Merge pull request #2538 from Kodiologist/digit-separator-docs

Document and test weirder uses of digit separators

Author: Kodiologist

docs/syntax.rst

@@ -146,7 +146,11 @@ few extensions:
 
 - Commas (``,``) can be used like underscores (``_``) to separate digits
   without changing the result. Thus, ``10_000_000_000`` may also be written
-  ``10,000,000,000``.
+  ``10,000,000,000``. Hy is also more permissive about the placement of
+  separators than Python: several may be in a row, and they may be after all
+  digits, after ``.``, ``e``, or ``j``, or even inside a radix prefix. Separators
+  before the first digit are still forbidden because e.g. ``_1`` is a legal
+  Python variable name, so it's a symbol in Hy rather than an integer.
 - Integers can begin with leading zeroes, even without a radix prefix like
   ``0x``. Leading zeroes don't automatically cause the literal to be
   interpreted in octal like they do in C. For octal, use the prefix ``0o``, as

tests/test_reader.py

@@ -259,14 +259,15 @@ def test_lex_digit_separators():
 
     assert tokenize("0x_af") == [Integer(0xAF)]
     assert tokenize("0x,af") == [Integer(0xAF)]
+    assert tokenize("0_xaf") == [Integer(0xAF)]
     assert tokenize("0b_010") == [Integer(0b010)]
     assert tokenize("0b,010") == [Integer(0b010)]
     assert tokenize("0o_373") == [Integer(0o373)]
     assert tokenize("0o,373") == [Integer(0o373)]
 
-    assert tokenize("1_2.3,4") == [Float(12.34)]
-    assert tokenize("1_2e3,4") == [Float(12e34)]
-    assert tokenize("1,0_00j") == [Complex(1000j)]
+    assert tokenize("1_2._3,4") == [Float(12.34)]
+    assert tokenize("1_2e_3,4") == [Float(12e34)]
+    assert tokenize("1,0_00j,") == [Complex(1000j)]
 
     assert tokenize("1,,,,___,____,,__,,2__,,,__") == [Integer(12)]
     assert tokenize("_1,,,,___,____,,__,,2__,,,__") == [