def _common_new(cls, function, *symbols, discrete, **assumptions):
"""Return either a special return value or the tuple,
(function, limits, orientation). This code is common to
both ExprWithLimits and AddWithLimits."""
function = sympify(function)
if isinstance(function, Equality):
# This transforms e.g. Integral(Eq(x, y)) to Eq(Integral(x), Integral(y))
# but that is only valid for definite integrals.
limits, orientation = _process_limits(*symbols, discrete=discrete)
if not (limits and all(len(limit) == 3 for limit in limits)):
sympy_deprecation_warning(
"""
Creating a indefinite integral with an Eq() argument is
deprecated.
This is because indefinite integrals do not preserve equality
due to the arbitrary constants. If you want an equality of
indefinite integrals, use Eq(Integral(a, x), Integral(b, x))
explicitly.
""",
deprecated_since_version="1.6",
active_deprecations_target="deprecated-indefinite-integral-eq",
stacklevel=5,
)
lhs = function.lhs
rhs = function.rhs
return Equality(cls(lhs, *symbols, **assumptions), \
cls(rhs, *symbols, **assumptions))
if function is S.NaN:
return S.NaN
if symbols:
limits, orientation = _process_limits(*symbols, discrete=discrete)
for i, li in enumerate(limits):
if len(li) == 4:
function = function.subs(li[0], li[-1])
limits[i] = Tuple(*li[:-1])
else:
# symbol not provided -- we can still try to compute a general form
free = function.free_symbols
if len(free) != 1:
raise ValueError("specify dummy variables for %s" % function)
limits, orientation = [Tuple(s) for s in free], 1
# denest any nested calls
while cls == type(function):
limits = list(function.limits) + limits
function = function.function
# Any embedded piecewise functions need to be brought out to the
# top level. We only fold Piecewise that contain the integration
# variable.
reps = {}
symbols_of_integration = {i[0] for i in limits}
for p in function.atoms(Piecewise):
if not p.has(*symbols_of_integration):
reps[p] = Dummy()
# mask off those that don't
function = function.xreplace(reps)
# do the fold
function = piecewise_fold(function)
# remove the masking
function = function.xreplace({v: k for k, v in reps.items()})
return function, limits, orientation
def _eval_subs(self, old, new):
"""
Perform substitutions over non-dummy variables
of an expression with limits. Also, can be used
to specify point-evaluation of an abstract antiderivative.
Examples
========
>>> from sympy import Sum, oo
>>> from sympy.abc import s, n
>>> Sum(1/n**s, (n, 1, oo)).subs(s, 2)
Sum(n**(-2), (n, 1, oo))
>>> from sympy import Integral
>>> from sympy.abc import x, a
>>> Integral(a*x**2, x).subs(x, 4)
Integral(a*x**2, (x, 4))
See Also
========
variables : Lists the integration variables
transform : Perform mapping on the dummy variable for integrals
change_index : Perform mapping on the sum and product dummy variables
"""
from sympy.core.function import AppliedUndef, UndefinedFunction
func, limits = self.function, list(self.limits)
# If one of the expressions we are replacing is used as a func index
# one of two things happens.
# - the old variable first appears as a free variable
# so we perform all free substitutions before it becomes
# a func index.
# - the old variable first appears as a func index, in
# which case we ignore. See change_index.
# Reorder limits to match standard mathematical practice for scoping
limits.reverse()
if not isinstance(old, Symbol) or \
old.free_symbols.intersection(self.free_symbols):
sub_into_func = True
for i, xab in enumerate(limits):
if 1 == len(xab) and old == xab[0]:
if new._diff_wrt:
xab = (new, )
else:
xab = (old, old)
limits[i] = Tuple(xab[0],
*[l._subs(old, new) for l in xab[1:]])
if len(xab[0].free_symbols.intersection(
old.free_symbols)) != 0:
sub_into_func = False
break
if isinstance(old, AppliedUndef) or isinstance(
old, UndefinedFunction):
sy2 = set(self.variables).intersection(set(new.atoms(Symbol)))
sy1 = set(self.variables).intersection(set(old.args))
if not sy2.issubset(sy1):
raise ValueError(
"substitution can not create dummy dependencies")
sub_into_func = True
if sub_into_func:
func = func.subs(old, new)
else:
# old is a Symbol and a dummy variable of some limit
for i, xab in enumerate(limits):
if len(xab) == 3:
limits[i] = Tuple(xab[0],
*[l._subs(old, new) for l in xab[1:]])
if old == xab[0]:
break
# simplify redundant limits (x, x) to (x, )
for i, xab in enumerate(limits):
if len(xab) == 2 and (xab[0] - xab[1]).is_zero:
limits[i] = Tuple(xab[0], )
# Reorder limits back to representation-form
limits.reverse()
return self.func(func, *limits)
def __new__(cls, *args, **kwargs):
evaluate = kwargs.get('evaluate', global_evaluate[0])
on_morph = kwargs.get('on_morph', 'ignore')
# unpack into coords
coords = args[0] if len(args) == 1 else args
# check args and handle quickly handle Point instances
if isinstance(coords, Point):
# even if we're mutating the dimension of a point, we
# don't reevaluate its coordinates
evaluate = False
if len(coords) == kwargs.get('dim', len(coords)):
return coords
if not is_sequence(coords):
raise TypeError(filldedent('''
Expecting sequence of coordinates, not `{}`'''
.format(func_name(coords))))
# A point where only `dim` is specified is initialized
# to zeros.
if len(coords) == 0 and kwargs.get('dim', None):
coords = (S.Zero,)*kwargs.get('dim')
coords = Tuple(*coords)
dim = kwargs.get('dim', len(coords))
if len(coords) < 2:
raise ValueError(filldedent('''
Point requires 2 or more coordinates or
keyword `dim` > 1.'''))
if len(coords) != dim:
message = ("Dimension of {} needs to be changed "
"from {} to {}.").format(coords, len(coords), dim)
if on_morph == 'ignore':
pass
elif on_morph == "error":
raise ValueError(message)
elif on_morph == 'warn':
warnings.warn(message)
else:
raise ValueError(filldedent('''
on_morph value should be 'error',
'warn' or 'ignore'.'''))
if any(coords[dim:]):
raise ValueError('Nonzero coordinates cannot be removed.')
if any(a.is_number and im(a) for a in coords):
raise ValueError('Imaginary coordinates are not permitted.')
if not all(isinstance(a, Expr) for a in coords):
raise TypeError('Coordinates must be valid SymPy expressions.')
# pad with zeros appropriately
coords = coords[:dim] + (S.Zero,)*(dim - len(coords))
# Turn any Floats into rationals and simplify
# any expressions before we instantiate
if evaluate:
coords = coords.xreplace(dict(
[(f, simplify(nsimplify(f, rational=True)))
for f in coords.atoms(Float)]))
# return 2D or 3D instances
if len(coords) == 2:
kwargs['_nocheck'] = True
return Point2D(*coords, **kwargs)
elif len(coords) == 3:
kwargs['_nocheck'] = True
return Point3D(*coords, **kwargs)
# the general Point
return GeometryEntity.__new__(cls, *coords)
def aother(self):
""" Second set of numerator parameters. """
return Tuple(*self.args[0][1])
def bq(self):
""" Combined denominator parameters. """
return Tuple(*(self.args[1][0] + self.args[1][1]))
def ap(self):
""" Numerator parameters of the hypergeometric function. """
return Tuple(*self.args[0])
def an(self):
""" First set of numerator parameters. """
return Tuple(*self.args[0][0])
def dsolve_system(eqs, funcs=None, t=None, ics=None, doit=False):
r"""
Solves any(supported) system of Ordinary Differential Equations
Explanation
===========
This function takes a system of ODEs as an input, determines if the
it is solvable by this function, and returns the solution if found any.
This function can handle:
1. Linear, First Order, Constant coefficient homogeneous system of ODEs
2. Linear, First Order, Constant coefficient non-homogeneous system of ODEs
3. Linear, First Order, non-constant coefficient homogeneous system of ODEs
4. Linear, First Order, non-constant coefficient non-homogeneous system of ODEs
5. Any implicit system which can be divided into system of ODEs which is of the above 4 forms
The types of systems described above aren't limited by the number of equations, i.e. this
function can solve the above types irrespective of the number of equations in the system passed.
This function returns a list of solutions. Each solution is a list of equations where LHS is
the dependent variable and RHS is an expression in terms of the independent variable.
Parameters
==========
eqs : List
system of ODEs to be solved
funcs : List or None
List of dependent variables that make up the system of ODEs
t : Symbol
Independent variable in the system of ODEs
ics : Dict or None
Set of initial boundary/conditions for the system of ODEs
doit : Boolean
Evaluate the solutions if True. Default value is False
Examples
========
>>> from sympy import symbols, Eq, Function
>>> from sympy.solvers.ode.systems import dsolve_system
>>> f, g = symbols("f g", cls=Function)
>>> x = symbols("x")
>>> eqs = [Eq(f(x).diff(x), g(x)), Eq(g(x).diff(x), f(x))]
>>> dsolve_system(eqs)
[[Eq(f(x), -C1*exp(-x) + C2*exp(x)), Eq(g(x), C1*exp(-x) + C2*exp(x))]]
You can also pass the initial conditions for the system of ODEs:
>>> dsolve_system(eqs, ics={f(0): 1, g(0): 0})
[[Eq(f(x), exp(x)/2 + exp(-x)/2), Eq(g(x), exp(x)/2 - exp(-x)/2)]]
Optionally, you can pass the dependent variables and the independent
variable for which the system is to be solved:
>>> funcs = [f(x), g(x)]
>>> dsolve_system(eqs, funcs=funcs, t=x)
[[Eq(f(x), -C1*exp(-x) + C2*exp(x)), Eq(g(x), C1*exp(-x) + C2*exp(x))]]
Lets look at an implicit system of ODEs:
>>> eqs = [Eq(f(x).diff(x)**2, g(x)**2), Eq(g(x).diff(x), g(x))]
>>> dsolve_system(eqs)
[[Eq(f(x), C1 - C2*exp(x)), Eq(g(x), C2*exp(x))], [Eq(f(x), C1 + C2*exp(x)), Eq(g(x), C2*exp(x))]]
Returns
=======
List of List of Equations
Raises
======
NotImplementedError
When the system of ODEs is not solvable by this function.
ValueError
When the parameters passed aren't in the required form.
"""
from sympy.solvers.ode.ode import solve_ics, _extract_funcs, constant_renumber
if not iterable(eqs):
raise ValueError(
filldedent('''
List of equations should be passed. The input is not valid.
'''))
eqs = _preprocess_eqs(eqs)
if funcs is not None and not isinstance(funcs, list):
raise ValueError(
filldedent('''
Input to the funcs should be a list of functions.
'''))
if funcs is None:
funcs = _extract_funcs(eqs)
if any(len(func.args) != 1 for func in funcs):
raise ValueError(
filldedent('''
dsolve_system can solve a system of ODEs with only one independent
variable.
'''))
if len(eqs) != len(funcs):
raise ValueError(
filldedent('''
Number of equations and number of functions don't match
'''))
if t is not None and not isinstance(t, Symbol):
raise ValueError(
filldedent('''
The indepedent variable must be of type Symbol
'''))
if t is None:
t = list(list(eqs[0].atoms(Derivative))[0].atoms(Symbol))[0]
sys_order = _get_func_order(eqs, funcs)
# To add higher order to first order reduction later
if not all(sys_order[func] == 1 for func in funcs):
raise NotImplementedError(
filldedent('''
Higher order ODEs aren't solvable by dsolve_system
'''))
canon_eqs = canonical_odes(eqs, funcs, t)
sols = []
for canon_eq in canon_eqs:
sol = _strong_component_solver(canon_eq, funcs, t)
if sol is None:
sol = _component_solver(canon_eq, funcs, t)
sols.append(sol)
if sols:
final_sols = []
for sol in sols:
# To preserve the order corresponding to the
# funcs list.
sol_dict = {s.lhs: s.rhs for s in sol}
sol = [Eq(var, sol_dict[var]) for var in funcs]
variables = Tuple(*eqs).free_symbols
sol = constant_renumber(sol, variables=variables)
if ics:
constants = Tuple(*sol).free_symbols - variables
solved_constants = solve_ics(sol, funcs, constants, ics)
sol = [s.subs(solved_constants) for s in sol]
if doit:
sol = [s.doit() for s in sol]
final_sols.append(sol)
sols = final_sols
return sols
def __init__(self, data, **kwarg):
"""
Creates a TableForm.
Parameters:
data ...
2D data to be put into the table; data can be
given as a Matrix
headings ...
gives the labels for rows and columns:
Can be a single argument that applies to both
dimensions:
- None ... no labels
- "automatic" ... labels are 1, 2, 3, ...
Can be a list of labels for rows and columns:
The labels for each dimension can be given
as None, "automatic", or [l1, l2, ...] e.g.
["automatic", None] will number the rows
[default: None]
alignments ...
alignment of the columns with:
- "left" or "<"
- "center" or "^"
- "right" or ">"
When given as a single value, the value is used for
all columns. The row headings (if given) will be
right justified unless an explicit alignment is
given for it and all other columns.
[default: "left"]
formats ...
a list of format strings or functions that accept
3 arguments (entry, row number, col number) and
return a string for the table entry. (If a function
returns None then the _print method will be used.)
wipe_zeros ...
Don't show zeros in the table.
[default: True]
pad ...
the string to use to indicate a missing value (e.g.
elements that are None or those that are missing
from the end of a row (i.e. any row that is shorter
than the rest is assumed to have missing values).
When None, nothing will be shown for values that
are missing from the end of a row; values that are
None, however, will be shown.
[default: None]
Examples
========
>>> from sympy import TableForm, Matrix
>>> TableForm([[5, 7], [4, 2], [10, 3]])
5 7
4 2
10 3
>>> TableForm([list('.'*i) for i in range(1, 4)], headings='automatic')
| 1 2 3
---------
1 | .
2 | . .
3 | . . .
>>> TableForm([['.'*(j if not i%2 else 1) for i in range(3)]
... for j in range(4)], alignments='rcl')
.
. . .
.. . ..
... . ...
"""
from sympy import Symbol, S, Matrix
from sympy.core.sympify import SympifyError
# We only support 2D data. Check the consistency:
if isinstance(data, Matrix):
data = data.tolist()
_h = len(data)
# fill out any short lines
pad = kwarg.get("pad", None)
ok_None = False
if pad is None:
pad = " "
ok_None = True
pad = Symbol(pad)
_w = max(len(line) for line in data)
for i, line in enumerate(data):
if len(line) != _w:
line.extend([pad] * (_w - len(line)))
for j, lj in enumerate(line):
if lj is None:
if not ok_None:
lj = pad
else:
try:
lj = S(lj)
except SympifyError:
lj = Symbol(str(lj))
line[j] = lj
data[i] = line
_lines = Tuple(*data)
headings = kwarg.get("headings", [None, None])
if headings == "automatic":
_headings = [range(1, _h + 1), range(1, _w + 1)]
else:
h1, h2 = headings
if h1 == "automatic":
h1 = range(1, _h + 1)
if h2 == "automatic":
h2 = range(1, _w + 1)
_headings = [h1, h2]
allow = ("l", "r", "c")
alignments = kwarg.get("alignments", "l")
def _std_align(a):
a = a.strip().lower()
if len(a) > 1:
return {"left": "l", "right": "r", "center": "c"}.get(a, a)
else:
return {"<": "l", ">": "r", "^": "c"}.get(a, a)
std_align = _std_align(alignments)
if std_align in allow:
_alignments = [std_align] * _w
else:
_alignments = []
for a in alignments:
std_align = _std_align(a)
_alignments.append(std_align)
if std_align not in ("l", "r", "c"):
raise ValueError('alignment "%s" unrecognized' %
alignments)
if _headings[0] and len(_alignments) == _w + 1:
_head_align = _alignments[0]
_alignments = _alignments[1:]
else:
_head_align = "r"
if len(_alignments) != _w:
raise ValueError(
"wrong number of alignments: expected %s but got %s" %
(_w, len(_alignments)))
_column_formats = kwarg.get("formats", [None] * _w)
_wipe_zeros = kwarg.get("wipe_zeros", True)
self._w = _w
self._h = _h
self._lines = _lines
self._headings = _headings
self._head_align = _head_align
self._alignments = _alignments
self._column_formats = _column_formats
self._wipe_zeros = _wipe_zeros
def disambiguate(*iter):
"""
Return a Tuple containing the passed expressions with symbols
that appear the same when printed replaced with numerically
subscripted symbols, and all Dummy symbols replaced with Symbols.
Parameters
==========
iter: list of symbols or expressions.
Examples
========
>>> from sympy.core.symbol import disambiguate
>>> from sympy import Dummy, Symbol, Tuple
>>> from sympy.abc import y
>>> tup = Symbol('_x'), Dummy('x'), Dummy('x')
>>> disambiguate(*tup)
(x_2, x, x_1)
>>> eqs = Tuple(Symbol('x')/y, Dummy('x')/y)
>>> disambiguate(*eqs)
(x_1/y, x/y)
>>> ix = Symbol('x', integer=True)
>>> vx = Symbol('x')
>>> disambiguate(vx + ix)
(x + x_1,)
To make your own mapping of symbols to use, pass only the free symbols
of the expressions and create a dictionary:
>>> free = eqs.free_symbols
>>> mapping = dict(zip(free, disambiguate(*free)))
>>> eqs.xreplace(mapping)
(x_1/y, x/y)
"""
new_iter = Tuple(*iter)
key = lambda x:tuple(sorted(x.assumptions0.items()))
syms = ordered(new_iter.free_symbols, keys=key)
mapping = {}
for s in syms:
mapping.setdefault(str(s).lstrip('_'), []).append(s)
reps = {}
for k in mapping:
# the first or only symbol doesn't get subscripted but make
# sure that it's a Symbol, not a Dummy
mapk0 = Symbol("%s" % (k), **mapping[k][0].assumptions0)
if mapping[k][0] != mapk0:
reps[mapping[k][0]] = mapk0
# the others get subscripts (and are made into Symbols)
skip = 0
for i in range(1, len(mapping[k])):
while True:
name = "%s_%i" % (k, i + skip)
if name not in mapping:
break
skip += 1
ki = mapping[k][i]
reps[ki] = Symbol(name, **ki.assumptions0)
return new_iter.xreplace(reps)
def test_nested():
expr = Basic(S(1), Basic(S(2)), S(3))
cmpd = Compound(Basic, (S(1), Compound(Basic, Tuple(2)), S(3)))
assert deconstruct(expr) == cmpd
assert construct(cmpd) == expr
def cse(exprs,
symbols=None,
optimizations=None,
postprocess=None,
order='canonical',
ignore=(),
list=True):
""" Perform common subexpression elimination on an expression.
Parameters
==========
exprs : list of SymPy expressions, or a single SymPy expression
The expressions to reduce.
symbols : infinite iterator yielding unique Symbols
The symbols used to label the common subexpressions which are pulled
out. The ``numbered_symbols`` generator is useful. The default is a
stream of symbols of the form "x0", "x1", etc. This must be an
infinite iterator.
optimizations : list of (callable, callable) pairs
The (preprocessor, postprocessor) pairs of external optimization
functions. Optionally 'basic' can be passed for a set of predefined
basic optimizations. Such 'basic' optimizations were used by default
in old implementation, however they can be really slow on larger
expressions. Now, no pre or post optimizations are made by default.
postprocess : a function which accepts the two return values of cse and
returns the desired form of output from cse, e.g. if you want the
replacements reversed the function might be the following lambda:
lambda r, e: return reversed(r), e
order : string, 'none' or 'canonical'
The order by which Mul and Add arguments are processed. If set to
'canonical', arguments will be canonically ordered. If set to 'none',
ordering will be faster but dependent on expressions hashes, thus
machine dependent and variable. For large expressions where speed is a
concern, use the setting order='none'.
ignore : iterable of Symbols
Substitutions containing any Symbol from ``ignore`` will be ignored.
list : bool, (default True)
Returns expression in list or else with same type as input (when False).
Returns
=======
replacements : list of (Symbol, expression) pairs
All of the common subexpressions that were replaced. Subexpressions
earlier in this list might show up in subexpressions later in this
list.
reduced_exprs : list of SymPy expressions
The reduced expressions with all of the replacements above.
Examples
========
>>> from sympy import cse, SparseMatrix
>>> from sympy.abc import x, y, z, w
>>> cse(((w + x + y + z)*(w + y + z))/(w + x)**3)
([(x0, y + z), (x1, w + x)], [(w + x0)*(x0 + x1)/x1**3])
List of expressions with recursive substitutions:
>>> m = SparseMatrix([x + y, x + y + z])
>>> cse([(x+y)**2, x + y + z, y + z, x + z + y, m])
([(x0, x + y), (x1, x0 + z)], [x0**2, x1, y + z, x1, Matrix([
[x0],
[x1]])])
Note: the type and mutability of input matrices is retained.
>>> isinstance(_[1][-1], SparseMatrix)
True
The user may disallow substitutions containing certain symbols:
>>> cse([y**2*(x + 1), 3*y**2*(x + 1)], ignore=(y,))
([(x0, x + 1)], [x0*y**2, 3*x0*y**2])
The default return value for the reduced expression(s) is a list, even if there is only
one expression. The `list` flag preserves the type of the input in the output:
>>> cse(x)
([], [x])
>>> cse(x, list=False)
([], x)
"""
from sympy.matrices import (MatrixBase, Matrix, ImmutableMatrix,
SparseMatrix, ImmutableSparseMatrix)
if not list:
return _cse_homogeneous(exprs,
symbols=symbols,
optimizations=optimizations,
postprocess=postprocess,
order=order,
ignore=ignore)
if isinstance(exprs, (int, float)):
exprs = sympify(exprs)
# Handle the case if just one expression was passed.
if isinstance(exprs, (Basic, MatrixBase)):
exprs = [exprs]
copy = exprs
temp = []
for e in exprs:
if isinstance(e, (Matrix, ImmutableMatrix)):
temp.append(Tuple(*e.flat()))
elif isinstance(e, (SparseMatrix, ImmutableSparseMatrix)):
temp.append(Tuple(*e.todok().items()))
else:
temp.append(e)
exprs = temp
del temp
if optimizations is None:
optimizations = []
elif optimizations == 'basic':
optimizations = basic_optimizations
# Preprocess the expressions to give us better optimization opportunities.
reduced_exprs = [preprocess_for_cse(e, optimizations) for e in exprs]
if symbols is None:
symbols = numbered_symbols(cls=Symbol)
else:
# In case we get passed an iterable with an __iter__ method instead of
# an actual iterator.
symbols = iter(symbols)
# Find other optimization opportunities.
opt_subs = opt_cse(reduced_exprs, order)
# Main CSE algorithm.
replacements, reduced_exprs = tree_cse(reduced_exprs, symbols, opt_subs,
order, ignore)
# Postprocess the expressions to return the expressions to canonical form.
exprs = copy
for i, (sym, subtree) in enumerate(replacements):
subtree = postprocess_for_cse(subtree, optimizations)
replacements[i] = (sym, subtree)
reduced_exprs = [
postprocess_for_cse(e, optimizations) for e in reduced_exprs
]
# Get the matrices back
for i, e in enumerate(exprs):
if isinstance(e, (Matrix, ImmutableMatrix)):
reduced_exprs[i] = Matrix(e.rows, e.cols, reduced_exprs[i])
if isinstance(e, ImmutableMatrix):
reduced_exprs[i] = reduced_exprs[i].as_immutable()
elif isinstance(e, (SparseMatrix, ImmutableSparseMatrix)):
m = SparseMatrix(e.rows, e.cols, {})
for k, v in reduced_exprs[i]:
m[k] = v
if isinstance(e, ImmutableSparseMatrix):
m = m.as_immutable()
reduced_exprs[i] = m
if postprocess is None:
return replacements, reduced_exprs
return postprocess(replacements, reduced_exprs)
def test_2d_scalar_2():
print('============== test_2d_scalar_2 ================')
x, y = symbols('x y')
u = Symbol('u')
v = Symbol('v')
c = Constant('c')
b0 = Constant('b0')
b1 = Constant('b1')
b = Tuple(b0, b1)
a = Lambda((x, y, v, u),
c * u * v + Dot(b, Grad(v)) * u + Dot(b, Grad(u)) * v)
print('> input := {0}'.format(a))
# ... create a finite element space
p1 = 2
p2 = 2
ne1 = 8
ne2 = 8
print('> Grid :: [{ne1},{ne2}]'.format(ne1=ne1, ne2=ne2))
print('> Degree :: [{p1},{p2}]'.format(p1=p1, p2=p2))
grid_1 = linspace(0., 1., ne1 + 1)
grid_2 = linspace(0., 1., ne2 + 1)
V1 = SplineSpace(p1, grid=grid_1)
V2 = SplineSpace(p2, grid=grid_2)
V = TensorFemSpace(V1, V2)
# ...
# ... create a glt symbol from a string without evaluation
expr = glt_symbol(a, space=V)
print('> glt symbol := {0}'.format(expr))
# ...
# ...
symbol_f90 = compile_symbol('symbol_scalar_2',
a,
V,
d_constants={
'b0': 0.1,
'b1': 1.,
'c': 0.2
},
backend='fortran')
# ...
# ... example of symbol evaluation
t1 = linspace(-pi, pi, ne1 + 1)
t2 = linspace(-pi, pi, ne2 + 1)
x1 = linspace(0., 1., ne1 + 1)
x2 = linspace(0., 1., ne2 + 1)
e = zeros((ne1 + 1, ne2 + 1), order='F')
symbol_f90(x1, x2, t1, t2, e)
# ...
print('')
def _process_limits(*symbols, discrete=None):
"""Process the list of symbols and convert them to canonical limits,
storing them as Tuple(symbol, lower, upper). The orientation of
the function is also returned when the upper limit is missing
so (x, 1, None) becomes (x, None, 1) and the orientation is changed.
In the case that a limit is specified as (symbol, Range), a list of
length 4 may be returned if a change of variables is needed; the
expression that should replace the symbol in the expression is
the fourth element in the list.
"""
limits = []
orientation = 1
if discrete is None:
err_msg = 'discrete must be True or False'
elif discrete:
err_msg = 'use Range, not Interval or Relational'
else:
err_msg = 'use Interval or Relational, not Range'
for V in symbols:
if isinstance(V, (Relational, BooleanFunction)):
if discrete:
raise TypeError(err_msg)
variable = V.atoms(Symbol).pop()
V = (variable, V.as_set())
elif isinstance(V, Symbol) or getattr(V, '_diff_wrt', False):
if isinstance(V, Idx):
if V.lower is None or V.upper is None:
limits.append(Tuple(V))
else:
limits.append(Tuple(V, V.lower, V.upper))
else:
limits.append(Tuple(V))
continue
if is_sequence(V) and not isinstance(V, Set):
if len(V) == 2 and isinstance(V[1], Set):
V = list(V)
if isinstance(V[1], Interval): # includes Reals
if discrete:
raise TypeError(err_msg)
V[1:] = V[1].inf, V[1].sup
elif isinstance(V[1], Range):
if not discrete:
raise TypeError(err_msg)
lo = V[1].inf
hi = V[1].sup
dx = abs(V[1].step) # direction doesn't matter
if dx == 1:
V[1:] = [lo, hi]
else:
if lo is not S.NegativeInfinity:
V = [V[0]] + [0, (hi - lo) // dx, dx * V[0] + lo]
else:
V = [V[0]] + [0, S.Infinity, -dx * V[0] + hi]
else:
# more complicated sets would require splitting, e.g.
# Union(Interval(1, 3), interval(6,10))
raise NotImplementedError(
'expecting Range'
if discrete else 'Relational or single Interval')
V = sympify(flatten(V)) # list of sympified elements/None
if isinstance(V[0],
(Symbol, Idx)) or getattr(V[0], '_diff_wrt', False):
newsymbol = V[0]
if len(V) == 3:
# general case
if V[2] is None and V[1] is not None:
orientation *= -1
V = [newsymbol] + [i for i in V[1:] if i is not None]
lenV = len(V)
if not isinstance(newsymbol, Idx) or lenV == 3:
if lenV == 4:
limits.append(Tuple(*V))
continue
if lenV == 3:
if isinstance(newsymbol, Idx):
# Idx represents an integer which may have
# specified values it can take on; if it is
# given such a value, an error is raised here
# if the summation would try to give it a larger
# or smaller value than permitted. None and Symbolic
# values will not raise an error.
lo, hi = newsymbol.lower, newsymbol.upper
try:
if lo is not None and not bool(V[1] >= lo):
raise ValueError(
"Summation will set Idx value too low."
)
except TypeError:
pass
try:
if hi is not None and not bool(V[2] <= hi):
raise ValueError(
"Summation will set Idx value too high."
)
except TypeError:
pass
limits.append(Tuple(*V))
continue
if lenV == 1 or (lenV == 2 and V[1] is None):
limits.append(Tuple(newsymbol))
continue
elif lenV == 2:
limits.append(Tuple(newsymbol, V[1]))
continue
raise ValueError('Invalid limits given: %s' % str(symbols))
return limits, orientation
def _common_new(cls, function, *symbols, **assumptions):
"""Return either a special return value or the tuple,
(function, limits, orientation). This code is common to
both ExprWithLimits and AddWithLimits."""
function = sympify(function)
if isinstance(function, Equality):
# This transforms e.g. Integral(Eq(x, y)) to Eq(Integral(x), Integral(y))
# but that is only valid for definite integrals.
limits, orientation = _process_limits(*symbols)
if not (limits and all(len(limit) == 3 for limit in limits)):
SymPyDeprecationWarning(
feature='Integral(Eq(x, y))',
useinstead='Eq(Integral(x, z), Integral(y, z))',
issue=18053,
deprecated_since_version=1.6,
).warn()
lhs = function.lhs
rhs = function.rhs
return Equality(cls(lhs, *symbols, **assumptions), \
cls(rhs, *symbols, **assumptions))
if function is S.NaN:
return S.NaN
if symbols:
limits, orientation = _process_limits(*symbols)
for i, li in enumerate(limits):
if len(li) == 4:
function = function.subs(li[0], li[-1])
limits[i] = Tuple(*li[:-1])
else:
# symbol not provided -- we can still try to compute a general form
free = function.free_symbols
if len(free) != 1:
raise ValueError(
"specify dummy variables for %s" % function)
limits, orientation = [Tuple(s) for s in free], 1
# denest any nested calls
while cls == type(function):
limits = list(function.limits) + limits
function = function.function
# Any embedded piecewise functions need to be brought out to the
# top level. We only fold Piecewise that contain the integration
# variable.
reps = {}
symbols_of_integration = {i[0] for i in limits}
for p in function.atoms(Piecewise):
if not p.has(*symbols_of_integration):
reps[p] = Dummy()
# mask off those that don't
function = function.xreplace(reps)
# do the fold
function = piecewise_fold(function)
# remove the masking
function = function.xreplace({v: k for k, v in reps.items()})
return function, limits, orientation
def _construct_declarations(cls, args):
return Tuple(*[Declaration(arg) for arg in args])
def _process_limits(*symbols):
"""Process the list of symbols and convert them to canonical limits,
storing them as Tuple(symbol, lower, upper). The orientation of
the function is also returned when the upper limit is missing
so (x, 1, None) becomes (x, None, 1) and the orientation is changed.
"""
limits = []
orientation = 1
for V in symbols:
if isinstance(V, (Relational, BooleanFunction)):
variable = V.atoms(Symbol).pop()
V = (variable, V.as_set())
if isinstance(V, Symbol) or getattr(V, '_diff_wrt', False):
if isinstance(V, Idx):
if V.lower is None or V.upper is None:
limits.append(Tuple(V))
else:
limits.append(Tuple(V, V.lower, V.upper))
else:
limits.append(Tuple(V))
continue
elif is_sequence(V, Tuple):
if len(V) == 2 and isinstance(V[1], Range):
lo = V[1].inf
hi = V[1].sup
dx = abs(V[1].step)
V = [V[0]] + [0, (hi - lo)//dx, dx*V[0] + lo]
V = sympify(flatten(V)) # a list of sympified elements
if isinstance(V[0], (Symbol, Idx)) or getattr(V[0], '_diff_wrt', False):
newsymbol = V[0]
if len(V) == 2 and isinstance(V[1], Interval): # 2 -> 3
# Interval
V[1:] = [V[1].start, V[1].end]
elif len(V) == 3:
# general case
if V[2] is None and not V[1] is None:
orientation *= -1
V = [newsymbol] + [i for i in V[1:] if i is not None]
if not isinstance(newsymbol, Idx) or len(V) == 3:
if len(V) == 4:
limits.append(Tuple(*V))
continue
if len(V) == 3:
if isinstance(newsymbol, Idx):
# Idx represents an integer which may have
# specified values it can take on; if it is
# given such a value, an error is raised here
# if the summation would try to give it a larger
# or smaller value than permitted. None and Symbolic
# values will not raise an error.
lo, hi = newsymbol.lower, newsymbol.upper
try:
if lo is not None and not bool(V[1] >= lo):
raise ValueError("Summation will set Idx value too low.")
except TypeError:
pass
try:
if hi is not None and not bool(V[2] <= hi):
raise ValueError("Summation will set Idx value too high.")
except TypeError:
pass
limits.append(Tuple(*V))
continue
if len(V) == 1 or (len(V) == 2 and V[1] is None):
limits.append(Tuple(newsymbol))
continue
elif len(V) == 2:
limits.append(Tuple(newsymbol, V[1]))
continue
raise ValueError('Invalid limits given: %s' % str(symbols))
return limits, orientation
def plot_implicit(expr,
x_var=None,
y_var=None,
adaptive=True,
depth=0,
points=300,
line_color="blue",
show=True,
**kwargs):
"""A plot function to plot implicit equations / inequalities.
Arguments
=========
- ``expr`` : The equation / inequality that is to be plotted.
- ``x_var`` (optional) : symbol to plot on x-axis or tuple giving symbol
and range as ``(symbol, xmin, xmax)``
- ``y_var`` (optional) : symbol to plot on y-axis or tuple giving symbol
and range as ``(symbol, ymin, ymax)``
If neither ``x_var`` nor ``y_var`` are given then the free symbols in the
expression will be assigned in the order they are sorted.
The following keyword arguments can also be used:
- ``adaptive`` Boolean. The default value is set to True. It has to be
set to False if you want to use a mesh grid.
- ``depth`` integer. The depth of recursion for adaptive mesh grid.
Default value is 0. Takes value in the range (0, 4).
- ``points`` integer. The number of points if adaptive mesh grid is not
used. Default value is 300.
- ``show`` Boolean. Default value is True. If set to False, the plot will
not be shown. See ``Plot`` for further information.
- ``title`` string. The title for the plot.
- ``xlabel`` string. The label for the x-axis
- ``ylabel`` string. The label for the y-axis
Aesthetics options:
- ``line_color``: float or string. Specifies the color for the plot.
See ``Plot`` to see how to set color for the plots.
Default value is "Blue"
plot_implicit, by default, uses interval arithmetic to plot functions. If
the expression cannot be plotted using interval arithmetic, it defaults to
a generating a contour using a mesh grid of fixed number of points. By
setting adaptive to False, you can force plot_implicit to use the mesh
grid. The mesh grid method can be effective when adaptive plotting using
interval arithmetic, fails to plot with small line width.
Examples
========
Plot expressions:
.. plot::
:context: reset
:format: doctest
:include-source: True
>>> from sympy import plot_implicit, symbols, Eq, And
>>> x, y = symbols('x y')
Without any ranges for the symbols in the expression:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p1 = plot_implicit(Eq(x**2 + y**2, 5))
With the range for the symbols:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p2 = plot_implicit(
... Eq(x**2 + y**2, 3), (x, -3, 3), (y, -3, 3))
With depth of recursion as argument:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p3 = plot_implicit(
... Eq(x**2 + y**2, 5), (x, -4, 4), (y, -4, 4), depth = 2)
Using mesh grid and not using adaptive meshing:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p4 = plot_implicit(
... Eq(x**2 + y**2, 5), (x, -5, 5), (y, -2, 2),
... adaptive=False)
Using mesh grid without using adaptive meshing with number of points
specified:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p5 = plot_implicit(
... Eq(x**2 + y**2, 5), (x, -5, 5), (y, -2, 2),
... adaptive=False, points=400)
Plotting regions:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p6 = plot_implicit(y > x**2)
Plotting Using boolean conjunctions:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p7 = plot_implicit(And(y > x, y > -x))
When plotting an expression with a single variable (y - 1, for example),
specify the x or the y variable explicitly:
.. plot::
:context: close-figs
:format: doctest
:include-source: True
>>> p8 = plot_implicit(y - 1, y_var=y)
>>> p9 = plot_implicit(x - 1, x_var=x)
"""
has_equality = False # Represents whether the expression contains an Equality,
#GreaterThan or LessThan
def arg_expand(bool_expr):
"""
Recursively expands the arguments of an Boolean Function
"""
for arg in bool_expr.args:
if isinstance(arg, BooleanFunction):
arg_expand(arg)
elif isinstance(arg, Relational):
arg_list.append(arg)
arg_list = []
if isinstance(expr, BooleanFunction):
arg_expand(expr)
#Check whether there is an equality in the expression provided.
if any(
isinstance(e, (Equality, GreaterThan, LessThan))
for e in arg_list):
has_equality = True
elif not isinstance(expr, Relational):
expr = Eq(expr, 0)
has_equality = True
elif isinstance(expr, (Equality, GreaterThan, LessThan)):
has_equality = True
xyvar = [i for i in (x_var, y_var) if i is not None]
free_symbols = expr.free_symbols
range_symbols = Tuple(*flatten(xyvar)).free_symbols
undeclared = free_symbols - range_symbols
if len(free_symbols & range_symbols) > 2:
raise NotImplementedError("Implicit plotting is not implemented for "
"more than 2 variables")
#Create default ranges if the range is not provided.
default_range = Tuple(-5, 5)
def _range_tuple(s):
if isinstance(s, Symbol):
return Tuple(s) + default_range
if len(s) == 3:
return Tuple(*s)
raise ValueError('symbol or `(symbol, min, max)` expected but got %s' %
s)
if len(xyvar) == 0:
xyvar = list(_sort_gens(free_symbols))
var_start_end_x = _range_tuple(xyvar[0])
x = var_start_end_x[0]
if len(xyvar) != 2:
if x in undeclared or not undeclared:
xyvar.append(Dummy('f(%s)' % x.name))
else:
xyvar.append(undeclared.pop())
var_start_end_y = _range_tuple(xyvar[1])
#Check whether the depth is greater than 4 or less than 0.
if depth > 4:
depth = 4
elif depth < 0:
depth = 0
series_argument = ImplicitSeries(expr, var_start_end_x, var_start_end_y,
has_equality, adaptive, depth, points,
line_color)
#set the x and y limits
kwargs['xlim'] = tuple(float(x) for x in var_start_end_x[1:])
kwargs['ylim'] = tuple(float(y) for y in var_start_end_y[1:])
# set the x and y labels
kwargs.setdefault('xlabel', var_start_end_x[0])
kwargs.setdefault('ylabel', var_start_end_y[0])
p = Plot(series_argument, **kwargs)
if show:
p.show()
return p
def bq(self):
""" Denominator parameters of the hypergeometric function. """
return Tuple(*self.args[1])
def tr(p):
if len(p) != 2:
raise TypeError("wrong argument")
return Tuple(_prep_tuple(p[0]), _prep_tuple(p[1]))
def ap(self):
""" Combined numerator parameters. """
return Tuple(*(self.args[0][0] + self.args[0][1]))
def __new__(cls, base_dims, derived_dims=(), dimensional_dependencies={}, name=None, descr=None):
dimensional_dependencies = dict(dimensional_dependencies)
if (name is not None) or (descr is not None):
SymPyDeprecationWarning(
deprecated_since_version="1.2",
issue=13336,
useinstead="do not define a `name` or `descr`",
).warn()
def parse_dim(dim):
if isinstance(dim, str):
dim = Dimension(Symbol(dim))
elif isinstance(dim, Dimension):
pass
elif isinstance(dim, Symbol):
dim = Dimension(dim)
else:
raise TypeError("%s wrong type" % dim)
return dim
base_dims = [parse_dim(i) for i in base_dims]
derived_dims = [parse_dim(i) for i in derived_dims]
for dim in base_dims:
dim = dim.name
if (dim in dimensional_dependencies
and (len(dimensional_dependencies[dim]) != 1 or
dimensional_dependencies[dim].get(dim, None) != 1)):
raise IndexError("Repeated value in base dimensions")
dimensional_dependencies[dim] = Dict({dim: 1})
def parse_dim_name(dim):
if isinstance(dim, Dimension):
return dim.name
elif isinstance(dim, str):
return Symbol(dim)
elif isinstance(dim, Symbol):
return dim
else:
raise TypeError("unrecognized type %s for %s" % (type(dim), dim))
for dim in dimensional_dependencies.keys():
dim = parse_dim(dim)
if (dim not in derived_dims) and (dim not in base_dims):
derived_dims.append(dim)
def parse_dict(d):
return Dict({parse_dim_name(i): j for i, j in d.items()})
# Make sure everything is a SymPy type:
dimensional_dependencies = {parse_dim_name(i): parse_dict(j) for i, j in
dimensional_dependencies.items()}
for dim in derived_dims:
if dim in base_dims:
raise ValueError("Dimension %s both in base and derived" % dim)
if dim.name not in dimensional_dependencies:
# TODO: should this raise a warning?
dimensional_dependencies[dim.name] = Dict({dim.name: 1})
base_dims.sort(key=default_sort_key)
derived_dims.sort(key=default_sort_key)
base_dims = Tuple(*base_dims)
derived_dims = Tuple(*derived_dims)
dimensional_dependencies = Dict({i: Dict(j) for i, j in dimensional_dependencies.items()})
obj = Basic.__new__(cls, base_dims, derived_dims, dimensional_dependencies)
return obj
def bm(self):
""" First set of denominator parameters. """
return Tuple(*self.args[1][0])
def test_count_ops_visual():
ADD, MUL, POW, SIN, COS, EXP, AND, D, G = symbols(
'Add Mul Pow sin cos exp And Derivative Integral'.upper())
DIV, SUB, NEG = symbols('DIV SUB NEG')
NOT, OR, AND, XOR, IMPLIES, EQUIVALENT, ITE, BASIC, TUPLE = symbols(
'Not Or And Xor Implies Equivalent ITE Basic Tuple'.upper())
def count(val):
return count_ops(val, visual=True)
assert count(7) is S.Zero
assert count(S(7)) is S.Zero
assert count(-1) == NEG
assert count(-2) == NEG
assert count(S(2) / 3) == DIV
assert count(pi / 3) == DIV
assert count(-pi / 3) == DIV + NEG
assert count(I - 1) == SUB
assert count(1 - I) == SUB
assert count(1 - 2 * I) == SUB + MUL
assert count(x) is S.Zero
assert count(-x) == NEG
assert count(-2 * x / 3) == NEG + DIV + MUL
assert count(1 / x) == DIV
assert count(1 / (x * y)) == DIV + MUL
assert count(-1 / x) == NEG + DIV
assert count(-2 / x) == NEG + DIV
assert count(x / y) == DIV
assert count(-x / y) == NEG + DIV
assert count(x**2) == POW
assert count(-x**2) == POW + NEG
assert count(-2 * x**2) == POW + MUL + NEG
assert count(x + pi / 3) == ADD + DIV
assert count(x + S(1) / 3) == ADD + DIV
assert count(x + y) == ADD
assert count(x - y) == SUB
assert count(y - x) == SUB
assert count(-1 / (x - y)) == DIV + NEG + SUB
assert count(-1 / (y - x)) == DIV + NEG + SUB
assert count(1 + x**y) == ADD + POW
assert count(1 + x + y) == 2 * ADD
assert count(1 + x + y + z) == 3 * ADD
assert count(1 + x**y + 2 * x * y + y**2) == 3 * ADD + 2 * POW + 2 * MUL
assert count(2 * z + y + x + 1) == 3 * ADD + MUL
assert count(2 * z + y**17 + x + 1) == 3 * ADD + MUL + POW
assert count(2 * z + y**17 + x + sin(x)) == 3 * ADD + POW + MUL + SIN
assert count(2 * z + y**17 + x +
sin(x**2)) == 3 * ADD + MUL + 2 * POW + SIN
assert count(2 * z + y**17 + x + sin(x**2) +
exp(cos(x))) == 4 * ADD + MUL + 2 * POW + EXP + COS + SIN
assert count(Derivative(x, x)) == D
assert count(Integral(x, x) + 2 * x / (1 + x)) == G + DIV + MUL + 2 * ADD
assert count(Basic()) is S.Zero
assert count({x + 1: sin(x)}) == ADD + SIN
assert count([x + 1, sin(x) + y, None]) == ADD + SIN + ADD
assert count({x + 1: sin(x), y: cos(x) + 1}) == SIN + COS + 2 * ADD
assert count({}) is S.Zero
assert count([x + 1, sin(x) * y, None]) == SIN + ADD + MUL
assert count([]) is S.Zero
assert count(Basic()) == 0
assert count(Basic(Basic(), Basic(x, x + y))) == ADD + 2 * BASIC
assert count(Basic(x, x + y)) == ADD + BASIC
assert count(Or(x, y)) == OR
assert count(And(x, y)) == AND
assert count(And(x**y, z)) == AND + POW
assert count(Or(x, Or(y, And(z, a)))) == AND + OR
assert count(Nor(x, y)) == NOT + OR
assert count(Nand(x, y)) == NOT + AND
assert count(Xor(x, y)) == XOR
assert count(Implies(x, y)) == IMPLIES
assert count(Equivalent(x, y)) == EQUIVALENT
assert count(ITE(x, y, z)) == ITE
assert count([Or(x, y), And(x, y), Basic(x + y)]) == ADD + AND + BASIC + OR
assert count(Basic(Tuple(x))) == BASIC + TUPLE
#It checks that TUPLE is counted as an operation.
assert count(Eq(x + y, S(2))) == ADD
def bother(self):
""" Second set of denominator parameters. """
return Tuple(*self.args[1][1])
def __new__(cls, expr, *args, **kwargs):
expr = sympify(expr)
if not args:
if expr.is_Order:
variables = expr.variables
point = expr.point
else:
variables = list(expr.free_symbols)
point = [S.Zero] * len(variables)
else:
args = list(args if is_sequence(args) else [args])
variables, point = [], []
if is_sequence(args[0]):
for a in args:
v, p = list(map(sympify, a))
variables.append(v)
point.append(p)
else:
variables = list(map(sympify, args))
point = [S.Zero] * len(variables)
if not all(v.is_symbol for v in variables):
raise TypeError('Variables are not symbols, got %s' % variables)
if len(list(uniq(variables))) != len(variables):
raise ValueError(
'Variables are supposed to be unique symbols, got %s' %
variables)
if expr.is_Order:
expr_vp = dict(expr.args[1:])
new_vp = dict(expr_vp)
vp = dict(zip(variables, point))
for v, p in vp.items():
if v in new_vp.keys():
if p != new_vp[v]:
raise NotImplementedError(
"Mixing Order at different points is not supported."
)
else:
new_vp[v] = p
if set(expr_vp.keys()) == set(new_vp.keys()):
return expr
else:
variables = list(new_vp.keys())
point = [new_vp[v] for v in variables]
if expr is S.NaN:
return S.NaN
if any(x in p.free_symbols for x in variables for p in point):
raise ValueError('Got %s as a point.' % point)
if variables:
if any(p != point[0] for p in point):
raise NotImplementedError(
"Multivariable orders at different points are not supported."
)
if point[0] is S.Infinity:
s = {k: 1 / Dummy() for k in variables}
rs = {1 / v: 1 / k for k, v in s.items()}
elif point[0] is S.NegativeInfinity:
s = {k: -1 / Dummy() for k in variables}
rs = {-1 / v: -1 / k for k, v in s.items()}
elif point[0] is not S.Zero:
s = dict((k, Dummy() + point[0]) for k in variables)
rs = dict((v - point[0], k - point[0]) for k, v in s.items())
else:
s = ()
rs = ()
expr = expr.subs(s)
if expr.is_Add:
from sympy import expand_multinomial
expr = expand_multinomial(expr)
if s:
args = tuple([r[0] for r in rs.items()])
else:
args = tuple(variables)
if len(variables) > 1:
# XXX: better way? We need this expand() to
# workaround e.g: expr = x*(x + y).
# (x*(x + y)).as_leading_term(x, y) currently returns
# x*y (wrong order term!). That's why we want to deal with
# expand()'ed expr (handled in "if expr.is_Add" branch below).
expr = expr.expand()
old_expr = None
while old_expr != expr:
old_expr = expr
if expr.is_Add:
lst = expr.extract_leading_order(args)
expr = Add(*[f.expr for (e, f) in lst])
elif expr:
expr = expr.as_leading_term(*args)
expr = expr.as_independent(*args, as_Add=False)[1]
expr = expand_power_base(expr)
expr = expand_log(expr)
if len(args) == 1:
# The definition of O(f(x)) symbol explicitly stated that
# the argument of f(x) is irrelevant. That's why we can
# combine some power exponents (only "on top" of the
# expression tree for f(x)), e.g.:
# x**p * (-x)**q -> x**(p+q) for real p, q.
x = args[0]
margs = list(
Mul.make_args(
expr.as_independent(x, as_Add=False)[1]))
for i, t in enumerate(margs):
if t.is_Pow:
b, q = t.args
if b in (x, -x) and q.is_real and not q.has(x):
margs[i] = x**q
elif b.is_Pow and not b.exp.has(x):
b, r = b.args
if b in (x, -x) and r.is_real:
margs[i] = x**(r * q)
elif b.is_Mul and b.args[0] is S.NegativeOne:
b = -b
if b.is_Pow and not b.exp.has(x):
b, r = b.args
if b in (x, -x) and r.is_real:
margs[i] = x**(r * q)
expr = Mul(*margs)
expr = expr.subs(rs)
if expr.is_zero:
return expr
if expr.is_Order:
expr = expr.expr
if not expr.has(*variables):
expr = S.One
# create Order instance:
vp = dict(zip(variables, point))
variables.sort(key=default_sort_key)
point = [vp[v] for v in variables]
args = (expr, ) + Tuple(*zip(variables, point))
obj = Expr.__new__(cls, *args)
return obj
def _process_limits(*symbols):
"""Process the list of symbols and convert them to canonical limits,
storing them as Tuple(symbol, lower, upper). The orientation of
the function is also returned when the upper limit is missing
so (x, 1, None) becomes (x, None, 1) and the orientation is changed.
"""
limits = []
orientation = 1
for V in symbols:
if isinstance(V, (Relational, BooleanFunction)):
variable = V.atoms(Symbol).pop()
V = (variable, V.as_set())
if isinstance(V, Symbol) or getattr(V, '_diff_wrt', False):
if isinstance(V, Idx):
if V.lower is None or V.upper is None:
limits.append(Tuple(V))
else:
limits.append(Tuple(V, V.lower, V.upper))
else:
limits.append(Tuple(V))
continue
elif is_sequence(V, Tuple):
V = sympify(flatten(V))
if isinstance(V[0],
(Symbol, Idx)) or getattr(V[0], '_diff_wrt', False):
newsymbol = V[0]
if len(V) == 2 and isinstance(V[1], Interval):
V[1:] = [V[1].start, V[1].end]
if len(V) == 3:
if V[1] is None and V[2] is not None:
nlim = [V[2]]
elif V[1] is not None and V[2] is None:
orientation *= -1
nlim = [V[1]]
elif V[1] is None and V[2] is None:
nlim = []
else:
nlim = V[1:]
limits.append(Tuple(newsymbol, *nlim))
if isinstance(V[0], Idx):
if V[0].lower is not None and not bool(
nlim[0] >= V[0].lower):
raise ValueError(
"Summation exceeds Idx lower range.")
if V[0].upper is not None and not bool(
nlim[1] <= V[0].upper):
raise ValueError(
"Summation exceeds Idx upper range.")
continue
elif len(V) == 1 or (len(V) == 2 and V[1] is None):
limits.append(Tuple(newsymbol))
continue
elif len(V) == 2:
limits.append(Tuple(newsymbol, V[1]))
continue
raise ValueError('Invalid limits given: %s' % str(symbols))
return limits, orientation
def __new__(cls):
x = C.Dummy('x')
#construct "by hand" to avoid infinite loop
return Expr.__new__(cls, Tuple(x), x)
def _eval_derivative(self, sym):
"""Evaluate the derivative of the current Integral object by
differentiating under the integral sign [1], using the Fundamental
Theorem of Calculus [2] when possible.
Whenever an Integral is encountered that is equivalent to zero or
has an integrand that is independent of the variable of integration
those integrals are performed. All others are returned as Integral
instances which can be resolved with doit() (provided they are integrable).
References:
[1] http://en.wikipedia.org/wiki/Differentiation_under_the_integral_sign
[2] http://en.wikipedia.org/wiki/Fundamental_theorem_of_calculus
Examples
========
>>> from sympy import Integral
>>> from sympy.abc import x, y
>>> i = Integral(x + y, y, (y, 1, x))
>>> i.diff(x)
Integral(x + y, (y, x)) + Integral(1, y, (y, 1, x))
>>> i.doit().diff(x) == i.diff(x).doit()
True
>>> i.diff(y)
0
The previous must be true since there is no y in the evaluated integral:
>>> i.free_symbols
set([x])
>>> i.doit()
2*x**3/3 - x/2 - 1/6
"""
# differentiate under the integral sign; we do not
# check for regularity conditions (TODO), see issue 4215
# get limits and the function
f, limits = self.function, list(self.limits)
# the order matters if variables of integration appear in the limits
# so work our way in from the outside to the inside.
limit = limits.pop(-1)
if len(limit) == 3:
x, a, b = limit
elif len(limit) == 2:
x, b = limit
a = None
else:
a = b = None
x = limit[0]
if limits: # f is the argument to an integral
f = self.func(f, *tuple(limits))
# assemble the pieces
def _do(f, ab):
dab_dsym = diff(ab, sym)
if not dab_dsym:
return S.Zero
if isinstance(f, Integral):
limits = [(x, x) if (len(l) == 1 and l[0] == x) else l
for l in f.limits]
f = self.func(f.function, *limits)
return f.subs(x, ab)*dab_dsym
rv = 0
if b is not None:
rv += _do(f, b)
if a is not None:
rv -= _do(f, a)
if len(limit) == 1 and sym == x:
# the dummy variable *is* also the real-world variable
arg = f
rv += arg
else:
# the dummy variable might match sym but it's
# only a dummy and the actual variable is determined
# by the limits, so mask off the variable of integration
# while differentiating
u = Dummy('u')
arg = f.subs(x, u).diff(sym).subs(u, x)
rv += self.func(arg, Tuple(x, a, b))
return rv
def test_factor_terms():
A = Symbol('A', commutative=False)
assert factor_terms(9*(x + x*y + 1) + (3*x + 3)**(2 + 2*x)) == \
9*x*y + 9*x + _keep_coeff(S(3), x + 1)**_keep_coeff(S(2), x + 1) + 9
assert factor_terms(9*(x + x*y + 1) + (3)**(2 + 2*x)) == \
_keep_coeff(S(9), 3**(2*x) + x*y + x + 1)
assert factor_terms(3**(2 + 2*x) + a*3**(2 + 2*x)) == \
9*3**(2*x)*(a + 1)
assert factor_terms(x + x*A) == \
x*(1 + A)
assert factor_terms(sin(x + x*A)) == \
sin(x*(1 + A))
assert factor_terms((3*x + 3)**((2 + 2*x)/3)) == \
_keep_coeff(S(3), x + 1)**_keep_coeff(Rational(2, 3), x + 1)
assert factor_terms(x + (x*y + x)**(3*x + 3)) == \
x + (x*(y + 1))**_keep_coeff(S(3), x + 1)
assert factor_terms(a*(x + x*y) + b*(x*2 + y*x*2)) == \
x*(a + 2*b)*(y + 1)
i = Integral(x, (x, 0, oo))
assert factor_terms(i) == i
assert factor_terms(x / 2 + y) == x / 2 + y
# fraction doesn't apply to integer denominators
assert factor_terms(x / 2 + y, fraction=True) == x / 2 + y
# clear *does* apply to the integer denominators
assert factor_terms(x / 2 + y, clear=True) == Mul(S.Half,
x + 2 * y,
evaluate=False)
# check radical extraction
eq = sqrt(2) + sqrt(10)
assert factor_terms(eq) == eq
assert factor_terms(eq, radical=True) == sqrt(2) * (1 + sqrt(5))
eq = root(-6, 3) + root(6, 3)
assert factor_terms(
eq, radical=True) == 6**(S.One / 3) * (1 + (-1)**(S.One / 3))
eq = [x + x * y]
ans = [x * (y + 1)]
for c in [list, tuple, set]:
assert factor_terms(c(eq)) == c(ans)
assert factor_terms(Tuple(x + x * y)) == Tuple(x * (y + 1))
assert factor_terms(Interval(0, 1)) == Interval(0, 1)
e = 1 / sqrt(a / 2 + 1)
assert factor_terms(e, clear=False) == 1 / sqrt(a / 2 + 1)
assert factor_terms(e, clear=True) == sqrt(2) / sqrt(a + 2)
eq = x / (x + 1 / x) + 1 / (x**2 + 1)
assert factor_terms(eq, fraction=False) == eq
assert factor_terms(eq, fraction=True) == 1
assert factor_terms((1/(x**3 + x**2) + 2/x**2)*y) == \
y*(2 + 1/(x + 1))/x**2
# if not True, then processesing for this in factor_terms is not necessary
assert gcd_terms(-x - y) == -x - y
assert factor_terms(-x - y) == Mul(-1, x + y, evaluate=False)
# if not True, then "special" processesing in factor_terms is not necessary
assert gcd_terms(exp(Mul(-1, x + 1))) == exp(-x - 1)
e = exp(-x - 2) + x
assert factor_terms(e) == exp(Mul(-1, x + 2, evaluate=False)) + x
assert factor_terms(e, sign=False) == e
assert factor_terms(exp(-4 * x - 2) -
x) == -x + exp(Mul(-2, 2 * x + 1, evaluate=False))
# sum/integral tests
for F in (Sum, Integral):
assert factor_terms(F(x, (y, 1, 10))) == x * F(1, (y, 1, 10))
assert factor_terms(F(x, (y, 1, 10)) + x) == x * (1 + F(1, (y, 1, 10)))
assert factor_terms(F(x * y + x * y**2,
(y, 1, 10))) == x * F(y * (y + 1), (y, 1, 10))
# expressions involving Pow terms with base 0
assert factor_terms(0**(x - 2) - 1) == 0**(x - 2) - 1
assert factor_terms(0**(x + 2) - 1) == 0**(x + 2) - 1
assert factor_terms((0**(x + 2) - 1).subs(x, -2)) == 0
