Two postponed TODOs in recent PRs... (#1148)

rocky · web-flow · commit dcc607ec3571 · 2024-10-28T07:34:13.000-04:00
* Move eval functions to mathics.eval.numbers.numbers
* Move "no_doc" after imports
diff --git a/mathics/builtin/atomic/numbers.py b/mathics/builtin/atomic/numbers.py
@@ -17,10 +17,7 @@
 However, things like 'N[Pi, 100]' should work as expected.
 """
 
-from functools import lru_cache
-
 import mpmath
-import sympy
 
 from mathics.core.atoms import (
     Integer,
@@ -33,6 +30,7 @@
 from mathics.core.attributes import A_LISTABLE, A_PROTECTED, A_READ_PROTECTED
 from mathics.core.builtin import Builtin, Predefined
 from mathics.core.convert.python import from_python
+from mathics.core.evaluation import Evaluation
 from mathics.core.expression import Expression
 from mathics.core.list import ListExpression
 from mathics.core.number import (
@@ -52,100 +50,19 @@
     SymbolRound,
 )
 from mathics.eval.nevaluator import eval_N
-from mathics.eval.numbers.numbers import eval_Accuracy, eval_Precision
+from mathics.eval.numbers.numbers import (
+    check_finite_decimal,
+    convert_float_base,
+    convert_repeating_decimal,
+    eval_Accuracy,
+    eval_Precision,
+    log_n_b,
+)
 
 SymbolIntegerDigits = Symbol("IntegerDigits")
 SymbolIntegerExponent = Symbol("IntegerExponent")
 
 
-@lru_cache()
-def log_n_b(py_n, py_b) -> int:
-    return (
-        int(mpmath.floor(mpmath.log(py_n, py_b))) + 1 if py_n != 0 and py_n != 1 else 1
-    )
-
-
-def check_finite_decimal(denominator):
-    # The rational number is finite decimal if the denominator has form 2^a * 5^b
-    while denominator % 5 == 0:
-        denominator = denominator / 5
-
-    while denominator % 2 == 0:
-        denominator = denominator / 2
-
-    return True if denominator == 1 else False
-
-
-def convert_repeating_decimal(numerator, denominator, base):
-    head = [x for x in str(numerator // denominator)]
-    tails = []
-    subresults = [numerator % denominator]
-    numerator %= denominator
-
-    while numerator != 0:  # only rational input can go to this case
-        numerator *= base
-        result_digit, numerator = divmod(numerator, denominator)
-        tails.append(str(result_digit))
-        if numerator not in subresults:
-            subresults.append(numerator)
-        else:
-            break
-
-    for i in range(len(head) - 1, -1, -1):
-        j = len(tails) - 1
-        if head[i] != tails[j]:
-            break
-        else:
-            del tails[j]
-            tails.insert(0, head[i])
-            del head[i]
-
-    # truncate all leading 0's
-    if all(elem == "0" for elem in head):
-        for i in range(0, len(tails)):
-            if tails[0] == "0":
-                tails = tails[1:] + [str(0)]
-            else:
-                break
-    return (head, tails)
-
-
-def convert_float_base(x, base, precision=10):
-    length_of_int = 0 if x == 0 else int(mpmath.log(x, base))
-    # iexps = list(range(length_of_int, -1, -1))
-
-    def convert_int(x, base, exponents):
-        out = []
-        for e in range(0, exponents + 1):
-            d = x % base
-            out.append(d)
-            x = x / base
-            if x == 0:
-                break
-        out.reverse()
-        return out
-
-    def convert_float(x, base, exponents):
-        out = []
-        for e in range(0, exponents):
-            d = int(x * base)
-            out.append(d)
-            x = (x * base) - d
-            if x == 0:
-                break
-        return out
-
-    int_part = convert_int(int(x), base, length_of_int)
-    if isinstance(x, (float, sympy.Float)):
-        # fexps = list(range(-1, -int(precision + 1), -1))
-        real_part = convert_float(x - int(x), base, precision + 1)
-        return int_part + real_part
-    elif isinstance(x, int):
-        return int_part
-    else:
-        raise TypeError(x)
-
-
 class Accuracy(Builtin):
     """
     <url>
@@ -255,7 +172,7 @@ class IntegerExponent(Builtin):
 
     summary_text = "number of trailing 0s in a given base"
 
-    def eval_two_arg_integers(self, n: Integer, b: Integer, evaluation):
+    def eval_two_arg_integers(self, n: Integer, b: Integer, evaluation: Evaluation):
         """IntegerExponent[n_Integer, b_Integer]"""
 
         py_n, py_b = n.value, b.value
diff --git a/mathics/builtin/box/expression.py b/mathics/builtin/box/expression.py
@@ -1,6 +1,3 @@
-# This is never intended to go in Mathics3 docs
-no_doc = True
-
 from typing import Optional, Sequence, Union
 
 from mathics.core.attributes import A_PROTECTED, A_READ_PROTECTED
@@ -10,6 +7,9 @@
 from mathics.core.list import ListExpression
 from mathics.core.symbols import Symbol, SymbolHoldForm, ensure_context
 
+# This is never intended to go in Mathics3 docs
+no_doc = True
+
 
 def split_name(name: str) -> str:
     """
diff --git a/mathics/eval/numbers/numbers.py b/mathics/eval/numbers/numbers.py
@@ -3,6 +3,7 @@
 Implementation of numbers handling functions.
 """
 
+from functools import lru_cache
 from typing import Optional
 
 import mpmath
@@ -20,6 +21,87 @@
 from mathics.core.symbols import SymbolPlus
 
 
+def check_finite_decimal(denominator):
+    # The rational number is finite decimal if the denominator has form 2^a * 5^b
+    while denominator % 5 == 0:
+        denominator = denominator / 5
+
+    while denominator % 2 == 0:
+        denominator = denominator / 2
+
+    return True if denominator == 1 else False
+
+
+def convert_repeating_decimal(numerator, denominator, base):
+    head = [x for x in str(numerator // denominator)]
+    tails = []
+    subresults = [numerator % denominator]
+    numerator %= denominator
+
+    while numerator != 0:  # only rational input can go to this case
+        numerator *= base
+        result_digit, numerator = divmod(numerator, denominator)
+        tails.append(str(result_digit))
+        if numerator not in subresults:
+            subresults.append(numerator)
+        else:
+            break
+
+    for i in range(len(head) - 1, -1, -1):
+        j = len(tails) - 1
+        if head[i] != tails[j]:
+            break
+        else:
+            del tails[j]
+            tails.insert(0, head[i])
+            del head[i]
+
+    # truncate all leading 0's
+    if all(elem == "0" for elem in head):
+        for i in range(0, len(tails)):
+            if tails[0] == "0":
+                tails = tails[1:] + [str(0)]
+            else:
+                break
+    return (head, tails)
+
+
+def convert_float_base(x, base, precision=10):
+    length_of_int = 0 if x == 0 else int(mpmath.log(x, base))
+    # iexps = list(range(length_of_int, -1, -1))
+
+    def convert_int(x, base, exponents):
+        out = []
+        for e in range(0, exponents + 1):
+            d = x % base
+            out.append(d)
+            x = x / base
+            if x == 0:
+                break
+        out.reverse()
+        return out
+
+    def convert_float(x, base, exponents):
+        out = []
+        for e in range(0, exponents):
+            d = int(x * base)
+            out.append(d)
+            x = (x * base) - d
+            if x == 0:
+                break
+        return out
+
+    int_part = convert_int(int(x), base, length_of_int)
+    if isinstance(x, (float, sympy.Float)):
+        # fexps = list(range(-1, -int(precision + 1), -1))
+        real_part = convert_float(x - int(x), base, precision + 1)
+        return int_part + real_part
+    elif isinstance(x, int):
+        return int_part
+    else:
+        raise TypeError(x)
+
+
 def eval_Accuracy(z: BaseElement) -> Optional[float]:
     """
     Determine the accuracy of an expression expr.
@@ -129,7 +211,7 @@ def cancel(expr):
                 return None
 
             # result = sympy.powsimp(result, deep=True)
-            result = sympy.cancel(result)
+            result = tracing.run_sympy(sympy.cancel, result)
 
             # cancel factors out rationals, so we factor them again
             result = sympy_factor(result)
@@ -140,25 +222,11 @@ def cancel(expr):
             return expr
 
 
-def cancel(expr):
-    if expr.has_form("Plus", None):
-        return Expression(SymbolPlus, *[cancel(element) for element in expr.elements])
-    else:
-        try:
-            result = expr.to_sympy()
-            if result is None:
-                return None
-
-            # result = sympy.powsimp(result, deep=True)
-            result = tracing.run_sympy(sympy.cancel, result)
-
-            # cancel factors out rationals, so we factor them again
-            result = sympy_factor(result)
-
-            return from_sympy(result)
-        except sympy.PolynomialError:
-            # e.g. for non-commutative expressions
-            return expr
+@lru_cache()
+def log_n_b(py_n, py_b) -> int:
+    return (
+        int(mpmath.floor(mpmath.log(py_n, py_b))) + 1 if py_n != 0 and py_n != 1 else 1
+    )
 
 
 def sympy_factor(expr_sympy):
