, evaluate=False):
+            return candidate
+        else:
+            return cosm1(x)
+
+
+def _powm1(x, y, *, evaluate=True):
+    return Add(Pow(x, y, evaluate=evaluate), -S.One, evaluate=evaluate)
+
+
+class powm1(Function):
+    """ Minus one plus x to the power of y, i.e. x**y - 1. For use when x is close to one or y is close to zero.
+
+    Helper class for use with e.g. scipy.special.powm1
+    See: https://docs.scipy.org/doc/scipy/reference/generated/scipy.special.powm1.html
+    """
+    nargs = 2
+
+    def fdiff(self, argindex=1):
+        """
+        Returns the first derivative of this function.
+        """
+        if argindex == 1:
+            return Pow(self.args[0], self.args[1])*self.args[1]/self.args[0]
+        elif argindex == 2:
+            return log(self.args[0])*Pow(*self.args)
+        else:
+            raise ArgumentIndexError(self, argindex)
+
+    def _eval_rewrite_as_Pow(self, x, y, **kwargs):
+        return _powm1(x, y)
+
+    def _eval_evalf(self, *args, **kwargs):
+        return self.rewrite(Pow).evalf(*args, **kwargs)
+
+    def _eval_simplify(self, **kwargs):
+        x, y = self.args
+        candidate = _powm1(x.simplify(**kwargs), y.simplify(**kwargs))
+        if candidate != _powm1(x, y, evaluate=False):
+            return candidate
+        else:
+            return powm1(x, y)

rl/Lib/site-packages/sympy/codegen/tests/test_abstract_nodes.py

@@ -0,0 +1,14 @@
+from sympy.core.symbol import symbols
+from sympy.codegen.abstract_nodes import List
+
+
+def test_List():
+    l = List(2, 3, 4)
+    assert l == List(2, 3, 4)
+    assert str(l) == "[2, 3, 4]"
+    x, y, z = symbols('x y z')
+    l = List(x**2,y**3,z**4)
+    # contrary to python's built-in list, we can call e.g. "replace" on List.
+    m = l.replace(lambda arg: arg.is_Pow and arg.exp>2, lambda p: p.base-p.exp)
+    assert m == [x**2, y-3, z-4]
+    hash(m)

rl/Lib/site-packages/sympy/codegen/tests/test_algorithms.py

@@ -0,0 +1,179 @@
+import tempfile
+from sympy import log, Min, Max, sqrt
+from sympy.core.numbers import Float
+from sympy.core.symbol import Symbol, symbols
+from sympy.functions.elementary.trigonometric import cos
+from sympy.codegen.ast import Assignment, Raise, RuntimeError_, QuotedString
+from sympy.codegen.algorithms import newtons_method, newtons_method_function
+from sympy.codegen.cfunctions import expm1
+from sympy.codegen.fnodes import bind_C
+from sympy.codegen.futils import render_as_module as f_module
+from sympy.codegen.pyutils import render_as_module as py_module
+from sympy.external import import_module
+from sympy.printing.codeprinter import ccode
+from sympy.utilities._compilation import compile_link_import_strings, has_c, has_fortran
+from sympy.utilities._compilation.util import may_xfail
+from sympy.testing.pytest import skip, raises
+
+cython = import_module('cython')
+wurlitzer = import_module('wurlitzer')
+
+def test_newtons_method():
+    x, dx, atol = symbols('x dx atol')
+    expr = cos(x) - x**3
+    algo = newtons_method(expr, x, atol, dx)
+    assert algo.has(Assignment(dx, -expr/expr.diff(x)))
+
+
+@may_xfail
+def test_newtons_method_function__ccode():
+    x = Symbol('x', real=True)
+    expr = cos(x) - x**3
+    func = newtons_method_function(expr, x)
+
+    if not cython:
+        skip("cython not installed.")
+    if not has_c():
+        skip("No C compiler found.")
+
+    compile_kw = {"std": 'c99'}
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = compile_link_import_strings([
+            ('newton.c', ('#include <math.h>\n'
+                          '#include <stdio.h>\n') + ccode(func)),
+            ('_newton.pyx', ("#cython: language_level={}\n".format("3") +
+                             "cdef extern double newton(double)\n"
+                             "def py_newton(x):\n"
+                             "    return newton(x)\n"))
+        ], build_dir=folder, compile_kwargs=compile_kw)
+        assert abs(mod.py_newton(0.5) - 0.865474033102) < 1e-12
+
+
+@may_xfail
+def test_newtons_method_function__fcode():
+    x = Symbol('x', real=True)
+    expr = cos(x) - x**3
+    func = newtons_method_function(expr, x, attrs=[bind_C(name='newton')])
+
+    if not cython:
+        skip("cython not installed.")
+    if not has_fortran():
+        skip("No Fortran compiler found.")
+
+    f_mod = f_module([func], 'mod_newton')
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = compile_link_import_strings([
+            ('newton.f90', f_mod),
+            ('_newton.pyx', ("#cython: language_level={}\n".format("3") +
+                             "cdef extern double newton(double*)\n"
+                             "def py_newton(double x):\n"
+                             "    return newton(&x)\n"))
+        ], build_dir=folder)
+        assert abs(mod.py_newton(0.5) - 0.865474033102) < 1e-12
+
+
+def test_newtons_method_function__pycode():
+    x = Symbol('x', real=True)
+    expr = cos(x) - x**3
+    func = newtons_method_function(expr, x)
+    py_mod = py_module(func)
+    namespace = {}
+    exec(py_mod, namespace, namespace)
+    res = eval('newton(0.5)', namespace)
+    assert abs(res - 0.865474033102) < 1e-12
+
+
+@may_xfail
+def test_newtons_method_function__ccode_parameters():
+    args = x, A, k, p = symbols('x A k p')
+    expr = A*cos(k*x) - p*x**3
+    raises(ValueError, lambda: newtons_method_function(expr, x))
+    use_wurlitzer = wurlitzer
+
+    func = newtons_method_function(expr, x, args, debug=use_wurlitzer)
+
+    if not has_c():
+        skip("No C compiler found.")
+    if not cython:
+        skip("cython not installed.")
+
+    compile_kw = {"std": 'c99'}
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = compile_link_import_strings([
+            ('newton_par.c', ('#include <math.h>\n'
+                          '#include <stdio.h>\n') + ccode(func)),
+            ('_newton_par.pyx', ("#cython: language_level={}\n".format("3") +
+                                 "cdef extern double newton(double, double, double, double)\n"
+                             "def py_newton(x, A=1, k=1, p=1):\n"
+                             "    return newton(x, A, k, p)\n"))
+        ], compile_kwargs=compile_kw, build_dir=folder)
+
+        if use_wurlitzer:
+            with wurlitzer.pipes() as (out, err):
+                result = mod.py_newton(0.5)
+        else:
+            result = mod.py_newton(0.5)
+
+        assert abs(result - 0.865474033102) < 1e-12
+
+        if not use_wurlitzer:
+            skip("C-level output only tested when package 'wurlitzer' is available.")
+
+        out, err = out.read(), err.read()
+        assert err == ''
+        assert out == """\
+x=         0.5
+x=      1.1121 d_x=     0.61214
+x=     0.90967 d_x=    -0.20247
+x=     0.86726 d_x=   -0.042409
+x=     0.86548 d_x=  -0.0017867
+x=     0.86547 d_x= -3.1022e-06
+x=     0.86547 d_x= -9.3421e-12
+x=     0.86547 d_x=  3.6902e-17
+"""  # try to run tests with LC_ALL=C if this assertion fails
+
+
+def test_newtons_method_function__rtol_cse_nan():
+    a, b, c, N_geo, N_tot = symbols('a b c N_geo N_tot', real=True, nonnegative=True)
+    i = Symbol('i', integer=True, nonnegative=True)
+    N_ari = N_tot - N_geo - 1
+    delta_ari = (c-b)/N_ari
+    ln_delta_geo = log(b) + log(-expm1((log(a)-log(b))/N_geo))
+    eqb_log = ln_delta_geo - log(delta_ari)
+
+    def _clamp(low, expr, high):
+        return Min(Max(low, expr), high)
+
+    meth_kw = {
+        'clamped_newton': {'delta_fn': lambda e, x: _clamp(
+            (sqrt(a*x)-x)*0.99,
+            -e/e.diff(x),
+            (sqrt(c*x)-x)*0.99
+        )},
+        'halley': {'delta_fn': lambda e, x: (-2*(e*e.diff(x))/(2*e.diff(x)**2 - e*e.diff(x, 2)))},
+        'halley_alt': {'delta_fn': lambda e, x: (-e/e.diff(x)/(1-e/e.diff(x)*e.diff(x,2)/2/e.diff(x)))},
+    }
+    args = eqb_log, b
+    for use_cse in [False, True]:
+        kwargs = {
+            'params': (b, a, c, N_geo, N_tot), 'itermax': 60, 'debug': True, 'cse': use_cse,
+            'counter': i, 'atol': 1e-100, 'rtol': 2e-16, 'bounds': (a,c),
+            'handle_nan': Raise(RuntimeError_(QuotedString("encountered NaN.")))
+        }
+        func = {k: newtons_method_function(*args, func_name=f"{k}_b", **dict(kwargs, **kw)) for k, kw in meth_kw.items()}
+        py_mod = {k: py_module(v) for k, v in func.items()}
+        namespace = {}
+        root_find_b = {}
+        for k, v in py_mod.items():
+            ns = namespace[k] = {}
+            exec(v, ns, ns)
+            root_find_b[k] = ns[f'{k}_b']
+        ref = Float('13.2261515064168768938151923226496')
+        reftol = {'clamped_newton': 2e-16, 'halley': 2e-16, 'halley_alt': 3e-16}
+        guess = 4.0
+        for meth, func in root_find_b.items():
+            result = func(guess, 1e-2, 1e2, 50, 100)
+            req = ref*reftol[meth]
+            if use_cse:
+                req *= 2
+            assert abs(result - ref) < req

rl/Lib/site-packages/sympy/codegen/tests/test_applications.py

@@ -0,0 +1,57 @@
+# This file contains tests that exercise multiple AST nodes
+
+import tempfile
+
+from sympy.external import import_module
+from sympy.printing.codeprinter import ccode
+from sympy.utilities._compilation import compile_link_import_strings, has_c
+from sympy.utilities._compilation.util import may_xfail
+from sympy.testing.pytest import skip
+from sympy.codegen.ast import (
+    FunctionDefinition, FunctionPrototype, Variable, Pointer, real, Assignment,
+    integer, CodeBlock, While
+)
+from sympy.codegen.cnodes import void, PreIncrement
+from sympy.codegen.cutils import render_as_source_file
+
+cython = import_module('cython')
+np = import_module('numpy')
+
+def _mk_func1():
+    declars = n, inp, out = Variable('n', integer), Pointer('inp', real), Pointer('out', real)
+    i = Variable('i', integer)
+    whl = While(i<n, [Assignment(out[i], inp[i]), PreIncrement(i)])
+    body = CodeBlock(i.as_Declaration(value=0), whl)
+    return FunctionDefinition(void, 'our_test_function', declars, body)
+
+
+def _render_compile_import(funcdef, build_dir):
+    code_str = render_as_source_file(funcdef, settings={"contract": False})
+    declar = ccode(FunctionPrototype.from_FunctionDefinition(funcdef))
+    return compile_link_import_strings([
+        ('our_test_func.c', code_str),
+        ('_our_test_func.pyx', ("#cython: language_level={}\n".format("3") +
+                                "cdef extern {declar}\n"
+                                "def _{fname}({typ}[:] inp, {typ}[:] out):\n"
+                                "    {fname}(inp.size, &inp[0], &out[0])").format(
+                                    declar=declar, fname=funcdef.name, typ='double'
+                                ))
+    ], build_dir=build_dir)
+
+
+@may_xfail
+def test_copying_function():
+    if not np:
+        skip("numpy not installed.")
+    if not has_c():
+        skip("No C compiler found.")
+    if not cython:
+        skip("Cython not found.")
+
+    info = None
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = _render_compile_import(_mk_func1(), build_dir=folder)
+        inp = np.arange(10.0)
+        out = np.empty_like(inp)
+        mod._our_test_function(inp, out)
+        assert np.allclose(inp, out)

rl/Lib/site-packages/sympy/codegen/tests/test_approximations.py

@@ -0,0 +1,53 @@
+import math
+from sympy.core.symbol import symbols
+from sympy.functions.elementary.exponential import exp
+from sympy.codegen.rewriting import optimize
+from sympy.codegen.approximations import SumApprox, SeriesApprox
+
+
+def test_SumApprox_trivial():
+    x = symbols('x')
+    expr1 = 1 + x
+    sum_approx = SumApprox(bounds={x: (-1e-20, 1e-20)}, reltol=1e-16)
+    apx1 = optimize(expr1, [sum_approx])
+    assert apx1 - 1 == 0
+
+
+def test_SumApprox_monotone_terms():
+    x, y, z = symbols('x y z')
+    expr1 = exp(z)*(x**2 + y**2 + 1)
+    bnds1 = {x: (0, 1e-3), y: (100, 1000)}
+    sum_approx_m2 = SumApprox(bounds=bnds1, reltol=1e-2)
+    sum_approx_m5 = SumApprox(bounds=bnds1, reltol=1e-5)
+    sum_approx_m11 = SumApprox(bounds=bnds1, reltol=1e-11)
+    assert (optimize(expr1, [sum_approx_m2])/exp(z) - (y**2)).simplify() == 0
+    assert (optimize(expr1, [sum_approx_m5])/exp(z) - (y**2 + 1)).simplify() == 0
+    assert (optimize(expr1, [sum_approx_m11])/exp(z) - (y**2 + 1 + x**2)).simplify() == 0
+
+
+def test_SeriesApprox_trivial():
+    x, z = symbols('x z')
+    for factor in [1, exp(z)]:
+        x = symbols('x')
+        expr1 = exp(x)*factor
+        bnds1 = {x: (-1, 1)}
+        series_approx_50 = SeriesApprox(bounds=bnds1, reltol=0.50)
+        series_approx_10 = SeriesApprox(bounds=bnds1, reltol=0.10)
+        series_approx_05 = SeriesApprox(bounds=bnds1, reltol=0.05)
+        c = (bnds1[x][1] + bnds1[x][0])/2  # 0.0
+        f0 = math.exp(c)  # 1.0
+
+        ref_50 = f0 + x + x**2/2
+        ref_10 = f0 + x + x**2/2 + x**3/6
+        ref_05 = f0 + x + x**2/2 + x**3/6 + x**4/24
+
+        res_50 = optimize(expr1, [series_approx_50])
+        res_10 = optimize(expr1, [series_approx_10])
+        res_05 = optimize(expr1, [series_approx_05])
+
+        assert (res_50/factor - ref_50).simplify() == 0
+        assert (res_10/factor - ref_10).simplify() == 0
+        assert (res_05/factor - ref_05).simplify() == 0
+
+        max_ord3 = SeriesApprox(bounds=bnds1, reltol=0.05, max_order=3)
+        assert optimize(expr1, [max_ord3]) == expr1

rl/Lib/site-packages/sympy/codegen/tests/test_ast.py

@@ -0,0 +1,661 @@
+import math
+from sympy.core.containers import Tuple
+from sympy.core.numbers import nan, oo, Float, Integer
+from sympy.core.relational import Lt
+from sympy.core.symbol import symbols, Symbol
+from sympy.functions.elementary.trigonometric import sin
+from sympy.matrices.dense import Matrix
+from sympy.matrices.expressions.matexpr import MatrixSymbol
+from sympy.sets.fancysets import Range
+from sympy.tensor.indexed import Idx, IndexedBase
+from sympy.testing.pytest import raises
+
+
+from sympy.codegen.ast import (
+    Assignment, Attribute, aug_assign, CodeBlock, For, Type, Variable, Pointer, Declaration,
+    AddAugmentedAssignment, SubAugmentedAssignment, MulAugmentedAssignment,
+    DivAugmentedAssignment, ModAugmentedAssignment, value_const, pointer_const,
+    integer, real, complex_, int8, uint8, float16 as f16, float32 as f32,
+    float64 as f64, float80 as f80, float128 as f128, complex64 as c64, complex128 as c128,
+    While, Scope, String, Print, QuotedString, FunctionPrototype, FunctionDefinition, Return,
+    FunctionCall, untyped, IntBaseType, intc, Node, none, NoneToken, Token, Comment
+)
+
+x, y, z, t, x0, x1, x2, a, b = symbols("x, y, z, t, x0, x1, x2, a, b")
+n = symbols("n", integer=True)
+A = MatrixSymbol('A', 3, 1)
+mat = Matrix([1, 2, 3])
+B = IndexedBase('B')
+i = Idx("i", n)
+A22 = MatrixSymbol('A22',2,2)
+B22 = MatrixSymbol('B22',2,2)
+
+
+def test_Assignment():
+    # Here we just do things to show they don't error
+    Assignment(x, y)
+    Assignment(x, 0)
+    Assignment(A, mat)
+    Assignment(A[1,0], 0)
+    Assignment(A[1,0], x)
+    Assignment(B[i], x)
+    Assignment(B[i], 0)
+    a = Assignment(x, y)
+    assert a.func(*a.args) == a
+    assert a.op == ':='
+    # Here we test things to show that they error
+    # Matrix to scalar
+    raises(ValueError, lambda: Assignment(B[i], A))
+    raises(ValueError, lambda: Assignment(B[i], mat))
+    raises(ValueError, lambda: Assignment(x, mat))
+    raises(ValueError, lambda: Assignment(x, A))
+    raises(ValueError, lambda: Assignment(A[1,0], mat))
+    # Scalar to matrix
+    raises(ValueError, lambda: Assignment(A, x))
+    raises(ValueError, lambda: Assignment(A, 0))
+    # Non-atomic lhs
+    raises(TypeError, lambda: Assignment(mat, A))
+    raises(TypeError, lambda: Assignment(0, x))
+    raises(TypeError, lambda: Assignment(x*x, 1))
+    raises(TypeError, lambda: Assignment(A + A, mat))
+    raises(TypeError, lambda: Assignment(B, 0))
+
+
+def test_AugAssign():
+    # Here we just do things to show they don't error
+    aug_assign(x, '+', y)
+    aug_assign(x, '+', 0)
+    aug_assign(A, '+', mat)
+    aug_assign(A[1, 0], '+', 0)
+    aug_assign(A[1, 0], '+', x)
+    aug_assign(B[i], '+', x)
+    aug_assign(B[i], '+', 0)
+
+    # Check creation via aug_assign vs constructor
+    for binop, cls in [
+            ('+', AddAugmentedAssignment),
+            ('-', SubAugmentedAssignment),
+            ('*', MulAugmentedAssignment),
+            ('/', DivAugmentedAssignment),
+            ('%', ModAugmentedAssignment),
+        ]:
+        a = aug_assign(x, binop, y)
+        b = cls(x, y)
+        assert a.func(*a.args) == a == b
+        assert a.binop == binop
+        assert a.op == binop + '='
+
+    # Here we test things to show that they error
+    # Matrix to scalar
+    raises(ValueError, lambda: aug_assign(B[i], '+', A))
+    raises(ValueError, lambda: aug_assign(B[i], '+', mat))
+    raises(ValueError, lambda: aug_assign(x, '+', mat))
+    raises(ValueError, lambda: aug_assign(x, '+', A))
+    raises(ValueError, lambda: aug_assign(A[1, 0], '+', mat))
+    # Scalar to matrix
+    raises(ValueError, lambda: aug_assign(A, '+', x))
+    raises(ValueError, lambda: aug_assign(A, '+', 0))
+    # Non-atomic lhs
+    raises(TypeError, lambda: aug_assign(mat, '+', A))
+    raises(TypeError, lambda: aug_assign(0, '+', x))
+    raises(TypeError, lambda: aug_assign(x * x, '+', 1))
+    raises(TypeError, lambda: aug_assign(A + A, '+', mat))
+    raises(TypeError, lambda: aug_assign(B, '+', 0))
+
+
+def test_Assignment_printing():
+    assignment_classes = [
+        Assignment,
+        AddAugmentedAssignment,
+        SubAugmentedAssignment,
+        MulAugmentedAssignment,
+        DivAugmentedAssignment,
+        ModAugmentedAssignment,
+    ]
+    pairs = [
+        (x, 2 * y + 2),
+        (B[i], x),
+        (A22, B22),
+        (A[0, 0], x),
+    ]
+
+    for cls in assignment_classes:
+        for lhs, rhs in pairs:
+            a = cls(lhs, rhs)
+            assert repr(a) == '%s(%s, %s)' % (cls.__name__, repr(lhs), repr(rhs))
+
+
+def test_CodeBlock():
+    c = CodeBlock(Assignment(x, 1), Assignment(y, x + 1))
+    assert c.func(*c.args) == c
+
+    assert c.left_hand_sides == Tuple(x, y)
+    assert c.right_hand_sides == Tuple(1, x + 1)
+
+def test_CodeBlock_topological_sort():
+    assignments = [
+        Assignment(x, y + z),
+        Assignment(z, 1),
+        Assignment(t, x),
+        Assignment(y, 2),
+        ]
+
+    ordered_assignments = [
+        # Note that the unrelated z=1 and y=2 are kept in that order
+        Assignment(z, 1),
+        Assignment(y, 2),
+        Assignment(x, y + z),
+        Assignment(t, x),
+        ]
+    c1 = CodeBlock.topological_sort(assignments)
+    assert c1 == CodeBlock(*ordered_assignments)
+
+    # Cycle
+    invalid_assignments = [
+        Assignment(x, y + z),
+        Assignment(z, 1),
+        Assignment(y, x),
+        Assignment(y, 2),
+        ]
+
+    raises(ValueError, lambda: CodeBlock.topological_sort(invalid_assignments))
+
+    # Free symbols
+    free_assignments = [
+        Assignment(x, y + z),
+        Assignment(z, a * b),
+        Assignment(t, x),
+        Assignment(y, b + 3),
+        ]
+
+    free_assignments_ordered = [
+        Assignment(z, a * b),
+        Assignment(y, b + 3),
+        Assignment(x, y + z),
+        Assignment(t, x),
+        ]
+
+    c2 = CodeBlock.topological_sort(free_assignments)
+    assert c2 == CodeBlock(*free_assignments_ordered)
+
+def test_CodeBlock_free_symbols():
+    c1 = CodeBlock(
+        Assignment(x, y + z),
+        Assignment(z, 1),
+        Assignment(t, x),
+        Assignment(y, 2),
+        )
+    assert c1.free_symbols == set()
+
+    c2 = CodeBlock(
+        Assignment(x, y + z),
+        Assignment(z, a * b),
+        Assignment(t, x),
+        Assignment(y, b + 3),
+    )
+    assert c2.free_symbols == {a, b}
+
+def test_CodeBlock_cse():
+    c1 = CodeBlock(
+        Assignment(y, 1),
+        Assignment(x, sin(y)),
+        Assignment(z, sin(y)),
+        Assignment(t, x*z),
+        )
+    assert c1.cse() == CodeBlock(
+        Assignment(y, 1),
+        Assignment(x0, sin(y)),
+        Assignment(x, x0),
+        Assignment(z, x0),
+        Assignment(t, x*z),
+    )
+
+    # Multiple assignments to same symbol not supported
+    raises(NotImplementedError, lambda: CodeBlock(
+        Assignment(x, 1),
+        Assignment(y, 1), Assignment(y, 2)
+    ).cse())
+
+    # Check auto-generated symbols do not collide with existing ones
+    c2 = CodeBlock(
+        Assignment(x0, sin(y) + 1),
+        Assignment(x1, 2 * sin(y)),
+        Assignment(z, x * y),
+        )
+    assert c2.cse() == CodeBlock(
+        Assignment(x2, sin(y)),
+        Assignment(x0, x2 + 1),
+        Assignment(x1, 2 * x2),
+        Assignment(z, x * y),
+        )
+
+
+def test_CodeBlock_cse__issue_14118():
+    # see https://github.com/sympy/sympy/issues/14118
+    c = CodeBlock(
+        Assignment(A22, Matrix([[x, sin(y)],[3, 4]])),
+        Assignment(B22, Matrix([[sin(y), 2*sin(y)], [sin(y)**2, 7]]))
+    )
+    assert c.cse() == CodeBlock(
+        Assignment(x0, sin(y)),
+        Assignment(A22, Matrix([[x, x0],[3, 4]])),
+        Assignment(B22, Matrix([[x0, 2*x0], [x0**2, 7]]))
+    )
+
+def test_For():
+    f = For(n, Range(0, 3), (Assignment(A[n, 0], x + n), aug_assign(x, '+', y)))
+    f = For(n, (1, 2, 3, 4, 5), (Assignment(A[n, 0], x + n),))
+    assert f.func(*f.args) == f
+    raises(TypeError, lambda: For(n, x, (x + y,)))
+
+
+def test_none():
+    assert none.is_Atom
+    assert none == none
+    class Foo(Token):
+        pass
+    foo = Foo()
+    assert foo != none
+    assert none == None
+    assert none == NoneToken()
+    assert none.func(*none.args) == none
+
+
+def test_String():
+    st = String('foobar')
+    assert st.is_Atom
+    assert st == String('foobar')
+    assert st.text == 'foobar'
+    assert st.func(**st.kwargs()) == st
+    assert st.func(*st.args) == st
+
+
+    class Signifier(String):
+        pass
+
+    si = Signifier('foobar')
+    assert si != st
+    assert si.text == st.text
+    s = String('foo')
+    assert str(s) == 'foo'
+    assert repr(s) == "String('foo')"
+
+def test_Comment():
+    c = Comment('foobar')
+    assert c.text == 'foobar'
+    assert str(c) == 'foobar'
+
+def test_Node():
+    n = Node()
+    assert n == Node()
+    assert n.func(*n.args) == n
+
+
+def test_Type():
+    t = Type('MyType')
+    assert len(t.args) == 1
+    assert t.name == String('MyType')
+    assert str(t) == 'MyType'
+    assert repr(t) == "Type(String('MyType'))"
+    assert Type(t) == t
+    assert t.func(*t.args) == t
+    t1 = Type('t1')
+    t2 = Type('t2')
+    assert t1 != t2
+    assert t1 == t1 and t2 == t2
+    t1b = Type('t1')
+    assert t1 == t1b
+    assert t2 != t1b
+
+
+def test_Type__from_expr():
+    assert Type.from_expr(i) == integer
+    u = symbols('u', real=True)
+    assert Type.from_expr(u) == real
+    assert Type.from_expr(n) == integer
+    assert Type.from_expr(3) == integer
+    assert Type.from_expr(3.0) == real
+    assert Type.from_expr(3+1j) == complex_
+    raises(ValueError, lambda: Type.from_expr(sum))
+
+
+def test_Type__cast_check__integers():
+    # Rounding
+    raises(ValueError, lambda: integer.cast_check(3.5))
+    assert integer.cast_check('3') == 3
+    assert integer.cast_check(Float('3.0000000000000000000')) == 3
+    assert integer.cast_check(Float('3.0000000000000000001')) == 3  # unintuitive maybe?
+
+    # Range
+    assert int8.cast_check(127.0) == 127
+    raises(ValueError, lambda: int8.cast_check(128))
+    assert int8.cast_check(-128) == -128
+    raises(ValueError, lambda: int8.cast_check(-129))
+
+    assert uint8.cast_check(0) == 0
+    assert uint8.cast_check(128) == 128
+    raises(ValueError, lambda: uint8.cast_check(256.0))
+    raises(ValueError, lambda: uint8.cast_check(-1))
+
+def test_Attribute():
+    noexcept = Attribute('noexcept')
+    assert noexcept == Attribute('noexcept')
+    alignas16 = Attribute('alignas', [16])
+    alignas32 = Attribute('alignas', [32])
+    assert alignas16 != alignas32
+    assert alignas16.func(*alignas16.args) == alignas16
+
+
+def test_Variable():
+    v = Variable(x, type=real)
+    assert v == Variable(v)
+    assert v == Variable('x', type=real)
+    assert v.symbol == x
+    assert v.type == real
+    assert value_const not in v.attrs
+    assert v.func(*v.args) == v
+    assert str(v) == 'Variable(x, type=real)'
+
+    w = Variable(y, f32, attrs={value_const})
+    assert w.symbol == y
+    assert w.type == f32
+    assert value_const in w.attrs
+    assert w.func(*w.args) == w
+
+    v_n = Variable(n, type=Type.from_expr(n))
+    assert v_n.type == integer
+    assert v_n.func(*v_n.args) == v_n
+    v_i = Variable(i, type=Type.from_expr(n))
+    assert v_i.type == integer
+    assert v_i != v_n
+
+    a_i = Variable.deduced(i)
+    assert a_i.type == integer
+    assert Variable.deduced(Symbol('x', real=True)).type == real
+    assert a_i.func(*a_i.args) == a_i
+
+    v_n2 = Variable.deduced(n, value=3.5, cast_check=False)
+    assert v_n2.func(*v_n2.args) == v_n2
+    assert abs(v_n2.value - 3.5) < 1e-15
+    raises(ValueError, lambda: Variable.deduced(n, value=3.5, cast_check=True))
+
+    v_n3 = Variable.deduced(n)
+    assert v_n3.type == integer
+    assert str(v_n3) == 'Variable(n, type=integer)'
+    assert Variable.deduced(z, value=3).type == integer
+    assert Variable.deduced(z, value=3.0).type == real
+    assert Variable.deduced(z, value=3.0+1j).type == complex_
+
+
+def test_Pointer():
+    p = Pointer(x)
+    assert p.symbol == x
+    assert p.type == untyped
+    assert value_const not in p.attrs
+    assert pointer_const not in p.attrs
+    assert p.func(*p.args) == p
+
+    u = symbols('u', real=True)
+    pu = Pointer(u, type=Type.from_expr(u), attrs={value_const, pointer_const})
+    assert pu.symbol is u
+    assert pu.type == real
+    assert value_const in pu.attrs
+    assert pointer_const in pu.attrs
+    assert pu.func(*pu.args) == pu
+
+    i = symbols('i', integer=True)
+    deref = pu[i]
+    assert deref.indices == (i,)
+
+
+def test_Declaration():
+    u = symbols('u', real=True)
+    vu = Variable(u, type=Type.from_expr(u))
+    assert Declaration(vu).variable.type == real
+    vn = Variable(n, type=Type.from_expr(n))
+    assert Declaration(vn).variable.type == integer
+
+    # PR 19107, does not allow comparison between expressions and Basic
+    # lt = StrictLessThan(vu, vn)
+    # assert isinstance(lt, StrictLessThan)
+
+    vuc = Variable(u, Type.from_expr(u), value=3.0, attrs={value_const})
+    assert value_const in vuc.attrs
+    assert pointer_const not in vuc.attrs
+    decl = Declaration(vuc)
+    assert decl.variable == vuc
+    assert isinstance(decl.variable.value, Float)
+    assert decl.variable.value == 3.0
+    assert decl.func(*decl.args) == decl
+    assert vuc.as_Declaration() == decl
+    assert vuc.as_Declaration(value=None, attrs=None) == Declaration(vu)
+
+    vy = Variable(y, type=integer, value=3)
+    decl2 = Declaration(vy)
+    assert decl2.variable == vy
+    assert decl2.variable.value == Integer(3)
+
+    vi = Variable(i, type=Type.from_expr(i), value=3.0)
+    decl3 = Declaration(vi)
+    assert decl3.variable.type == integer
+    assert decl3.variable.value == 3.0
+
+    raises(ValueError, lambda: Declaration(vi, 42))
+
+
+def test_IntBaseType():
+    assert intc.name == String('intc')
+    assert intc.args == (intc.name,)
+    assert str(IntBaseType('a').name) == 'a'
+
+
+def test_FloatType():
+    assert f16.dig == 3
+    assert f32.dig == 6
+    assert f64.dig == 15
+    assert f80.dig == 18
+    assert f128.dig == 33
+
+    assert f16.decimal_dig == 5
+    assert f32.decimal_dig == 9
+    assert f64.decimal_dig == 17
+    assert f80.decimal_dig == 21
+    assert f128.decimal_dig == 36
+
+    assert f16.max_exponent == 16
+    assert f32.max_exponent == 128
+    assert f64.max_exponent == 1024
+    assert f80.max_exponent == 16384
+    assert f128.max_exponent == 16384
+
+    assert f16.min_exponent == -13
+    assert f32.min_exponent == -125
+    assert f64.min_exponent == -1021
+    assert f80.min_exponent == -16381
+    assert f128.min_exponent == -16381
+
+    assert abs(f16.eps / Float('0.00097656', precision=16) - 1) < 0.1*10**-f16.dig
+    assert abs(f32.eps / Float('1.1920929e-07', precision=32) - 1) < 0.1*10**-f32.dig
+    assert abs(f64.eps / Float('2.2204460492503131e-16', precision=64) - 1) < 0.1*10**-f64.dig
+    assert abs(f80.eps / Float('1.08420217248550443401e-19', precision=80) - 1) < 0.1*10**-f80.dig
+    assert abs(f128.eps / Float(' 1.92592994438723585305597794258492732e-34', precision=128) - 1) < 0.1*10**-f128.dig
+
+    assert abs(f16.max / Float('65504', precision=16) - 1) < .1*10**-f16.dig
+    assert abs(f32.max / Float('3.40282347e+38', precision=32) - 1) < 0.1*10**-f32.dig
+    assert abs(f64.max / Float('1.79769313486231571e+308', precision=64) - 1) < 0.1*10**-f64.dig  # cf. np.finfo(np.float64).max
+    assert abs(f80.max / Float('1.18973149535723176502e+4932', precision=80) - 1) < 0.1*10**-f80.dig
+    assert abs(f128.max / Float('1.18973149535723176508575932662800702e+4932', precision=128) - 1) < 0.1*10**-f128.dig
+
+    # cf. np.finfo(np.float32).tiny
+    assert abs(f16.tiny / Float('6.1035e-05', precision=16) - 1) < 0.1*10**-f16.dig
+    assert abs(f32.tiny / Float('1.17549435e-38', precision=32) - 1) < 0.1*10**-f32.dig
+    assert abs(f64.tiny / Float('2.22507385850720138e-308', precision=64) - 1) < 0.1*10**-f64.dig
+    assert abs(f80.tiny / Float('3.36210314311209350626e-4932', precision=80) - 1) < 0.1*10**-f80.dig
+    assert abs(f128.tiny / Float('3.3621031431120935062626778173217526e-4932', precision=128) - 1) < 0.1*10**-f128.dig
+
+    assert f64.cast_check(0.5) == Float(0.5, 17)
+    assert abs(f64.cast_check(3.7) - 3.7) < 3e-17
+    assert isinstance(f64.cast_check(3), (Float, float))
+
+    assert f64.cast_nocheck(oo) == float('inf')
+    assert f64.cast_nocheck(-oo) == float('-inf')
+    assert f64.cast_nocheck(float(oo)) == float('inf')
+    assert f64.cast_nocheck(float(-oo)) == float('-inf')
+    assert math.isnan(f64.cast_nocheck(nan))
+
+    assert f32 != f64
+    assert f64 == f64.func(*f64.args)
+
+
+def test_Type__cast_check__floating_point():
+    raises(ValueError, lambda: f32.cast_check(123.45678949))
+    raises(ValueError, lambda: f32.cast_check(12.345678949))
+    raises(ValueError, lambda: f32.cast_check(1.2345678949))
+    raises(ValueError, lambda: f32.cast_check(.12345678949))
+    assert abs(123.456789049 - f32.cast_check(123.456789049) - 4.9e-8) < 1e-8
+    assert abs(0.12345678904 - f32.cast_check(0.12345678904) - 4e-11) < 1e-11
+
+    dcm21 = Float('0.123456789012345670499')  # 21 decimals
+    assert abs(dcm21 - f64.cast_check(dcm21) - 4.99e-19) < 1e-19
+
+    f80.cast_check(Float('0.12345678901234567890103', precision=88))
+    raises(ValueError, lambda: f80.cast_check(Float('0.12345678901234567890149', precision=88)))
+
+    v10 = 12345.67894
+    raises(ValueError, lambda: f32.cast_check(v10))
+    assert abs(Float(str(v10), precision=64+8) - f64.cast_check(v10)) < v10*1e-16
+
+    assert abs(f32.cast_check(2147483647) - 2147483650) < 1
+
+
+def test_Type__cast_check__complex_floating_point():
+    val9_11 = 123.456789049 + 0.123456789049j
+    raises(ValueError, lambda: c64.cast_check(.12345678949 + .12345678949j))
+    assert abs(val9_11 - c64.cast_check(val9_11) - 4.9e-8) < 1e-8
+
+    dcm21 = Float('0.123456789012345670499') + 1e-20j  # 21 decimals
+    assert abs(dcm21 - c128.cast_check(dcm21) - 4.99e-19) < 1e-19
+    v19 = Float('0.1234567890123456749') + 1j*Float('0.1234567890123456749')
+    raises(ValueError, lambda: c128.cast_check(v19))
+
+
+def test_While():
+    xpp = AddAugmentedAssignment(x, 1)
+    whl1 = While(x < 2, [xpp])
+    assert whl1.condition.args[0] == x
+    assert whl1.condition.args[1] == 2
+    assert whl1.condition == Lt(x, 2, evaluate=False)
+    assert whl1.body.args == (xpp,)
+    assert whl1.func(*whl1.args) == whl1
+
+    cblk = CodeBlock(AddAugmentedAssignment(x, 1))
+    whl2 = While(x < 2, cblk)
+    assert whl1 == whl2
+    assert whl1 != While(x < 3, [xpp])
+
+
+def test_Scope():
+    assign = Assignment(x, y)
+    incr = AddAugmentedAssignment(x, 1)
+    scp = Scope([assign, incr])
+    cblk = CodeBlock(assign, incr)
+    assert scp.body == cblk
+    assert scp == Scope(cblk)
+    assert scp != Scope([incr, assign])
+    assert scp.func(*scp.args) == scp
+
+
+def test_Print():
+    fmt = "%d %.3f"
+    ps = Print([n, x], fmt)
+    assert str(ps.format_string) == fmt
+    assert ps.print_args == Tuple(n, x)
+    assert ps.args == (Tuple(n, x), QuotedString(fmt), none)
+    assert ps == Print((n, x), fmt)
+    assert ps != Print([x, n], fmt)
+    assert ps.func(*ps.args) == ps
+
+    ps2 = Print([n, x])
+    assert ps2 == Print([n, x])
+    assert ps2 != ps
+    assert ps2.format_string == None
+
+
+def test_FunctionPrototype_and_FunctionDefinition():
+    vx = Variable(x, type=real)
+    vn = Variable(n, type=integer)
+    fp1 = FunctionPrototype(real, 'power', [vx, vn])
+    assert fp1.return_type == real
+    assert fp1.name == String('power')
+    assert fp1.parameters == Tuple(vx, vn)
+    assert fp1 == FunctionPrototype(real, 'power', [vx, vn])
+    assert fp1 != FunctionPrototype(real, 'power', [vn, vx])
+    assert fp1.func(*fp1.args) == fp1
+
+
+    body = [Assignment(x, x**n), Return(x)]
+    fd1 = FunctionDefinition(real, 'power', [vx, vn], body)
+    assert fd1.return_type == real
+    assert str(fd1.name) == 'power'
+    assert fd1.parameters == Tuple(vx, vn)
+    assert fd1.body == CodeBlock(*body)
+    assert fd1 == FunctionDefinition(real, 'power', [vx, vn], body)
+    assert fd1 != FunctionDefinition(real, 'power', [vx, vn], body[::-1])
+    assert fd1.func(*fd1.args) == fd1
+
+    fp2 = FunctionPrototype.from_FunctionDefinition(fd1)
+    assert fp2 == fp1
+
+    fd2 = FunctionDefinition.from_FunctionPrototype(fp1, body)
+    assert fd2 == fd1
+
+
+def test_Return():
+    rs = Return(x)
+    assert rs.args == (x,)
+    assert rs == Return(x)
+    assert rs != Return(y)
+    assert rs.func(*rs.args) == rs
+
+
+def test_FunctionCall():
+    fc = FunctionCall('power', (x, 3))
+    assert fc.function_args[0] == x
+    assert fc.function_args[1] == 3
+    assert len(fc.function_args) == 2
+    assert isinstance(fc.function_args[1], Integer)
+    assert fc == FunctionCall('power', (x, 3))
+    assert fc != FunctionCall('power', (3, x))
+    assert fc != FunctionCall('Power', (x, 3))
+    assert fc.func(*fc.args) == fc
+
+    fc2 = FunctionCall('fma', [2, 3, 4])
+    assert len(fc2.function_args) == 3
+    assert fc2.function_args[0] == 2
+    assert fc2.function_args[1] == 3
+    assert fc2.function_args[2] == 4
+    assert str(fc2) in ( # not sure if QuotedString is a better default...
+        'FunctionCall(fma, function_args=(2, 3, 4))',
+        'FunctionCall("fma", function_args=(2, 3, 4))',
+    )
+
+def test_ast_replace():
+    x = Variable('x', real)
+    y = Variable('y', real)
+    n = Variable('n', integer)
+
+    pwer = FunctionDefinition(real, 'pwer', [x, n], [pow(x.symbol, n.symbol)])
+    pname = pwer.name
+    pcall = FunctionCall('pwer', [y, 3])
+
+    tree1 = CodeBlock(pwer, pcall)
+    assert str(tree1.args[0].name) == 'pwer'
+    assert str(tree1.args[1].name) == 'pwer'
+    for a, b in zip(tree1, [pwer, pcall]):
+        assert a == b
+
+    tree2 = tree1.replace(pname, String('power'))
+    assert str(tree1.args[0].name) == 'pwer'
+    assert str(tree1.args[1].name) == 'pwer'
+    assert str(tree2.args[0].name) == 'power'
+    assert str(tree2.args[1].name) == 'power'

rl/Lib/site-packages/sympy/codegen/tests/test_cfunctions.py

@@ -0,0 +1,165 @@
+from sympy.core.numbers import (Rational, pi)
+from sympy.core.singleton import S
+from sympy.core.symbol import (Symbol, symbols)
+from sympy.functions.elementary.exponential import (exp, log)
+from sympy.codegen.cfunctions import (
+    expm1, log1p, exp2, log2, fma, log10, Sqrt, Cbrt, hypot
+)
+from sympy.core.function import expand_log
+
+
+def test_expm1():
+    # Eval
+    assert expm1(0) == 0
+
+    x = Symbol('x', real=True)
+
+    # Expand and rewrite
+    assert expm1(x).expand(func=True) - exp(x) == -1
+    assert expm1(x).rewrite('tractable') - exp(x) == -1
+    assert expm1(x).rewrite('exp') - exp(x) == -1
+
+    # Precision
+    assert not ((exp(1e-10).evalf() - 1) - 1e-10 - 5e-21) < 1e-22  # for comparison
+    assert abs(expm1(1e-10).evalf() - 1e-10 - 5e-21) < 1e-22
+
+    # Properties
+    assert expm1(x).is_real
+    assert expm1(x).is_finite
+
+    # Diff
+    assert expm1(42*x).diff(x) - 42*exp(42*x) == 0
+    assert expm1(42*x).diff(x) - expm1(42*x).expand(func=True).diff(x) == 0
+
+
+def test_log1p():
+    # Eval
+    assert log1p(0) == 0
+    d = S(10)
+    assert expand_log(log1p(d**-1000) - log(d**1000 + 1) + log(d**1000)) == 0
+
+    x = Symbol('x', real=True)
+
+    # Expand and rewrite
+    assert log1p(x).expand(func=True) - log(x + 1) == 0
+    assert log1p(x).rewrite('tractable') - log(x + 1) == 0
+    assert log1p(x).rewrite('log') - log(x + 1) == 0
+
+    # Precision
+    assert not abs(log(1e-99 + 1).evalf() - 1e-99) < 1e-100  # for comparison
+    assert abs(expand_log(log1p(1e-99)).evalf() - 1e-99) < 1e-100
+
+    # Properties
+    assert log1p(-2**Rational(-1, 2)).is_real
+
+    assert not log1p(-1).is_finite
+    assert log1p(pi).is_finite
+
+    assert not log1p(x).is_positive
+    assert log1p(Symbol('y', positive=True)).is_positive
+
+    assert not log1p(x).is_zero
+    assert log1p(Symbol('z', zero=True)).is_zero
+
+    assert not log1p(x).is_nonnegative
+    assert log1p(Symbol('o', nonnegative=True)).is_nonnegative
+
+    # Diff
+    assert log1p(42*x).diff(x) - 42/(42*x + 1) == 0
+    assert log1p(42*x).diff(x) - log1p(42*x).expand(func=True).diff(x) == 0
+
+
+def test_exp2():
+    # Eval
+    assert exp2(2) == 4
+
+    x = Symbol('x', real=True)
+
+    # Expand
+    assert exp2(x).expand(func=True) - 2**x == 0
+
+    # Diff
+    assert exp2(42*x).diff(x) - 42*exp2(42*x)*log(2) == 0
+    assert exp2(42*x).diff(x) - exp2(42*x).diff(x) == 0
+
+
+def test_log2():
+    # Eval
+    assert log2(8) == 3
+    assert log2(pi) != log(pi)/log(2)  # log2 should *save* (CPU) instructions
+
+    x = Symbol('x', real=True)
+    assert log2(x) != log(x)/log(2)
+    assert log2(2**x) == x
+
+    # Expand
+    assert log2(x).expand(func=True) - log(x)/log(2) == 0
+
+    # Diff
+    assert log2(42*x).diff() - 1/(log(2)*x) == 0
+    assert log2(42*x).diff() - log2(42*x).expand(func=True).diff(x) == 0
+
+
+def test_fma():
+    x, y, z = symbols('x y z')
+
+    # Expand
+    assert fma(x, y, z).expand(func=True) - x*y - z == 0
+
+    expr = fma(17*x, 42*y, 101*z)
+
+    # Diff
+    assert expr.diff(x) - expr.expand(func=True).diff(x) == 0
+    assert expr.diff(y) - expr.expand(func=True).diff(y) == 0
+    assert expr.diff(z) - expr.expand(func=True).diff(z) == 0
+
+    assert expr.diff(x) - 17*42*y == 0
+    assert expr.diff(y) - 17*42*x == 0
+    assert expr.diff(z) - 101 == 0
+
+
+def test_log10():
+    x = Symbol('x')
+
+    # Expand
+    assert log10(x).expand(func=True) - log(x)/log(10) == 0
+
+    # Diff
+    assert log10(42*x).diff(x) - 1/(log(10)*x) == 0
+    assert log10(42*x).diff(x) - log10(42*x).expand(func=True).diff(x) == 0
+
+
+def test_Cbrt():
+    x = Symbol('x')
+
+    # Expand
+    assert Cbrt(x).expand(func=True) - x**Rational(1, 3) == 0
+
+    # Diff
+    assert Cbrt(42*x).diff(x) - 42*(42*x)**(Rational(1, 3) - 1)/3 == 0
+    assert Cbrt(42*x).diff(x) - Cbrt(42*x).expand(func=True).diff(x) == 0
+
+
+def test_Sqrt():
+    x = Symbol('x')
+
+    # Expand
+    assert Sqrt(x).expand(func=True) - x**S.Half == 0
+
+    # Diff
+    assert Sqrt(42*x).diff(x) - 42*(42*x)**(S.Half - 1)/2 == 0
+    assert Sqrt(42*x).diff(x) - Sqrt(42*x).expand(func=True).diff(x) == 0
+
+
+def test_hypot():
+    x, y = symbols('x y')
+
+    # Expand
+    assert hypot(x, y).expand(func=True) - (x**2 + y**2)**S.Half == 0
+
+    # Diff
+    assert hypot(17*x, 42*y).diff(x).expand(func=True) - hypot(17*x, 42*y).expand(func=True).diff(x) == 0
+    assert hypot(17*x, 42*y).diff(y).expand(func=True) - hypot(17*x, 42*y).expand(func=True).diff(y) == 0
+
+    assert hypot(17*x, 42*y).diff(x).expand(func=True) - 2*17*17*x*((17*x)**2 + (42*y)**2)**Rational(-1, 2)/2 == 0
+    assert hypot(17*x, 42*y).diff(y).expand(func=True) - 2*42*42*y*((17*x)**2 + (42*y)**2)**Rational(-1, 2)/2 == 0

rl/Lib/site-packages/sympy/codegen/tests/test_cnodes.py

@@ -0,0 +1,112 @@
+from sympy.core.symbol import symbols
+from sympy.printing.codeprinter import ccode
+from sympy.codegen.ast import Declaration, Variable, float64, int64, String, CodeBlock
+from sympy.codegen.cnodes import (
+    alignof, CommaOperator, goto, Label, PreDecrement, PostDecrement, PreIncrement, PostIncrement,
+    sizeof, union, struct
+)
+
+x, y = symbols('x y')
+
+
+def test_alignof():
+    ax = alignof(x)
+    assert ccode(ax) == 'alignof(x)'
+    assert ax.func(*ax.args) == ax
+
+
+def test_CommaOperator():
+    expr = CommaOperator(PreIncrement(x), 2*x)
+    assert ccode(expr) == '(++(x), 2*x)'
+    assert expr.func(*expr.args) == expr
+
+
+def test_goto_Label():
+    s = 'early_exit'
+    g = goto(s)
+    assert g.func(*g.args) == g
+    assert g != goto('foobar')
+    assert ccode(g) == 'goto early_exit'
+
+    l1 = Label(s)
+    assert ccode(l1) == 'early_exit:'
+    assert l1 == Label('early_exit')
+    assert l1 != Label('foobar')
+
+    body = [PreIncrement(x)]
+    l2 = Label(s, body)
+    assert l2.name == String("early_exit")
+    assert l2.body == CodeBlock(PreIncrement(x))
+    assert ccode(l2) == ("early_exit:\n"
+        "++(x);")
+
+    body = [PreIncrement(x), PreDecrement(y)]
+    l2 = Label(s, body)
+    assert l2.name == String("early_exit")
+    assert l2.body == CodeBlock(PreIncrement(x), PreDecrement(y))
+    assert ccode(l2) == ("early_exit:\n"
+        "{\n   ++(x);\n   --(y);\n}")
+
+
+def test_PreDecrement():
+    p = PreDecrement(x)
+    assert p.func(*p.args) == p
+    assert ccode(p) == '--(x)'
+
+
+def test_PostDecrement():
+    p = PostDecrement(x)
+    assert p.func(*p.args) == p
+    assert ccode(p) == '(x)--'
+
+
+def test_PreIncrement():
+    p = PreIncrement(x)
+    assert p.func(*p.args) == p
+    assert ccode(p) == '++(x)'
+
+
+def test_PostIncrement():
+    p = PostIncrement(x)
+    assert p.func(*p.args) == p
+    assert ccode(p) == '(x)++'
+
+
+def test_sizeof():
+    typename = 'unsigned int'
+    sz = sizeof(typename)
+    assert ccode(sz) == 'sizeof(%s)' % typename
+    assert sz.func(*sz.args) == sz
+    assert not sz.is_Atom
+    assert sz.atoms() == {String('unsigned int'), String('sizeof')}
+
+
+def test_struct():
+    vx, vy = Variable(x, type=float64), Variable(y, type=float64)
+    s = struct('vec2', [vx, vy])
+    assert s.func(*s.args) == s
+    assert s == struct('vec2', (vx, vy))
+    assert s != struct('vec2', (vy, vx))
+    assert str(s.name) == 'vec2'
+    assert len(s.declarations) == 2
+    assert all(isinstance(arg, Declaration) for arg in s.declarations)
+    assert ccode(s) == (
+        "struct vec2 {\n"
+        "   double x;\n"
+        "   double y;\n"
+        "}")
+
+
+def test_union():
+    vx, vy = Variable(x, type=float64), Variable(y, type=int64)
+    u = union('dualuse', [vx, vy])
+    assert u.func(*u.args) == u
+    assert u == union('dualuse', (vx, vy))
+    assert str(u.name) == 'dualuse'
+    assert len(u.declarations) == 2
+    assert all(isinstance(arg, Declaration) for arg in u.declarations)
+    assert ccode(u) == (
+        "union dualuse {\n"
+        "   double x;\n"
+        "   int64_t y;\n"
+        "}")

rl/Lib/site-packages/sympy/codegen/tests/test_cxxnodes.py

@@ -0,0 +1,14 @@
+from sympy.core.symbol import Symbol
+from sympy.codegen.ast import Type
+from sympy.codegen.cxxnodes import using
+from sympy.printing.codeprinter import cxxcode
+
+x = Symbol('x')
+
+def test_using():
+    v = Type('std::vector')
+    u1 = using(v)
+    assert cxxcode(u1) == 'using std::vector'
+
+    u2 = using(v, 'vec')
+    assert cxxcode(u2) == 'using vec = std::vector'

rl/Lib/site-packages/sympy/codegen/tests/test_fnodes.py

@@ -0,0 +1,213 @@
+import os
+import tempfile
+from sympy.core.symbol import (Symbol, symbols)
+from sympy.codegen.ast import (
+    Assignment, Print, Declaration, FunctionDefinition, Return, real,
+    FunctionCall, Variable, Element, integer
+)
+from sympy.codegen.fnodes import (
+    allocatable, ArrayConstructor, isign, dsign, cmplx, kind, literal_dp,
+    Program, Module, use, Subroutine, dimension, assumed_extent, ImpliedDoLoop,
+    intent_out, size, Do, SubroutineCall, sum_, array, bind_C
+)
+from sympy.codegen.futils import render_as_module
+from sympy.core.expr import unchanged
+from sympy.external import import_module
+from sympy.printing.codeprinter import fcode
+from sympy.utilities._compilation import has_fortran, compile_run_strings, compile_link_import_strings
+from sympy.utilities._compilation.util import may_xfail
+from sympy.testing.pytest import skip, XFAIL
+
+cython = import_module('cython')
+np = import_module('numpy')
+
+
+def test_size():
+    x = Symbol('x', real=True)
+    sx = size(x)
+    assert fcode(sx, source_format='free') == 'size(x)'
+
+
+@may_xfail
+def test_size_assumed_shape():
+    if not has_fortran():
+        skip("No fortran compiler found.")
+    a = Symbol('a', real=True)
+    body = [Return((sum_(a**2)/size(a))**.5)]
+    arr = array(a, dim=[':'], intent='in')
+    fd = FunctionDefinition(real, 'rms', [arr], body)
+    render_as_module([fd], 'mod_rms')
+
+    (stdout, stderr), info = compile_run_strings([
+        ('rms.f90', render_as_module([fd], 'mod_rms')),
+        ('main.f90', (
+            'program myprog\n'
+            'use mod_rms, only: rms\n'
+            'real*8, dimension(4), parameter :: x = [4, 2, 2, 2]\n'
+            'print *, dsqrt(7d0) - rms(x)\n'
+            'end program\n'
+        ))
+    ], clean=True)
+    assert '0.00000' in stdout
+    assert stderr == ''
+    assert info['exit_status'] == os.EX_OK
+
+
+@XFAIL  # https://github.com/sympy/sympy/issues/20265
+@may_xfail
+def test_ImpliedDoLoop():
+    if not has_fortran():
+        skip("No fortran compiler found.")
+
+    a, i = symbols('a i', integer=True)
+    idl = ImpliedDoLoop(i**3, i, -3, 3, 2)
+    ac = ArrayConstructor([-28, idl, 28])
+    a = array(a, dim=[':'], attrs=[allocatable])
+    prog = Program('idlprog', [
+        a.as_Declaration(),
+        Assignment(a, ac),
+        Print([a])
+    ])
+    fsrc = fcode(prog, standard=2003, source_format='free')
+    (stdout, stderr), info = compile_run_strings([('main.f90', fsrc)], clean=True)
+    for numstr in '-28 -27 -1 1 27 28'.split():
+        assert numstr in stdout
+    assert stderr == ''
+    assert info['exit_status'] == os.EX_OK
+
+
+@may_xfail
+def test_Program():
+    x = Symbol('x', real=True)
+    vx = Variable.deduced(x, 42)
+    decl = Declaration(vx)
+    prnt = Print([x, x+1])
+    prog = Program('foo', [decl, prnt])
+    if not has_fortran():
+        skip("No fortran compiler found.")
+
+    (stdout, stderr), info = compile_run_strings([('main.f90', fcode(prog, standard=90))], clean=True)
+    assert '42' in stdout
+    assert '43' in stdout
+    assert stderr == ''
+    assert info['exit_status'] == os.EX_OK
+
+
+@may_xfail
+def test_Module():
+    x = Symbol('x', real=True)
+    v_x = Variable.deduced(x)
+    sq = FunctionDefinition(real, 'sqr', [v_x], [Return(x**2)])
+    mod_sq = Module('mod_sq', [], [sq])
+    sq_call = FunctionCall('sqr', [42.])
+    prg_sq = Program('foobar', [
+        use('mod_sq', only=['sqr']),
+        Print(['"Square of 42 = "', sq_call])
+    ])
+    if not has_fortran():
+        skip("No fortran compiler found.")
+    (stdout, stderr), info = compile_run_strings([
+        ('mod_sq.f90', fcode(mod_sq, standard=90)),
+        ('main.f90', fcode(prg_sq, standard=90))
+    ], clean=True)
+    assert '42' in stdout
+    assert str(42**2) in stdout
+    assert stderr == ''
+
+
+@XFAIL  # https://github.com/sympy/sympy/issues/20265
+@may_xfail
+def test_Subroutine():
+    # Code to generate the subroutine in the example from
+    # http://www.fortran90.org/src/best-practices.html#arrays
+    r = Symbol('r', real=True)
+    i = Symbol('i', integer=True)
+    v_r = Variable.deduced(r, attrs=(dimension(assumed_extent), intent_out))
+    v_i = Variable.deduced(i)
+    v_n = Variable('n', integer)
+    do_loop = Do([
+        Assignment(Element(r, [i]), literal_dp(1)/i**2)
+    ], i, 1, v_n)
+    sub = Subroutine("f", [v_r], [
+        Declaration(v_n),
+        Declaration(v_i),
+        Assignment(v_n, size(r)),
+        do_loop
+    ])
+    x = Symbol('x', real=True)
+    v_x3 = Variable.deduced(x, attrs=[dimension(3)])
+    mod = Module('mymod', definitions=[sub])
+    prog = Program('foo', [
+        use(mod, only=[sub]),
+        Declaration(v_x3),
+        SubroutineCall(sub, [v_x3]),
+        Print([sum_(v_x3), v_x3])
+    ])
+
+    if not has_fortran():
+        skip("No fortran compiler found.")
+
+    (stdout, stderr), info = compile_run_strings([
+        ('a.f90', fcode(mod, standard=90)),
+        ('b.f90', fcode(prog, standard=90))
+    ], clean=True)
+    ref = [1.0/i**2 for i in range(1, 4)]
+    assert str(sum(ref))[:-3] in stdout
+    for _ in ref:
+        assert str(_)[:-3] in stdout
+    assert stderr == ''
+
+
+def test_isign():
+    x = Symbol('x', integer=True)
+    assert unchanged(isign, 1, x)
+    assert fcode(isign(1, x), standard=95, source_format='free') == 'isign(1, x)'
+
+
+def test_dsign():
+    x = Symbol('x')
+    assert unchanged(dsign, 1, x)
+    assert fcode(dsign(literal_dp(1), x), standard=95, source_format='free') == 'dsign(1d0, x)'
+
+
+def test_cmplx():
+    x = Symbol('x')
+    assert unchanged(cmplx, 1, x)
+
+
+def test_kind():
+    x = Symbol('x')
+    assert unchanged(kind, x)
+
+
+def test_literal_dp():
+    assert fcode(literal_dp(0), source_format='free') == '0d0'
+
+
+@may_xfail
+def test_bind_C():
+    if not has_fortran():
+        skip("No fortran compiler found.")
+    if not cython:
+        skip("Cython not found.")
+    if not np:
+        skip("NumPy not found.")
+
+    a = Symbol('a', real=True)
+    s = Symbol('s', integer=True)
+    body = [Return((sum_(a**2)/s)**.5)]
+    arr = array(a, dim=[s], intent='in')
+    fd = FunctionDefinition(real, 'rms', [arr, s], body, attrs=[bind_C('rms')])
+    f_mod = render_as_module([fd], 'mod_rms')
+
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = compile_link_import_strings([
+            ('rms.f90', f_mod),
+            ('_rms.pyx', (
+                "#cython: language_level={}\n".format("3") +
+                "cdef extern double rms(double*, int*)\n"
+                "def py_rms(double[::1] x):\n"
+                "    cdef int s = x.size\n"
+                "    return rms(&x[0], &s)\n"))
+        ], build_dir=folder)
+        assert abs(mod.py_rms(np.array([2., 4., 2., 2.])) - 7**0.5) < 1e-14

rl/Lib/site-packages/sympy/codegen/tests/test_matrix_nodes.py

@@ -0,0 +1,50 @@
+from sympy.core.symbol import symbols
+from sympy.core.function import Function
+from sympy.matrices.dense import Matrix
+from sympy.matrices.dense import zeros
+from sympy.simplify.simplify import simplify
+from sympy.codegen.matrix_nodes import MatrixSolve
+from sympy.utilities.lambdify import lambdify
+from sympy.printing.numpy import NumPyPrinter
+from sympy.testing.pytest import skip
+from sympy.external import import_module
+
+
+def test_matrix_solve_issue_24862():
+    A = Matrix(3, 3, symbols('a:9'))
+    b = Matrix(3, 1, symbols('b:3'))
+    hash(MatrixSolve(A, b))
+
+
+def test_matrix_solve_derivative_exact():
+    q = symbols('q')
+    a11, a12, a21, a22, b1, b2 = (
+        f(q) for f in symbols('a11 a12 a21 a22 b1 b2', cls=Function))
+    A = Matrix([[a11, a12], [a21, a22]])
+    b = Matrix([b1, b2])
+    x_lu = A.LUsolve(b)
+    dxdq_lu = A.LUsolve(b.diff(q) - A.diff(q) * A.LUsolve(b))
+    assert simplify(x_lu.diff(q) - dxdq_lu) == zeros(2, 1)
+    # dxdq_ms is the MatrixSolve equivalent of dxdq_lu
+    dxdq_ms = MatrixSolve(A, b.diff(q) - A.diff(q) * MatrixSolve(A, b))
+    assert MatrixSolve(A, b).diff(q) == dxdq_ms
+
+
+def test_matrix_solve_derivative_numpy():
+    np = import_module('numpy')
+    if not np:
+        skip("numpy not installed.")
+    q = symbols('q')
+    a11, a12, a21, a22, b1, b2 = (
+        f(q) for f in symbols('a11 a12 a21 a22 b1 b2', cls=Function))
+    A = Matrix([[a11, a12], [a21, a22]])
+    b = Matrix([b1, b2])
+    dx_lu = A.LUsolve(b).diff(q)
+    subs = {a11.diff(q): 0.2, a12.diff(q): 0.3, a21.diff(q): 0.1,
+            a22.diff(q): 0.5, b1.diff(q): 0.4, b2.diff(q): 0.9,
+            a11: 1.3, a12: 0.5, a21: 1.2, a22: 4, b1: 6.2, b2: 3.5}
+    p, p_vals = zip(*subs.items())
+    dx_sm = MatrixSolve(A, b).diff(q)
+    np.testing.assert_allclose(
+        lambdify(p, dx_sm, printer=NumPyPrinter)(*p_vals),
+        lambdify(p, dx_lu, printer=NumPyPrinter)(*p_vals))

rl/Lib/site-packages/sympy/codegen/tests/test_numpy_nodes.py

@@ -0,0 +1,50 @@
+from itertools import product
+from sympy.core.singleton import S
+from sympy.core.symbol import symbols
+from sympy.functions.elementary.exponential import (exp, log)
+from sympy.printing.repr import srepr
+from sympy.codegen.numpy_nodes import logaddexp, logaddexp2
+
+x, y, z = symbols('x y z')
+
+def test_logaddexp():
+    lae_xy = logaddexp(x, y)
+    ref_xy = log(exp(x) + exp(y))
+    for wrt, deriv_order in product([x, y, z], range(3)):
+        assert (
+            lae_xy.diff(wrt, deriv_order) -
+            ref_xy.diff(wrt, deriv_order)
+        ).rewrite(log).simplify() == 0
+
+    one_third_e = 1*exp(1)/3
+    two_thirds_e = 2*exp(1)/3
+    logThirdE = log(one_third_e)
+    logTwoThirdsE = log(two_thirds_e)
+    lae_sum_to_e = logaddexp(logThirdE, logTwoThirdsE)
+    assert lae_sum_to_e.rewrite(log) == 1
+    assert lae_sum_to_e.simplify() == 1
+    was = logaddexp(2, 3)
+    assert srepr(was) == srepr(was.simplify())  # cannot simplify with 2, 3
+
+
+def test_logaddexp2():
+    lae2_xy = logaddexp2(x, y)
+    ref2_xy = log(2**x + 2**y)/log(2)
+    for wrt, deriv_order in product([x, y, z], range(3)):
+        assert (
+            lae2_xy.diff(wrt, deriv_order) -
+            ref2_xy.diff(wrt, deriv_order)
+        ).rewrite(log).cancel() == 0
+
+    def lb(x):
+        return log(x)/log(2)
+
+    two_thirds = S.One*2/3
+    four_thirds = 2*two_thirds
+    lbTwoThirds = lb(two_thirds)
+    lbFourThirds = lb(four_thirds)
+    lae2_sum_to_2 = logaddexp2(lbTwoThirds, lbFourThirds)
+    assert lae2_sum_to_2.rewrite(log) == 1
+    assert lae2_sum_to_2.simplify() == 1
+    was = logaddexp2(x, y)
+    assert srepr(was) == srepr(was.simplify())  # cannot simplify with x, y

rl/Lib/site-packages/sympy/codegen/tests/test_pynodes.py

@@ -0,0 +1,13 @@
+from sympy.core.symbol import symbols
+from sympy.codegen.pynodes import List
+
+
+def test_List():
+    l = List(2, 3, 4)
+    assert l == List(2, 3, 4)
+    assert str(l) == "[2, 3, 4]"
+    x, y, z = symbols('x y z')
+    l = List(x**2,y**3,z**4)
+    # contrary to python's built-in list, we can call e.g. "replace" on List.
+    m = l.replace(lambda arg: arg.is_Pow and arg.exp>2, lambda p: p.base-p.exp)
+    assert m == [x**2, y-3, z-4]

rl/Lib/site-packages/sympy/codegen/tests/test_pyutils.py

@@ -0,0 +1,7 @@
+from sympy.codegen.ast import Print
+from sympy.codegen.pyutils import render_as_module
+
+def test_standard():
+    ast = Print('x y'.split(), r"coordinate: %12.5g %12.5g\n")
+    assert render_as_module(ast, standard='python3') == \
+        '\n\nprint("coordinate: %12.5g %12.5g\\n" % (x, y), end="")'

rl/Lib/site-packages/sympy/codegen/tests/test_rewriting.py

@@ -0,0 +1,479 @@
+import tempfile
+from sympy.core.numbers import pi, Rational
+from sympy.core.power import Pow
+from sympy.core.singleton import S
+from sympy.core.symbol import Symbol
+from sympy.functions.elementary.complexes import Abs
+from sympy.functions.elementary.exponential import (exp, log)
+from sympy.functions.elementary.trigonometric import (cos, sin, sinc)
+from sympy.matrices.expressions.matexpr import MatrixSymbol
+from sympy.assumptions import assuming, Q
+from sympy.external import import_module
+from sympy.printing.codeprinter import ccode
+from sympy.codegen.matrix_nodes import MatrixSolve
+from sympy.codegen.cfunctions import log2, exp2, expm1, log1p
+from sympy.codegen.numpy_nodes import logaddexp, logaddexp2
+from sympy.codegen.scipy_nodes import cosm1, powm1
+from sympy.codegen.rewriting import (
+    optimize, cosm1_opt, log2_opt, exp2_opt, expm1_opt, log1p_opt, powm1_opt, optims_c99,
+    create_expand_pow_optimization, matinv_opt, logaddexp_opt, logaddexp2_opt,
+    optims_numpy, optims_scipy, sinc_opts, FuncMinusOneOptim
+)
+from sympy.testing.pytest import XFAIL, skip
+from sympy.utilities import lambdify
+from sympy.utilities._compilation import compile_link_import_strings, has_c
+from sympy.utilities._compilation.util import may_xfail
+
+cython = import_module('cython')
+numpy = import_module('numpy')
+scipy = import_module('scipy')
+
+
+def test_log2_opt():
+    x = Symbol('x')
+    expr1 = 7*log(3*x + 5)/(log(2))
+    opt1 = optimize(expr1, [log2_opt])
+    assert opt1 == 7*log2(3*x + 5)
+    assert opt1.rewrite(log) == expr1
+
+    expr2 = 3*log(5*x + 7)/(13*log(2))
+    opt2 = optimize(expr2, [log2_opt])
+    assert opt2 == 3*log2(5*x + 7)/13
+    assert opt2.rewrite(log) == expr2
+
+    expr3 = log(x)/log(2)
+    opt3 = optimize(expr3, [log2_opt])
+    assert opt3 == log2(x)
+    assert opt3.rewrite(log) == expr3
+
+    expr4 = log(x)/log(2) + log(x+1)
+    opt4 = optimize(expr4, [log2_opt])
+    assert opt4 == log2(x) + log(2)*log2(x+1)
+    assert opt4.rewrite(log) == expr4
+
+    expr5 = log(17)
+    opt5 = optimize(expr5, [log2_opt])
+    assert opt5 == expr5
+
+    expr6 = log(x + 3)/log(2)
+    opt6 = optimize(expr6, [log2_opt])
+    assert str(opt6) == 'log2(x + 3)'
+    assert opt6.rewrite(log) == expr6
+
+
+def test_exp2_opt():
+    x = Symbol('x')
+    expr1 = 1 + 2**x
+    opt1 = optimize(expr1, [exp2_opt])
+    assert opt1 == 1 + exp2(x)
+    assert opt1.rewrite(Pow) == expr1
+
+    expr2 = 1 + 3**x
+    assert expr2 == optimize(expr2, [exp2_opt])
+
+
+def test_expm1_opt():
+    x = Symbol('x')
+
+    expr1 = exp(x) - 1
+    opt1 = optimize(expr1, [expm1_opt])
+    assert expm1(x) - opt1 == 0
+    assert opt1.rewrite(exp) == expr1
+
+    expr2 = 3*exp(x) - 3
+    opt2 = optimize(expr2, [expm1_opt])
+    assert 3*expm1(x) == opt2
+    assert opt2.rewrite(exp) == expr2
+
+    expr3 = 3*exp(x) - 5
+    opt3 = optimize(expr3, [expm1_opt])
+    assert 3*expm1(x) - 2 == opt3
+    assert opt3.rewrite(exp) == expr3
+    expm1_opt_non_opportunistic = FuncMinusOneOptim(exp, expm1, opportunistic=False)
+    assert expr3 == optimize(expr3, [expm1_opt_non_opportunistic])
+    assert opt1 == optimize(expr1, [expm1_opt_non_opportunistic])
+    assert opt2 == optimize(expr2, [expm1_opt_non_opportunistic])
+
+    expr4 = 3*exp(x) + log(x) - 3
+    opt4 = optimize(expr4, [expm1_opt])
+    assert 3*expm1(x) + log(x) == opt4
+    assert opt4.rewrite(exp) == expr4
+
+    expr5 = 3*exp(2*x) - 3
+    opt5 = optimize(expr5, [expm1_opt])
+    assert 3*expm1(2*x) == opt5
+    assert opt5.rewrite(exp) == expr5
+
+    expr6 = (2*exp(x) + 1)/(exp(x) + 1) + 1
+    opt6 = optimize(expr6, [expm1_opt])
+    assert opt6.count_ops() <= expr6.count_ops()
+
+    def ev(e):
+        return e.subs(x, 3).evalf()
+    assert abs(ev(expr6) - ev(opt6)) < 1e-15
+
+    y = Symbol('y')
+    expr7 = (2*exp(x) - 1)/(1 - exp(y)) - 1/(1-exp(y))
+    opt7 = optimize(expr7, [expm1_opt])
+    assert -2*expm1(x)/expm1(y) == opt7
+    assert (opt7.rewrite(exp) - expr7).factor() == 0
+
+    expr8 = (1+exp(x))**2 - 4
+    opt8 = optimize(expr8, [expm1_opt])
+    tgt8a = (exp(x) + 3)*expm1(x)
+    tgt8b = 2*expm1(x) + expm1(2*x)
+    # Both tgt8a & tgt8b seem to give full precision (~16 digits for double)
+    # for x=1e-7 (compare with expr8 which only achieves ~8 significant digits).
+    # If we can show that either tgt8a or tgt8b is preferable, we can
+    # change this test to ensure the preferable version is returned.
+    assert (tgt8a - tgt8b).rewrite(exp).factor() == 0
+    assert opt8 in (tgt8a, tgt8b)
+    assert (opt8.rewrite(exp) - expr8).factor() == 0
+
+    expr9 = sin(expr8)
+    opt9 = optimize(expr9, [expm1_opt])
+    tgt9a = sin(tgt8a)
+    tgt9b = sin(tgt8b)
+    assert opt9 in (tgt9a, tgt9b)
+    assert (opt9.rewrite(exp) - expr9.rewrite(exp)).factor().is_zero
+
+
+def test_expm1_two_exp_terms():
+    x, y = map(Symbol, 'x y'.split())
+    expr1 = exp(x) + exp(y) - 2
+    opt1 = optimize(expr1, [expm1_opt])
+    assert opt1 == expm1(x) + expm1(y)
+
+
+def test_cosm1_opt():
+    x = Symbol('x')
+
+    expr1 = cos(x) - 1
+    opt1 = optimize(expr1, [cosm1_opt])
+    assert cosm1(x) - opt1 == 0
+    assert opt1.rewrite(cos) == expr1
+
+    expr2 = 3*cos(x) - 3
+    opt2 = optimize(expr2, [cosm1_opt])
+    assert 3*cosm1(x) == opt2
+    assert opt2.rewrite(cos) == expr2
+
+    expr3 = 3*cos(x) - 5
+    opt3 = optimize(expr3, [cosm1_opt])
+    assert 3*cosm1(x) - 2 == opt3
+    assert opt3.rewrite(cos) == expr3
+    cosm1_opt_non_opportunistic = FuncMinusOneOptim(cos, cosm1, opportunistic=False)
+    assert expr3 == optimize(expr3, [cosm1_opt_non_opportunistic])
+    assert opt1 == optimize(expr1, [cosm1_opt_non_opportunistic])
+    assert opt2 == optimize(expr2, [cosm1_opt_non_opportunistic])
+
+    expr4 = 3*cos(x) + log(x) - 3
+    opt4 = optimize(expr4, [cosm1_opt])
+    assert 3*cosm1(x) + log(x) == opt4
+    assert opt4.rewrite(cos) == expr4
+
+    expr5 = 3*cos(2*x) - 3
+    opt5 = optimize(expr5, [cosm1_opt])
+    assert 3*cosm1(2*x) == opt5
+    assert opt5.rewrite(cos) == expr5
+
+    expr6 = 2 - 2*cos(x)
+    opt6 = optimize(expr6, [cosm1_opt])
+    assert -2*cosm1(x) == opt6
+    assert opt6.rewrite(cos) == expr6
+
+
+def test_cosm1_two_cos_terms():
+    x, y = map(Symbol, 'x y'.split())
+    expr1 = cos(x) + cos(y) - 2
+    opt1 = optimize(expr1, [cosm1_opt])
+    assert opt1 == cosm1(x) + cosm1(y)
+
+
+def test_expm1_cosm1_mixed():
+    x = Symbol('x')
+    expr1 = exp(x) + cos(x) - 2
+    opt1 = optimize(expr1, [expm1_opt, cosm1_opt])
+    assert opt1 == cosm1(x) + expm1(x)
+
+
+def _check_num_lambdify(expr, opt, val_subs, approx_ref, lambdify_kw=None, poorness=1e10):
+    """ poorness=1e10 signifies that `expr` loses precision of at least ten decimal digits. """
+    num_ref = expr.subs(val_subs).evalf()
+    eps = numpy.finfo(numpy.float64).eps
+    assert abs(num_ref - approx_ref) < approx_ref*eps
+    f1 = lambdify(list(val_subs.keys()), opt, **(lambdify_kw or {}))
+    args_float = tuple(map(float, val_subs.values()))
+    num_err1 = abs(f1(*args_float) - approx_ref)
+    assert num_err1 < abs(num_ref*eps)
+    f2 = lambdify(list(val_subs.keys()), expr, **(lambdify_kw or {}))
+    num_err2 = abs(f2(*args_float) - approx_ref)
+    assert num_err2 > abs(num_ref*eps*poorness)   # this only ensures that the *test* works as intended
+
+
+def test_cosm1_apart():
+    x = Symbol('x')
+
+    expr1 = 1/cos(x) - 1
+    opt1 = optimize(expr1, [cosm1_opt])
+    assert opt1 == -cosm1(x)/cos(x)
+    if scipy:
+        _check_num_lambdify(expr1, opt1, {x: S(10)**-30}, 5e-61, lambdify_kw={"modules": 'scipy'})
+
+    expr2 = 2/cos(x) - 2
+    opt2 = optimize(expr2, optims_scipy)
+    assert opt2 == -2*cosm1(x)/cos(x)
+    if scipy:
+        _check_num_lambdify(expr2, opt2, {x: S(10)**-30}, 1e-60, lambdify_kw={"modules": 'scipy'})
+
+    expr3 = pi/cos(3*x) - pi
+    opt3 = optimize(expr3, [cosm1_opt])
+    assert opt3 == -pi*cosm1(3*x)/cos(3*x)
+    if scipy:
+        _check_num_lambdify(expr3, opt3, {x: S(10)**-30/3}, float(5e-61*pi), lambdify_kw={"modules": 'scipy'})
+
+
+def test_powm1():
+    args = x, y = map(Symbol, "xy")
+
+    expr1 = x**y - 1
+    opt1 = optimize(expr1, [powm1_opt])
+    assert opt1 == powm1(x, y)
+    for arg in args:
+        assert expr1.diff(arg) == opt1.diff(arg)
+    if scipy and tuple(map(int, scipy.version.version.split('.')[:3])) >= (1, 10, 0):
+        subs1_a = {x: Rational(*(1.0+1e-13).as_integer_ratio()), y: pi}
+        ref1_f64_a = 3.139081648208105e-13
+        _check_num_lambdify(expr1, opt1, subs1_a, ref1_f64_a, lambdify_kw={"modules": 'scipy'}, poorness=10**11)
+
+        subs1_b = {x: pi, y: Rational(*(1e-10).as_integer_ratio())}
+        ref1_f64_b = 1.1447298859149205e-10
+        _check_num_lambdify(expr1, opt1, subs1_b, ref1_f64_b, lambdify_kw={"modules": 'scipy'}, poorness=10**9)
+
+
+def test_log1p_opt():
+    x = Symbol('x')
+    expr1 = log(x + 1)
+    opt1 = optimize(expr1, [log1p_opt])
+    assert log1p(x) - opt1 == 0
+    assert opt1.rewrite(log) == expr1
+
+    expr2 = log(3*x + 3)
+    opt2 = optimize(expr2, [log1p_opt])
+    assert log1p(x) + log(3) == opt2
+    assert (opt2.rewrite(log) - expr2).simplify() == 0
+
+    expr3 = log(2*x + 1)
+    opt3 = optimize(expr3, [log1p_opt])
+    assert log1p(2*x) - opt3 == 0
+    assert opt3.rewrite(log) == expr3
+
+    expr4 = log(x+3)
+    opt4 = optimize(expr4, [log1p_opt])
+    assert str(opt4) == 'log(x + 3)'
+
+
+def test_optims_c99():
+    x = Symbol('x')
+
+    expr1 = 2**x + log(x)/log(2) + log(x + 1) + exp(x) - 1
+    opt1 = optimize(expr1, optims_c99).simplify()
+    assert opt1 == exp2(x) + log2(x) + log1p(x) + expm1(x)
+    assert opt1.rewrite(exp).rewrite(log).rewrite(Pow) == expr1
+
+    expr2 = log(x)/log(2) + log(x + 1)
+    opt2 = optimize(expr2, optims_c99)
+    assert opt2 == log2(x) + log1p(x)
+    assert opt2.rewrite(log) == expr2
+
+    expr3 = log(x)/log(2) + log(17*x + 17)
+    opt3 = optimize(expr3, optims_c99)
+    delta3 = opt3 - (log2(x) + log(17) + log1p(x))
+    assert delta3 == 0
+    assert (opt3.rewrite(log) - expr3).simplify() == 0
+
+    expr4 = 2**x + 3*log(5*x + 7)/(13*log(2)) + 11*exp(x) - 11 + log(17*x + 17)
+    opt4 = optimize(expr4, optims_c99).simplify()
+    delta4 = opt4 - (exp2(x) + 3*log2(5*x + 7)/13 + 11*expm1(x) + log(17) + log1p(x))
+    assert delta4 == 0
+    assert (opt4.rewrite(exp).rewrite(log).rewrite(Pow) - expr4).simplify() == 0
+
+    expr5 = 3*exp(2*x) - 3
+    opt5 = optimize(expr5, optims_c99)
+    delta5 = opt5 - 3*expm1(2*x)
+    assert delta5 == 0
+    assert opt5.rewrite(exp) == expr5
+
+    expr6 = exp(2*x) - 3
+    opt6 = optimize(expr6, optims_c99)
+    assert opt6 in (expm1(2*x) - 2, expr6)  # expm1(2*x) - 2 is not better or worse
+
+    expr7 = log(3*x + 3)
+    opt7 = optimize(expr7, optims_c99)
+    delta7 = opt7 - (log(3) + log1p(x))
+    assert delta7 == 0
+    assert (opt7.rewrite(log) - expr7).simplify() == 0
+
+    expr8 = log(2*x + 3)
+    opt8 = optimize(expr8, optims_c99)
+    assert opt8 == expr8
+
+
+def test_create_expand_pow_optimization():
+    cc = lambda x: ccode(
+        optimize(x, [create_expand_pow_optimization(4)]))
+    x = Symbol('x')
+    assert cc(x**4) == 'x*x*x*x'
+    assert cc(x**4 + x**2) == 'x*x + x*x*x*x'
+    assert cc(x**5 + x**4) == 'pow(x, 5) + x*x*x*x'
+    assert cc(sin(x)**4) == 'pow(sin(x), 4)'
+    # gh issue 15335
+    assert cc(x**(-4)) == '1.0/(x*x*x*x)'
+    assert cc(x**(-5)) == 'pow(x, -5)'
+    assert cc(-x**4) == '-(x*x*x*x)'
+    assert cc(x**4 - x**2) == '-(x*x) + x*x*x*x'
+    i = Symbol('i', integer=True)
+    assert cc(x**i - x**2) == 'pow(x, i) - (x*x)'
+    y = Symbol('y', real=True)
+    assert cc(Abs(exp(y**4))) == "exp(y*y*y*y)"
+
+    # gh issue 20753
+    cc2 = lambda x: ccode(optimize(x, [create_expand_pow_optimization(
+        4, base_req=lambda b: b.is_Function)]))
+    assert cc2(x**3 + sin(x)**3) == "pow(x, 3) + sin(x)*sin(x)*sin(x)"
+
+
+def test_matsolve():
+    n = Symbol('n', integer=True)
+    A = MatrixSymbol('A', n, n)
+    x = MatrixSymbol('x', n, 1)
+
+    with assuming(Q.fullrank(A)):
+        assert optimize(A**(-1) * x, [matinv_opt]) == MatrixSolve(A, x)
+        assert optimize(A**(-1) * x + x, [matinv_opt]) == MatrixSolve(A, x) + x
+
+
+def test_logaddexp_opt():
+    x, y = map(Symbol, 'x y'.split())
+    expr1 = log(exp(x) + exp(y))
+    opt1 = optimize(expr1, [logaddexp_opt])
+    assert logaddexp(x, y) - opt1 == 0
+    assert logaddexp(y, x) - opt1 == 0
+    assert opt1.rewrite(log) == expr1
+
+
+def test_logaddexp2_opt():
+    x, y = map(Symbol, 'x y'.split())
+    expr1 = log(2**x + 2**y)/log(2)
+    opt1 = optimize(expr1, [logaddexp2_opt])
+    assert logaddexp2(x, y) - opt1 == 0
+    assert logaddexp2(y, x) - opt1 == 0
+    assert opt1.rewrite(log) == expr1
+
+
+def test_sinc_opts():
+    def check(d):
+        for k, v in d.items():
+            assert optimize(k, sinc_opts) == v
+
+    x = Symbol('x')
+    check({
+        sin(x)/x       : sinc(x),
+        sin(2*x)/(2*x) : sinc(2*x),
+        sin(3*x)/x     : 3*sinc(3*x),
+        x*sin(x)       : x*sin(x)
+    })
+
+    y = Symbol('y')
+    check({
+        sin(x*y)/(x*y)       : sinc(x*y),
+        y*sin(x/y)/x         : sinc(x/y),
+        sin(sin(x))/sin(x)   : sinc(sin(x)),
+        sin(3*sin(x))/sin(x) : 3*sinc(3*sin(x)),
+        sin(x)/y             : sin(x)/y
+    })
+
+
+def test_optims_numpy():
+    def check(d):
+        for k, v in d.items():
+            assert optimize(k, optims_numpy) == v
+
+    x = Symbol('x')
+    check({
+        sin(2*x)/(2*x) + exp(2*x) - 1: sinc(2*x) + expm1(2*x),
+        log(x+3)/log(2) + log(x**2 + 1): log1p(x**2) + log2(x+3)
+    })
+
+
+@XFAIL  # room for improvement, ideally this test case should pass.
+def test_optims_numpy_TODO():
+    def check(d):
+        for k, v in d.items():
+            assert optimize(k, optims_numpy) == v
+
+    x, y = map(Symbol, 'x y'.split())
+    check({
+        log(x*y)*sin(x*y)*log(x*y+1)/(log(2)*x*y): log2(x*y)*sinc(x*y)*log1p(x*y),
+        exp(x*sin(y)/y) - 1: expm1(x*sinc(y))
+    })
+
+
+@may_xfail
+def test_compiled_ccode_with_rewriting():
+    if not cython:
+        skip("cython not installed.")
+    if not has_c():
+        skip("No C compiler found.")
+
+    x = Symbol('x')
+    about_two = 2**(58/S(117))*3**(97/S(117))*5**(4/S(39))*7**(92/S(117))/S(30)*pi
+    # about_two: 1.999999999999581826
+    unchanged = 2*exp(x) - about_two
+    xval = S(10)**-11
+    ref = unchanged.subs(x, xval).n(19) # 2.0418173913673213e-11
+
+    rewritten = optimize(2*exp(x) - about_two, [expm1_opt])
+
+    # Unfortunately, we need to call ``.n()`` on our expressions before we hand them
+    # to ``ccode``, and we need to request a large number of significant digits.
+    # In this test, results converged for double precision when the following number
+    # of significant digits were chosen:
+    NUMBER_OF_DIGITS = 25   # TODO: this should ideally be automatically handled.
+
+    func_c = '''
+#include <math.h>
+
+double func_unchanged(double x) {
+    return %(unchanged)s;
+}
+double func_rewritten(double x) {
+    return %(rewritten)s;
+}
+''' % {"unchanged": ccode(unchanged.n(NUMBER_OF_DIGITS)),
+           "rewritten": ccode(rewritten.n(NUMBER_OF_DIGITS))}
+
+    func_pyx = '''
+#cython: language_level=3
+cdef extern double func_unchanged(double)
+cdef extern double func_rewritten(double)
+def py_unchanged(x):
+    return func_unchanged(x)
+def py_rewritten(x):
+    return func_rewritten(x)
+'''
+    with tempfile.TemporaryDirectory() as folder:
+        mod, info = compile_link_import_strings(
+            [('func.c', func_c), ('_func.pyx', func_pyx)],
+            build_dir=folder, compile_kwargs={"std": 'c99'}
+        )
+        err_rewritten = abs(mod.py_rewritten(1e-11) - ref)
+        err_unchanged = abs(mod.py_unchanged(1e-11) - ref)
+        assert 1e-27 < err_rewritten < 1e-25  # highly accurate.
+        assert 1e-19 < err_unchanged < 1e-16  # quite poor.
+
+    # Tolerances used above were determined as follows:
+    # >>> no_opt = unchanged.subs(x, xval.evalf()).evalf()
+    # >>> with_opt = rewritten.n(25).subs(x, 1e-11).evalf()
+    # >>> with_opt - ref, no_opt - ref
+    # (1.1536301877952077e-26, 1.6547074214222335e-18)

rl/Lib/site-packages/sympy/codegen/tests/test_scipy_nodes.py

@@ -0,0 +1,44 @@
+from itertools import product
+from sympy.core.power import Pow
+from sympy.core.symbol import symbols
+from sympy.functions.elementary.exponential import exp, log
+from sympy.functions.elementary.trigonometric import cos
+from sympy.core.numbers import pi
+from sympy.codegen.scipy_nodes import cosm1, powm1
+
+x, y, z = symbols('x y z')
+
+
+def test_cosm1():
+    cm1_xy = cosm1(x*y)
+    ref_xy = cos(x*y) - 1
+    for wrt, deriv_order in product([x, y, z], range(3)):
+        assert (
+            cm1_xy.diff(wrt, deriv_order) -
+            ref_xy.diff(wrt, deriv_order)
+        ).rewrite(cos).simplify() == 0
+
+    expr_minus2 = cosm1(pi)
+    assert expr_minus2.rewrite(cos) == -2
+    assert cosm1(3.14).simplify() == cosm1(3.14)  # cannot simplify with 3.14
+    assert cosm1(pi/2).simplify() == -1
+    assert (1/cos(x) - 1 + cosm1(x)/cos(x)).simplify() == 0
+
+
+def test_powm1():
+    cases = {
+            powm1(x, y): x**y - 1,
+            powm1(x*y, z): (x*y)**z - 1,
+            powm1(x, y*z): x**(y*z)-1,
+            powm1(x*y*z, x*y*z): (x*y*z)**(x*y*z)-1
+    }
+    for pm1_e, ref_e in cases.items():
+        for wrt, deriv_order in product([x, y, z], range(3)):
+            der = pm1_e.diff(wrt, deriv_order)
+            ref = ref_e.diff(wrt, deriv_order)
+            delta = (der - ref).rewrite(Pow)
+            assert delta.simplify() == 0
+
+    eulers_constant_m1 = powm1(x, 1/log(x))
+    assert eulers_constant_m1.rewrite(Pow) == exp(1) - 1
+    assert eulers_constant_m1.simplify() == exp(1) - 1