From ae4ff9bc05203dce92f1fe0de4ec4edc9791583a Mon Sep 17 00:00:00 2001
From: Laurent Rineau <laurent.rineau@cgal.org>
Date: Wed, 13 Nov 2013 11:15:27 +0100
Subject: [PATCH] Minor rephrasing about the exactness of Mpfz(double)

---
 Number_types/doc/Number_types/CGAL/Mpzf.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Number_types/doc/Number_types/CGAL/Mpzf.h b/Number_types/doc/Number_types/CGAL/Mpzf.h
index d165b4e7dc2..86dd53c6613 100644
--- a/Number_types/doc/Number_types/CGAL/Mpzf.h
+++ b/Number_types/doc/Number_types/CGAL/Mpzf.h
@@ -9,8 +9,8 @@ number which can represent numbers of the form \f$ m*2^e\f$, where \f$
 m\f$ is an arbitrary precision integer based on the <span
 class="textsc">GMP</span> library, and \f$ e\f$ is of type `int`. This
 type can be considered exact, even if the exponent is not a
-multiple-precision number. A `Mpzf` constructed from integer and a floating point
-number type represents exactly this number, the input is valid and finite.
+multiple-precision number. A `Mpzf` constructed from an integer or a
+normalized finite floating point number represents exactly that number.
 This number type offers functionality very
 similar to `MP_Float` and `Gmpzf` but is faster.