Re: bignum branch

[Paul Eggert <eggert@cs.ucla.edu>]

[-- Attachment #1: Type: text/plain, Size: 337 bytes --]

On 07/20/2018 01:02 PM, Achim Gratz wrote:
> maybe the doc string for eql should
> directly indicate that the representations of the FP numbers are
> compared bit-wise, which is distinct from their numerical values as
> prescribed by IEEE754 (and that comparison is done via =)?

Sure, I took a stab at that and installed the attached.


[-- Attachment #2: 0001-src-fns.c-Feql-Fequal-Improve-floating-point-doc.patch --]
[-- Type: text/x-patch, Size: 1602 bytes --]

From 64a168519df8c3f842df11dab01eabbfc35a42c4 Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert@Penguin.CS.UCLA.EDU>
Date: Fri, 20 Jul 2018 13:55:12 -0700
Subject: [PATCH] * src/fns.c (Feql, Fequal): Improve floating-point doc.

---
 src/fns.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/src/fns.c b/src/fns.c
index 7d120a90f7..e7424c3471 100644
--- a/src/fns.c
+++ b/src/fns.c
@@ -2193,8 +2193,10 @@ The PLIST is modified by side effects.  */)
 }
 \f
 DEFUN ("eql", Feql, Seql, 2, 2, 0,
-       doc: /* Return t if the two args are the same Lisp object.
-Floating-point numbers of equal value are `eql', but they may not be `eq'.  */)
+       doc: /* Return t if the two args are `eq' or are indistinguishable numbers.
+Floating-point values with the same sign, exponent and fraction are `eql'.
+This differs from numeric comparison: (eql 0.0 -0.0) returns nil and
+\(eql 0.0e+NaN 0.0e+NaN) returns t, whereas `=' does the opposite.  */)
   (Lisp_Object obj1, Lisp_Object obj2)
 {
   if (FLOATP (obj1))
@@ -2208,8 +2210,8 @@ DEFUN ("equal", Fequal, Sequal, 2, 2, 0,
 They must have the same data type.
 Conses are compared by comparing the cars and the cdrs.
 Vectors and strings are compared element by element.
-Numbers are compared by value, but integers cannot equal floats.
- (Use `=' if you want integers and floats to be able to be equal.)
+Numbers are compared via `eql', so integers do not equal floats.
+\(Use `=' if you want integers and floats to be able to be equal.)
 Symbols must match exactly.  */)
   (Lisp_Object o1, Lisp_Object o2)
 {
-- 
2.17.1