user-guide: Improve documentation of NumDecimalswip/docs-fixes-20018

1 files changed, 16 insertions, 6 deletions

diff --git a/docs/users_guide/exts/num_decimals.rst b/docs/users_guide/exts/num_decimals.rst
index d606599f74..3f31192112 100644
--- a/docs/users_guide/exts/num_decimals.rst
+++ b/docs/users_guide/exts/num_decimals.rst
@@ -8,13 +8,23 @@ Fractional looking integer literals
 
     :since: 7.8.1
 
-    Allow the use of floating-point literal syntax for integral types.
+    Allow the use of scientific notation style borrowed from floating-point literal syntax for integral types.
 
 Haskell 2010 and Haskell 98 define floating literals with the syntax
-``1.2e6``. These literals have the type ``Fractional a => a``.
+``1.2e6``, resembling scientific notation. These literals have the type ``Fractional a => a``.
 
 The language extension :extension:`NumDecimals` allows you to also use the
-floating literal syntax for instances of ``Integral``, and have values
-like ``(1.2e6 :: Num a => a)``
-
-
+scientific notation and floating point literal syntax for instances of
+``Num``, and have values like ``1.2e6 :: Num a => a`` and ``5e10 :: Num a => a``
+. This applies only to literals that really turn out to have integral
+values. For example ``1.23e1 :: Fractional a => a`` since ``1.23e1 == 12.3``,
+however ``1.23e2 :: Num a => a`` as ``1.23e2 == 123``.
+
+Integral literals written using scientific notation will be desugared using
+``fromInteger``, whereas any literals which aren't integral will be desugared
+using ``fromRational`` as usual.
+
+Note that regular floating point literals (without exponents) will also be
+desugared via ``fromInteger`` and assigned type ``Num a => a`` if they
+represent an integral value. For example ``1.0 :: Num a => a``, but
+``1.1 :: Fractional a => a``.