Miscellaneous functions

AUTHORS:

  • William Stein
  • William Stein (2006-04-26): added workaround for Windows where most users’ home directory has a space in it.
  • Robert Bradshaw (2007-09-20): Ellipsis range/iterator.

TESTS:

The following test, verifying that trac ticket #16181 has been resolved, needs to stay at the beginning of this file so that its context is not poisoned by other tests:

sage: sage.misc.misc.inject_variable('a', 0)
sage: a
0

Check the fix from trac ticket #8323:

sage: 'name' in globals()
False
sage: 'func' in globals()
False
class sage.misc.misc.AttrCallObject(name, args, kwds)

Bases: object

TESTS:

sage: f = attrcall('core', 3); f
*.core(3)
sage: TestSuite(f).run()
class sage.misc.misc.GlobalCputime(t)

Container for CPU times of subprocesses.

AUTHOR:

  • Martin Albrecht - (2008-12): initial version

EXAMPLE:

Objects of this type are returned if subprocesses=True is passed to cputime():

sage: cputime(subprocesses=True) # indirect doctest, output random
0.2347431

We can use it to keep track of the CPU time spent in Singular for example:

sage: t = cputime(subprocesses=True)
sage: P = PolynomialRing(QQ,7,'x')
sage: I = sage.rings.ideal.Katsura(P)
sage: gb = I.groebner_basis() # calls Singular
sage: cputime(subprocesses=True) - t # output random
0.462987

For further processing we can then convert this container to a float:

sage: t = cputime(subprocesses=True)
sage: float(t) #output somewhat random
2.1088339999999999

See also

cputime()

sage.misc.misc.alarm(seconds)

Raise an AlarmInterrupt exception in a given number of seconds. This is useful for automatically interrupting long computations and can be trapped using exception handling.

Use cancel_alarm() to cancel a previously scheduled alarm.

INPUT:

  • seconds – positive number, may be floating point

EXAMPLES:

sage: alarm(0.5); factor(2^1031-1)
Traceback (most recent call last):
...
AlarmInterrupt
sage: alarm(0)
Traceback (most recent call last):
...
ValueError: alarm() time must be positive
sage.misc.misc.assert_attribute(x, attr, init=None)

If the object x has the attribute attr, do nothing. If not, set x.attr to init.

sage.misc.misc.attrcall(name, *args, **kwds)

Returns a callable which takes in an object, gets the method named name from that object, and calls it with the specified arguments and keywords.

INPUT:

  • name - a string of the name of the method you want to call
  • args, kwds - arguments and keywords to be passed to the method

EXAMPLES:

sage: f = attrcall('core', 3); f
*.core(3)
sage: [f(p) for p in Partitions(5)]
[[2], [1, 1], [1, 1], [3, 1, 1], [2], [2], [1, 1]]
class sage.misc.misc.cached_attribute(method, name=None)

Bases: object

Computes attribute value and caches it in the instance.

class sage.misc.misc.cached_class_attribute(method, name=None)

Bases: sage.misc.misc.cached_attribute

Computes attribute value and caches it in the class.

sage.misc.misc.call_method(obj, name, *args, **kwds)

Call the method name on obj.

This has to exist somewhere in Python!!!

See also

operator.methodcaller() attrcal()

EXAMPLES:

sage: from sage.misc.misc import call_method
sage: call_method(1, "__add__", 2)
3
sage.misc.misc.cancel_alarm()

Cancel a previously scheduled alarm (if any) set by alarm().

EXAMPLES:

sage: alarm(0.5)
sage: cancel_alarm()
sage: cancel_alarm()  # Calling more than once doesn't matter
sage: sleep(0.6)      # sleep succeeds
sage.misc.misc.cmp_props(left, right, props)

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.coeff_repr(c, is_latex=False)

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.compose(f, g)

Return the composition of one-variable functions: \(f \circ g\)

See also self_compose() and nest()

INPUT:
  • \(f\) – a function of one variable
  • \(g\) – another function of one variable
OUTPUT:
A function, such that compose(f,g)(x) = f(g(x))

EXAMPLES:

sage: def g(x): return 3*x
sage: def f(x): return x + 1
sage: h1 = compose(f,g)
sage: h2 = compose(g,f)
sage: _ = var ('x')
sage: h1(x)
3*x + 1
sage: h2(x)
3*x + 3
::
sage: _ = function(‘f g’) sage: _ = var (‘x’) sage: compose(f,g)(x) f(g(x))
sage.misc.misc.cputime(t=0, subprocesses=False)

Return the time in CPU seconds since Sage started, or with optional argument t, return the time since t. This is how much time Sage has spent using the CPU. If subprocesses=False this does not count time spent in subprocesses spawned by Sage (e.g., Gap, Singular, etc.). If subprocesses=True this function tries to take all subprocesses with a working cputime() implementation into account.

The measurement for the main Sage process is done via a call to resource.getrusage(), so it avoids the wraparound problems in time.clock() on Cygwin.

INPUT:

  • t - (optional) time in CPU seconds, if t is a result from an earlier call with subprocesses=True, then subprocesses=True is assumed.
  • subprocesses – (optional), include subprocesses (default: False)

OUTPUT:

  • float - time in CPU seconds if subprocesses=False
  • GlobalCputime - object which holds CPU times of subprocesses otherwise

EXAMPLES:

sage: t = cputime()
sage: F = gp.factor(2^199-1)
sage: cputime(t)          # somewhat random
0.010999000000000092

sage: t = cputime(subprocesses=True)
sage: F = gp.factor(2^199-1)
sage: cputime(t) # somewhat random
0.091999

sage: w = walltime()
sage: F = gp.factor(2^199-1)
sage: walltime(w)         # somewhat random
0.58425593376159668

Note

Even with subprocesses=True there is no guarantee that the CPU time is reported correctly because subprocesses can be started and terminated at any given time.

sage.misc.misc.ellipsis_iter(*args, **kwds)

Same as ellipsis_range, but as an iterator (and may end with an Ellipsis).

See also ellipsis_range.

Use (1,2,...) notation.

EXAMPLES:

sage: A = ellipsis_iter(1,2,Ellipsis)
sage: [A.next() for _ in range(10)]
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
sage: A.next()
11
sage: A = ellipsis_iter(1,3,5,Ellipsis)
sage: [A.next() for _ in range(10)]
[1, 3, 5, 7, 9, 11, 13, 15, 17, 19]
sage: A = ellipsis_iter(1,2,Ellipsis,5,10,Ellipsis)
sage: [A.next() for _ in range(10)]
[1, 2, 3, 4, 5, 10, 11, 12, 13, 14]

TESTS:

These were carefully chosen tests, only to be changed if the semantics of ellipsis ranges change. In other words, if they don’t pass it’s probably a bug in the implementation, not in the doctest.

sage: list(1,..,10)
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
sage: list(1,3,..,10)
[1, 3, 5, 7, 9]
sage: list(1,..,10,..,20)
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
sage: list(1,3,..,10,..,20)
[1, 3, 5, 7, 9, 10, 12, 14, 16, 18, 20]
sage: list(1,3,..,10,10,..,20)
[1, 3, 5, 7, 9, 10, 12, 14, 16, 18, 20]
sage: list(0,2,..,10,10,..,20,20,..,25)
[0, 2, 4, 6, 8, 10, 10, 12, 14, 16, 18, 20, 20, 22, 24]
sage: list(10,..,1)
[]
sage: list(10,11,..,1)
[]
sage: list(10,9,..,1)
[10, 9, 8, 7, 6, 5, 4, 3, 2, 1]
sage: list(100,..,10,..,20)
[10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
sage: list(0,..,10,..,-20)
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
sage: list(100,..,10,..,-20)
[]
sage: list(100,102,..,10,..,20)
[10, 12, 14, 16, 18, 20]
sage.misc.misc.ellipsis_range(*args, **kwds)

Return arithmetic sequence determined by the numeric arguments and ellipsis. Best illustrated by examples.

Use [1,2,..,n] notation.

EXAMPLES:

sage: ellipsis_range(1,Ellipsis,11,100)
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 100]
sage: ellipsis_range(0,2,Ellipsis,10,Ellipsis,20)
[0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20]
sage: ellipsis_range(0,2,Ellipsis,11,Ellipsis,20)
[0, 2, 4, 6, 8, 10, 11, 13, 15, 17, 19]
sage: ellipsis_range(0,2,Ellipsis,11,Ellipsis,20, step=3)
[0, 2, 5, 8, 11, 14, 17, 20]
sage: ellipsis_range(10,Ellipsis,0)
[]

TESTS: These were carefully chosen tests, only to be changed if the semantics of ellipsis ranges change. In other words, if they don’t pass it’s probably a bug in the implementation, not in the doctest.

Note 10 only appears once (though it is in both ranges).

sage: ellipsis_range(0,Ellipsis,10,Ellipsis,20,step=2)
[0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20]

Sometimes one or more ranges is empty.

sage: ellipsis_range(100,Ellipsis,10,Ellipsis,20,step=2)
[10, 12, 14, 16, 18, 20]
sage: ellipsis_range(0,Ellipsis,10,Ellipsis,-20,step=2)
[0, 2, 4, 6, 8, 10]
sage: ellipsis_range(100,Ellipsis,10,Ellipsis,-20,step=2)
[]

We always start on the leftmost point of the range.

sage: ellipsis_range(0,Ellipsis,10,Ellipsis,20,step=3)
[0, 3, 6, 9, 10, 13, 16, 19]
sage: ellipsis_range(100,Ellipsis,10,Ellipsis,20,step=3)
[10, 13, 16, 19]
sage: ellipsis_range(0,Ellipsis,10,Ellipsis,-20,step=3)
[0, 3, 6, 9]
sage: ellipsis_range(100,Ellipsis,10,Ellipsis,-20,step=3)
[]
sage: ellipsis_range(0,1,Ellipsis,-10)
[]
sage: ellipsis_range(0,1,Ellipsis,-10,step=1)
[0]
sage: ellipsis_range(100,0,1,Ellipsis,-10)
[100]

Note the duplicate 5 in the output.

sage: ellipsis_range(0,Ellipsis,5,5,Ellipsis,10)
[0, 1, 2, 3, 4, 5, 5, 6, 7, 8, 9, 10]

Examples in which the step determines the parent of the elements:

sage: [1..3, step=0.5]
[1.00000000000000, 1.50000000000000, 2.00000000000000, 2.50000000000000, 3.00000000000000]
sage: v = [1..5, step=1/1]; v
[1, 2, 3, 4, 5]
sage: parent(v[2])
Rational Field
sage.misc.misc.embedded()

Return True if this copy of Sage is running embedded in the Sage notebook.

EXAMPLES:

sage: sage.misc.misc.embedded()    # output True if in the notebook
False
sage.misc.misc.exists(S, P)

If S contains an element x such that P(x) is True, this function returns True and the element x. Otherwise it returns False and None.

Note that this function is NOT suitable to be used in an if-statement or in any place where a boolean expression is expected. For those situations, use the Python built-in

any(P(x) for x in S)

INPUT:

  • S - object (that supports enumeration)
  • P - function that returns True or False

OUTPUT:

  • bool - whether or not P is True for some element x of S
  • object - x

EXAMPLES: lambda functions are very useful when using the exists function:

sage: exists([1,2,5], lambda x : x > 7)
(False, None)
sage: exists([1,2,5], lambda x : x > 3)
(True, 5)

The following example is similar to one in the MAGMA handbook. We check whether certain integers are a sum of two (small) cubes:

sage: cubes = [t**3 for t in range(-10,11)]
sage: exists([(x,y) for x in cubes for y in cubes], lambda v : v[0]+v[1] == 218)
(True, (-125, 343))
sage: exists([(x,y) for x in cubes for y in cubes], lambda v : v[0]+v[1] == 219)
(False, None)
sage.misc.misc.forall(S, P)

If P(x) is true every x in S, return True and None. If there is some element x in S such that P is not True, return False and x.

Note that this function is NOT suitable to be used in an if-statement or in any place where a boolean expression is expected. For those situations, use the Python built-in

all(P(x) for x in S)

INPUT:

  • S - object (that supports enumeration)
  • P - function that returns True or False

OUTPUT:

  • bool - whether or not P is True for all elements of S
  • object - x

EXAMPLES: lambda functions are very useful when using the forall function. As a toy example we test whether certain integers are greater than 3.

sage: forall([1,2,5], lambda x : x > 3)
(False, 1)
sage: forall([1,2,5], lambda x : x > 0)
(True, None)

Next we ask whether every positive integer less than 100 is a product of at most 2 prime factors:

sage: forall(range(1,100),  lambda n : len(factor(n)) <= 2)
(False, 30)

The answer is no, and 30 is a counterexample. However, every positive integer 100 is a product of at most 3 primes.

sage: forall(range(1,100),  lambda n : len(factor(n)) <= 3)
(True, None)
sage.misc.misc.generic_cmp(x, y)

Compare x and y and return -1, 0, or 1.

This is similar to x.__cmp__(y), but works even in some cases when a .__cmp__ method isn’t defined.

sage.misc.misc.get_main_globals()

Return the main global namespace.

EXAMPLES:

sage: from sage.misc.misc import get_main_globals
sage: G = get_main_globals()
sage: bla = 1
sage: G['bla']
1
sage: bla = 2
sage: G['bla']
2
sage: G['ble'] = 5
sage: ble
5

This is analogous to globals(), except that it can be called from any function, even if it is in a Python module:

sage: def f():
....:     G = get_main_globals()
....:     assert G['bli'] == 14
....:     G['blo'] = 42
sage: bli = 14
sage: f()
sage: blo
42

ALGORITHM:

The main global namespace is discovered by going up the frame stack until the frame for the __main__ module is found. Should this frame not be found (this should not occur in normal operation), an exception “ValueError: call stack is not deep enough” will be raised by _getframe.

See inject_variable_test() for a real test that this works within deeply nested calls in a function defined in a Python module.

sage.misc.misc.get_verbose()

Return the global Sage verbosity level.

INPUT: int level: an integer between 0 and 2, inclusive.

OUTPUT: changes the state of the verbosity flag.

EXAMPLES:

sage: get_verbose()
0
sage: set_verbose(2)
sage: get_verbose()
2
sage: set_verbose(0)
sage.misc.misc.get_verbose_files()
sage.misc.misc.getitem(v, n)

Variant of getitem that coerces to an int if a TypeError is raised.

(This is not needed anymore - classes should define an __index__ method.)

Thus, e.g., getitem(v,n) will work even if \(v\) is a Python list and \(n\) is a Sage integer.

EXAMPLES:

sage: v = [1,2,3]

The following used to fail in Sage <= 1.3.7. Now it works fine:

sage: v[ZZ(1)]
2

This always worked.

sage: getitem(v, ZZ(1))
2
sage.misc.misc.inject_variable(name, value)

Inject a variable into the main global namespace.

INPUT:

  • name – a string
  • value – anything

EXAMPLES:

sage: from sage.misc.misc import inject_variable
sage: inject_variable("a", 314)
sage: a
314

A warning is issued the first time an existing value is overwritten:

sage: inject_variable("a", 271)
doctest:...: RuntimeWarning: redefining global value `a`
sage: a
271
sage: inject_variable("a", 272)
sage: a
272

That’s because warn seem to not reissue twice the same warning:

sage: from warnings import warn sage: warn(“blah”) doctest:...: UserWarning: blah sage: warn(“blah”)

Use with care!

sage.misc.misc.inject_variable_test(name, value, depth)

A function for testing deep calls to inject_variable

TESTS:

sage: from sage.misc.misc import inject_variable_test
sage: inject_variable_test("a0", 314, 0)
sage: a0
314
sage: inject_variable_test("a1", 314, 1)
sage: a1
314
sage: inject_variable_test("a2", 314, 2)
sage: a2
314
sage: inject_variable_test("a2", 271, 2)
doctest:...: RuntimeWarning: redefining global value `a2`
sage: a2
271
sage.misc.misc.is_in_string(line, pos)

Returns True if the character at position pos in line occurs within a string.

EXAMPLES:

sage: from sage.misc.misc import is_in_string
sage: line = 'test(\'#\')'
sage: is_in_string(line, line.rfind('#'))
True
sage: is_in_string(line, line.rfind(')'))
False
sage.misc.misc.is_iterator(it)

Tests if it is an iterator.

The mantra if hasattr(it, 'next') was used to tests if it is an iterator. This is not quite correct since it could have a next methods with a different semantic.

EXAMPLES:

sage: it = iter([1,2,3])
sage: is_iterator(it)
True

sage: class wrong():
...      def __init__(self): self.n = 5
...      def next(self):
...          self.n -= 1
...          if self.n == 0: raise StopIteration
...          return self.n
sage: x = wrong()
sage: is_iterator(x)
False
sage: list(x)
Traceback (most recent call last):
...
TypeError: iteration over non-sequence

sage: class good(wrong):
...      def __iter__(self): return self
sage: x = good()
sage: is_iterator(x)
True
sage: list(x)
[4, 3, 2, 1]

sage: P = Partitions(3)
sage: is_iterator(P)
False
sage: is_iterator(iter(P))
True
class sage.misc.misc.lazy_prop(calculate_function)

Bases: object

sage.misc.misc.nest(f, n, x)

Return \(f(f(...f(x)...))\), where the composition occurs n times.

See also compose() and self_compose()

INPUT:
  • \(f\) – a function of one variable
  • \(n\) – a nonnegative integer
  • \(x\) – any input for \(f\)
OUTPUT:
\(f(f(...f(x)...))\), where the composition occurs n times

EXAMPLES:

sage: def f(x): return x^2 + 1
sage: x = var('x')
sage: nest(f, 3, x)
((x^2 + 1)^2 + 1)^2 + 1
sage: _ = function('f')
sage: _ = var('x')
sage: nest(f, 10, x)
f(f(f(f(f(f(f(f(f(f(x))))))))))
sage: _ = function('f')
sage: _ = var('x')
sage: nest(f, 0, x)
x
sage.misc.misc.newton_method_sizes(N)

Returns a sequence of integers \(1 = a_1 \leq a_2 \leq \cdots \leq a_n = N\) such that \(a_j = \lceil a_{j+1} / 2 \rceil\) for all \(j\).

This is useful for Newton-style algorithms that double the precision at each stage. For example if you start at precision 1 and want an answer to precision 17, then it’s better to use the intermediate stages 1, 2, 3, 5, 9, 17 than to use 1, 2, 4, 8, 16, 17.

INPUT:

  • N - positive integer

EXAMPLES:

sage: newton_method_sizes(17)
[1, 2, 3, 5, 9, 17]
sage: newton_method_sizes(16)
[1, 2, 4, 8, 16]
sage: newton_method_sizes(1)
[1]

AUTHORS:

  • David Harvey (2006-09-09)
sage.misc.misc.pad_zeros(s, size=3)

EXAMPLES:

sage: pad_zeros(100)
'100'
sage: pad_zeros(10)
'010'
sage: pad_zeros(10, 5)
'00010'
sage: pad_zeros(389, 5)
'00389'
sage: pad_zeros(389, 10)
'0000000389'
sage.misc.misc.powerset(X)

Iterator over the list of all subsets of the iterable X, in no particular order. Each list appears exactly once, up to order.

INPUT:

  • X - an iterable

OUTPUT: iterator of lists

EXAMPLES:

sage: list(powerset([1,2,3]))
[[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]
sage: [z for z in powerset([0,[1,2]])]
[[], [0], [[1, 2]], [0, [1, 2]]]

Iterating over the power set of an infinite set is also allowed:

sage: i = 0
sage: for x in powerset(ZZ):
...    if i > 10:
...       break
...    else:
...       i += 1
...    print x,
[] [0] [1] [0, 1] [-1] [0, -1] [1, -1] [0, 1, -1] [2] [0, 2] [1, 2]

You may also use subsets as an alias for powerset:

sage: subsets([1,2,3])
<generator object powerset at 0x...>
sage: list(subsets([1,2,3]))
[[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]

The reason we return lists instead of sets is that the elements of
sets must be hashable and many structures on which one wants the
powerset consist of non-hashable objects.

AUTHORS:

  • William Stein
  • Nils Bruin (2006-12-19): rewrite to work for not-necessarily finite objects X.
sage.misc.misc.prop(f)

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.random_sublist(X, s)

Return a pseudo-random sublist of the list X where the probability of including a particular element is s.

INPUT:

  • X - list
  • s - floating point number between 0 and 1

OUTPUT: list

EXAMPLES:

sage: S = [1,7,3,4,18]
sage: random_sublist(S, 0.5)
[1, 3, 4]
sage: random_sublist(S, 0.5)
[1, 3]
sage.misc.misc.repr_lincomb(terms, coeffs=None, is_latex=False, scalar_mult='*', strip_one=False, repr_monomial=None, latex_scalar_mult=None)

Compute a string representation of a linear combination of some formal symbols.

INPUT:

  • terms – list of terms, as pairs (support, coefficient)
  • is_latex – whether to produce latex (default: False)
  • scalar_mult – string representing the multiplication (default:'*')
  • latex_scalar_mult – latex string representing the multiplication (default: '' if scalar_mult is '*'; otherwise scalar_mult)
  • coeffs – for backward compatibility

OUTPUT:

  • str - a string

EXAMPLES:

sage: repr_lincomb([('a',1), ('b',-2), ('c',3)])
'a - 2*b + 3*c'
sage: repr_lincomb([('a',0), ('b',-2), ('c',3)])
'-2*b + 3*c'
sage: repr_lincomb([('a',0), ('b',2), ('c',3)])
'2*b + 3*c'
sage: repr_lincomb([('a',1), ('b',0), ('c',3)])
'a + 3*c'
sage: repr_lincomb([('a',-1), ('b','2+3*x'), ('c',3)])
'-a + (2+3*x)*b + 3*c'
sage: repr_lincomb([('a', '1+x^2'), ('b', '2+3*x'), ('c', 3)])
'(1+x^2)*a + (2+3*x)*b + 3*c'
sage: repr_lincomb([('a', '1+x^2'), ('b', '-2+3*x'), ('c', 3)])
'(1+x^2)*a + (-2+3*x)*b + 3*c'
sage: repr_lincomb([('a', 1), ('b', -2), ('c', -3)])
'a - 2*b - 3*c'
sage: t = PolynomialRing(RationalField(),'t').gen()
sage: repr_lincomb([('a', -t), ('s', t - 2), ('', t^2 + 2)])
'-t*a + (t-2)*s + (t^2+2)'

Examples for scalar_mult:

sage: repr_lincomb([('a',1), ('b',2), ('c',3)], scalar_mult='*')
'a + 2*b + 3*c'
sage: repr_lincomb([('a',2), ('b',0), ('c',-3)], scalar_mult='**')
'2**a - 3**c'
sage: repr_lincomb([('a',-1), ('b',2), ('c',3)], scalar_mult='**')
'-a + 2**b + 3**c'

Examples for scalar_mult and is_latex:

sage: repr_lincomb([('a',-1), ('b',2), ('c',3)], is_latex=True)
'-a + 2b + 3c'
sage: repr_lincomb([('a',-1), ('b',-1), ('c',3)], is_latex=True, scalar_mult='*')
'-a - b + 3c'
sage: repr_lincomb([('a',-1), ('b',2), ('c',-3)], is_latex=True, scalar_mult='**')
'-a + 2**b - 3**c'
sage: repr_lincomb([('a',-2), ('b',-1), ('c',-3)], is_latex=True, latex_scalar_mult='*')
'-2*a - b - 3*c'

Examples for strip_one:

sage: repr_lincomb([ ('a',1), (1,-2), ('3',3) ])
'a - 2*1 + 3*3'
sage: repr_lincomb([ ('a',-1), (1,1), ('3',3) ])
'-a + 1 + 3*3'
sage: repr_lincomb([ ('a',1), (1,-2), ('3',3) ], strip_one = True)
'a - 2 + 3*3'
sage: repr_lincomb([ ('a',-1), (1,1), ('3',3) ], strip_one = True)
'-a + 1 + 3*3'
sage: repr_lincomb([ ('a',1), (1,-1), ('3',3) ], strip_one = True)
'a - 1 + 3*3'

Examples for repr_monomial:

sage: repr_lincomb([('a',1), ('b',2), ('c',3)], repr_monomial = lambda s: s+"1")
'a1 + 2*b1 + 3*c1'

TESTS:

For backward compatibility (will be deprecated):

sage: repr_lincomb(['a','b','c'], [1,2,3])
doctest:...: DeprecationWarning: calling `repr_lincomb(monoms, coeffs)` is deprecated; please specify a list of tuples (monom, coeff) instead
See http://trac.sagemath.org/12484 for details.
'a + 2*b + 3*c'
sage.misc.misc.sage_makedirs(dir)

Python version of mkdir -p: try to create a directory, and also create all intermediate directories as necessary. Succeed silently if the directory already exists (unlike os.makedirs()). Raise other errors (like permission errors) normally.

EXAMPLES:

sage: from sage.misc.misc import sage_makedirs
sage: sage_makedirs(DOT_SAGE) # no output

The following fails because we are trying to create a directory in place of an ordinary file (the main Sage executable):

sage: sage_executable = os.path.join(SAGE_ROOT, 'sage')
sage: sage_makedirs(sage_executable)
Traceback (most recent call last):
...
OSError: ...
sage.misc.misc.self_compose(f, n)

Return the function \(f\) composed with itself \(n\) times.

See nest() if you want \(f(f(...(f(x))...))\) for known \(x\).

INPUT:
  • \(f\) – a function of one variable
  • \(n\) – a nonnegative integer
OUTPUT:
A function, the result of composing \(f\) with itself \(n\) times

EXAMPLES:

sage: def f(x): return x^2 + 1
sage: g = self_compose(f, 3)
sage: x = var('x')
sage: g(x)
((x^2 + 1)^2 + 1)^2 + 1
sage: def f(x): return x + 1
sage: g = self_compose(f, 10000)
sage: g(0)
10000
sage: x = var('x')
sage: self_compose(sin, 0)(x)
x
sage.misc.misc.set_verbose(level, files='all')

Set the global Sage verbosity level.

INPUT:

  • level - an integer between 0 and 2, inclusive.

  • files (default: ‘all’): list of files to make verbose, or

    ‘all’ to make ALL files verbose (the default).

OUTPUT: changes the state of the verbosity flag and possibly appends to the list of files that are verbose.

EXAMPLES:

sage: set_verbose(2)
sage: verbose("This is Sage.", level=1)  # not tested
VERBOSE1 (?): This is Sage.
sage: verbose("This is Sage.", level=2)  # not tested
VERBOSE2 (?): This is Sage.
sage: verbose("This is Sage.", level=3)  # not tested
[no output]
sage: set_verbose(0)
sage.misc.misc.set_verbose_files(file_name)
sage.misc.misc.sourcefile(object)

Work out which source or compiled file an object was defined in.

sage.misc.misc.srange(start, end=None, step=1, universe=None, check=True, include_endpoint=False, endpoint_tolerance=1e-05)

Return list of numbers a, a+step, ..., a+k*step, where a+k*step < b and a+(k+1)*step >= b over exact rings, and makes a best attempt for inexact rings (see note below).

This provides one way to iterate over Sage integers as opposed to Python int’s. It also allows you to specify step sizes for such an iteration. Note, however, that what is returned is a full list of Integers and not an iterator. It is potentially much slower than the Python range function, depending on the application. The function xsrange() provides an iterator with similar functionality which would usually be more efficient than using srange().

INPUT:

  • a - number
  • b - number (default: None)
  • step - number (default: 1)
  • universe - Parent or type where all the elements should live (default: deduce from inputs)
  • check - make sure a, b, and step all lie in the same universe
  • include_endpoint - whether or not to include the endpoint (default: False)
  • endpoint_tolerance - used to determine whether or not the endpoint is hit for inexact rings (default 1e-5)

OUTPUT:

  • list

If b is None, then b is set equal to a and a is set equal to the 0 in the parent of b.

Unlike range, a and b can be any type of numbers, and the resulting list involves numbers of that type.

Note

The list elements are computed via repeated addition rather than multiplication, which may produce slightly different results with inexact rings. For example:

sage: sum([1.1] * 10) == 1.1 * 10
False

Also, the question of whether the endpoint is hit exactly for a given a + k*step is fuzzy for an inexact ring. If a + k*step = b for some k within endpoint_tolerance of being integral, it is considered an exact hit, thus avoiding spurious values falling just below the endpoint.

Note

This function is called srange to distinguish it from the built-in Python range command. The s at the beginning of the name stands for “Sage”.

EXAMPLES:

sage: v = srange(5); v
[0, 1, 2, 3, 4]
sage: type(v[2])
<type 'sage.rings.integer.Integer'>
sage: srange(1, 10)
[1, 2, 3, 4, 5, 6, 7, 8, 9]
sage: srange(10, 1, -1)
[10, 9, 8, 7, 6, 5, 4, 3, 2]
sage: srange(10,1,-1, include_endpoint=True)
[10, 9, 8, 7, 6, 5, 4, 3, 2, 1]
sage: srange(1, 10, universe=RDF)
[1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0]

sage: srange(1, 10, 1/2)
[1, 3/2, 2, 5/2, 3, 7/2, 4, 9/2, 5, 11/2, 6, 13/2, 7, 15/2, 8, 17/2, 9, 19/2]
sage: srange(1, 5, 0.5)
[1.00000000000000, 1.50000000000000, 2.00000000000000, 2.50000000000000, 3.00000000000000, 3.50000000000000, 4.00000000000000, 4.50000000000000]
sage: srange(0, 1, 0.4)
[0.000000000000000, 0.400000000000000, 0.800000000000000]
sage: srange(1.0, 5.0, include_endpoint=True)
[1.00000000000000, 2.00000000000000, 3.00000000000000, 4.00000000000000, 5.00000000000000]
sage: srange(1.0, 1.1)
[1.00000000000000]
sage: srange(1.0, 1.0)
[]
sage: V = VectorSpace(QQ, 2)
sage: srange(V([0,0]), V([5,5]), step=V([2,2]))
[(0, 0), (2, 2), (4, 4)]

Including the endpoint:

sage: srange(0, 10, step=2, include_endpoint=True)
[0, 2, 4, 6, 8, 10]
sage: srange(0, 10, step=3, include_endpoint=True)
[0, 3, 6, 9]

Try some inexact rings:

sage: srange(0.5, 1.1, 0.1, universe=RDF, include_endpoint=False)
[0.5, 0.6, 0.7, 0.7999999999999999, 0.8999999999999999, 0.9999999999999999]
sage: srange(0.5, 1, 0.1, universe=RDF, include_endpoint=False)
[0.5, 0.6, 0.7, 0.7999999999999999, 0.8999999999999999]
sage: srange(0.5, 0.9, 0.1, universe=RDF, include_endpoint=False)
[0.5, 0.6, 0.7, 0.7999999999999999]
sage: srange(0, 1.1, 0.1, universe=RDF, include_endpoint=True)
[0.0, 0.1, 0.2, 0.30000000000000004, 0.4, 0.5, 0.6, 0.7, 0.7999999999999999, 0.8999999999999999, 0.9999999999999999, 1.1]
sage: srange(0, 0.2, 0.1, universe=RDF, include_endpoint=True)
[0.0, 0.1, 0.2]
sage: srange(0, 0.3, 0.1, universe=RDF, include_endpoint=True)
[0.0, 0.1, 0.2, 0.3]

TESTS:

These are doctests from trac ticket #6409:

sage: srange(1,0,include_endpoint=True)
[]
sage: srange(1,QQ(0),include_endpoint=True)
[]
sage: srange(3,0,-1,include_endpoint=True)
[3, 2, 1, 0]
sage: srange(1,1,0) # trac ticket #11753
Traceback (most recent call last):
...
ValueError: srange() step argument must not be zero
sage.misc.misc.strunc(s, n=60)

Truncate at first space after position n, adding ‘...’ if nontrivial truncation.

sage.misc.misc.subsets(X)

Iterator over the list of all subsets of the iterable X, in no particular order. Each list appears exactly once, up to order.

INPUT:

  • X - an iterable

OUTPUT: iterator of lists

EXAMPLES:

sage: list(powerset([1,2,3]))
[[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]
sage: [z for z in powerset([0,[1,2]])]
[[], [0], [[1, 2]], [0, [1, 2]]]

Iterating over the power set of an infinite set is also allowed:

sage: i = 0
sage: for x in powerset(ZZ):
...    if i > 10:
...       break
...    else:
...       i += 1
...    print x,
[] [0] [1] [0, 1] [-1] [0, -1] [1, -1] [0, 1, -1] [2] [0, 2] [1, 2]

You may also use subsets as an alias for powerset:

sage: subsets([1,2,3])
<generator object powerset at 0x...>
sage: list(subsets([1,2,3]))
[[], [1], [2], [1, 2], [3], [1, 3], [2, 3], [1, 2, 3]]

The reason we return lists instead of sets is that the elements of
sets must be hashable and many structures on which one wants the
powerset consist of non-hashable objects.

AUTHORS:

  • William Stein
  • Nils Bruin (2006-12-19): rewrite to work for not-necessarily finite objects X.
sage.misc.misc.sxrange(start, end=None, step=1, universe=None, check=True, include_endpoint=False, endpoint_tolerance=1e-05)

Return an iterator over numbers a, a+step, ...,  a+k*step, where a+k*step < b and a+(k+1)*step > b.

INPUT:
universe – Parent or type where all the elements should live (default: deduce from inputs) check – make sure a, b, and step all lie in the same universe include_endpoint – whether or not to include the endpoint (default: False) endpoint_tolerance – used to determine whether or not the endpoint is hit for inexact rings (default 1e-5)
  • a - number
  • b - number
  • step - number (default: 1)

OUTPUT: iterator

Unlike range, a and b can be any type of numbers, and the resulting iterator involves numbers of that type.

See also

srange()

Note

This function is called xsrange to distinguish it from the builtin Python xrange command.

EXAMPLES:

sage: list(xsrange(1,10))
[1, 2, 3, 4, 5, 6, 7, 8, 9]
sage: Q = RationalField()
sage: list(xsrange(1, 10, Q('1/2')))
[1, 3/2, 2, 5/2, 3, 7/2, 4, 9/2, 5, 11/2, 6, 13/2, 7, 15/2, 8, 17/2, 9, 19/2]
sage: list(xsrange(1, 5, 0.5))
[1.00000000000000, 1.50000000000000, 2.00000000000000, 2.50000000000000, 3.00000000000000, 3.50000000000000, 4.00000000000000, 4.50000000000000]
sage: list(xsrange(0, 1, 0.4))
[0.000000000000000, 0.400000000000000, 0.800000000000000]

Negative ranges are also allowed:

sage: list(xrange(4,1,-1))
[4, 3, 2]
sage: list(sxrange(4,1,-1))
[4, 3, 2]
sage: list(sxrange(4,1,-1/2))
[4, 7/2, 3, 5/2, 2, 3/2]

TESTS:

These are doctests from trac ticket #6409:

sage: list(xsrange(1,QQ(0),include_endpoint=True))
[]
sage: list(xsrange(1,QQ(0),-1,include_endpoint=True))
[1, 0]
sage: xsrange(1,1,0) # trac ticket #11753
Traceback (most recent call last):
...
ValueError: xsrange() step argument must not be zero
sage.misc.misc.to_gmp_hex(n)

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.todo(mesg='')

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.typecheck(x, C, var='x')

Check that x is of instance C. If not raise a TypeError with an error message.

sage.misc.misc.union(x, y=None)

Return the union of x and y, as a list. The resulting list need not be sorted and can change from call to call.

INPUT:

  • x - iterable
  • y - iterable (may optionally omitted)

OUTPUT: list

EXAMPLES:

sage: answer = union([1,2,3,4], [5,6]); answer
[1, 2, 3, 4, 5, 6]
sage: union([1,2,3,4,5,6], [5,6]) == answer
True
sage: union((1,2,3,4,5,6), [5,6]) == answer
True
sage: union((1,2,3,4,5,6), set([5,6])) == answer
True
sage.misc.misc.uniq(x)

Return the sublist of all elements in the list x that is sorted and is such that the entries in the sublist are unique.

EXAMPLES:

sage: v = uniq([1,1,8,-5,3,-5,'a','x','a'])
sage: v            # potentially random ordering of output
['a', 'x', -5, 1, 3, 8]
sage: set(v) == set(['a', 'x', -5, 1, 3, 8])
True
sage.misc.misc.unset_verbose_files(file_name)
sage.misc.misc.verbose(mesg='', t=0, level=1, caller_name=None)

Print a message if the current verbosity is at least level.

INPUT:

  • mesg - str, a message to print
  • t - int, optional, if included, will also print cputime(t), - which is the time since time t. Thus t should have been obtained with t=cputime()
  • level - int, (default: 1) the verbosity level of what we are printing
  • caller_name - string (default: None), the name of the calling function; in most cases Python can deduce this, so it need not be provided.

OUTPUT: possibly prints a message to stdout; also returns cputime()

EXAMPLE:

sage: set_verbose(1)
sage: t = cputime()
sage: t = verbose("This is Sage.", t, level=1, caller_name="william")       # not tested
VERBOSE1 (william): This is Sage. (time = 0.0)
sage: set_verbose(0)
sage.misc.misc.walltime(t=0)

Return the wall time in second, or with optional argument t, return the wall time since time t. “Wall time” means the time on a wall clock, i.e., the actual time.

INPUT:

  • t - (optional) float, time in CPU seconds

OUTPUT:

  • float - time in seconds

EXAMPLES:

sage: w = walltime()
sage: F = factor(2^199-1)
sage: walltime(w)   # somewhat random
0.8823847770690918
sage.misc.misc.word_wrap(s, ncols=85)

x.__init__(...) initializes x; see help(type(x)) for signature

sage.misc.misc.xsrange(start, end=None, step=1, universe=None, check=True, include_endpoint=False, endpoint_tolerance=1e-05)

Return an iterator over numbers a, a+step, ...,  a+k*step, where a+k*step < b and a+(k+1)*step > b.

INPUT:
universe – Parent or type where all the elements should live (default: deduce from inputs) check – make sure a, b, and step all lie in the same universe include_endpoint – whether or not to include the endpoint (default: False) endpoint_tolerance – used to determine whether or not the endpoint is hit for inexact rings (default 1e-5)
  • a - number
  • b - number
  • step - number (default: 1)

OUTPUT: iterator

Unlike range, a and b can be any type of numbers, and the resulting iterator involves numbers of that type.

See also

srange()

Note

This function is called xsrange to distinguish it from the builtin Python xrange command.

EXAMPLES:

sage: list(xsrange(1,10))
[1, 2, 3, 4, 5, 6, 7, 8, 9]
sage: Q = RationalField()
sage: list(xsrange(1, 10, Q('1/2')))
[1, 3/2, 2, 5/2, 3, 7/2, 4, 9/2, 5, 11/2, 6, 13/2, 7, 15/2, 8, 17/2, 9, 19/2]
sage: list(xsrange(1, 5, 0.5))
[1.00000000000000, 1.50000000000000, 2.00000000000000, 2.50000000000000, 3.00000000000000, 3.50000000000000, 4.00000000000000, 4.50000000000000]
sage: list(xsrange(0, 1, 0.4))
[0.000000000000000, 0.400000000000000, 0.800000000000000]

Negative ranges are also allowed:

sage: list(xrange(4,1,-1))
[4, 3, 2]
sage: list(sxrange(4,1,-1))
[4, 3, 2]
sage: list(sxrange(4,1,-1/2))
[4, 7/2, 3, 5/2, 2, 3/2]

TESTS:

These are doctests from trac ticket #6409:

sage: list(xsrange(1,QQ(0),include_endpoint=True))
[]
sage: list(xsrange(1,QQ(0),-1,include_endpoint=True))
[1, 0]
sage: xsrange(1,1,0) # trac ticket #11753
Traceback (most recent call last):
...
ValueError: xsrange() step argument must not be zero

Previous topic

Exceptions

Next topic

Temporary file handling

This Page