This site is a static rendering of the Trac instance that was used by R7RS-WG1 for its work on R7RS-small (PDF), which was ratified in 2013. For more information, see Home.

Source for ticket #229

cc

changetime

2012-10-09 13:47:39

component

WG1 - Core

description

For good reasons, `+nan.0` is not `=` to any other number, including itself.  However, `eqv?` is about "sameness" rather than "equality".  I propose that we add two clauses to the definition of `eqv?`, one saying that if both arguments are `+nan.0`, `eqv?` must return `#t`, and if one argument is `+nan.0` and the other is not, `eqv?` must return `#f`.

All of my usual test Schemes already behave this way, with these exceptions:

 * MIT, scsh, SigScheme, Scheme 9 do not have non-finite numbers.
 * Chicken returns `#f` if either argument is `+nan.0`.
 * SISC returns `#t` if either argument is `+nan.0`.

id

keywords

milestone

owner

cowan

priority

major

reporter

cowan

resolution

fixed

severity

status

closed

summary

EQV? and NaN

time

2011-07-11 01:41:51

type

defect

Changes

Change at time 2012-10-09 13:47:39

author

cowan

field

comment

newvalue

oldvalue

raw-time

1349765259614993

ticket

time

2012-10-09 13:47:39

Change at time 2012-10-09 13:47:39

author

cowan

field

resolution

newvalue

fixed

oldvalue

raw-time

1349765259614993

ticket

time

2012-10-09 13:47:39

Change at time 2012-10-09 13:47:39

author

cowan

field

status

newvalue

closed

oldvalue

writing

raw-time

1349765259614993

ticket

time

2012-10-09 13:47:39

Change at time 2012-08-27 07:04:38

author

cowan

field

comment

newvalue

oldvalue

raw-time

1346025878529299

ticket

time

2012-08-27 07:04:38

Change at time 2012-08-27 07:04:38

author

cowan

field

status

newvalue

writing

oldvalue

decided

raw-time

1346025878529299

ticket

time

2012-08-27 07:04:38

Change at time 2012-08-25 00:27:44

author

cowan

field

_comment0

newvalue

1345829336571305

oldvalue

WG1 voted to make no change: the behavior of `eqv?` on procedures is explicitly unspecified.

raw-time

1345829264709953

ticket

time

2012-08-25 00:27:44

Change at time 2012-08-25 00:27:44

author

cowan

field

comment

newvalue

WG1 voted to make no change: the behavior of `eqv?` on !NaNs is explicitly unspecified.

oldvalue

raw-time

1345829264709953

ticket

time

2012-08-25 00:27:44

Change at time 2012-08-25 00:27:44

author

cowan

field

status

newvalue

decided

oldvalue

reopened

raw-time

1345829264709953

ticket

time

2012-08-25 00:27:44

Change at time 2012-07-15 01:19:17

author

cowan

field

comment

newvalue

oldvalue

raw-time

1342289957996047

ticket

time

2012-07-15 01:19:17

Change at time 2012-07-15 01:19:17

author

cowan

field

resolution

newvalue

oldvalue

fixed

raw-time

1342289957996047

ticket

time

2012-07-15 01:19:17

Change at time 2012-07-15 01:19:17

author

cowan

field

status

newvalue

reopened

oldvalue

closed

raw-time

1342289957996047

ticket

time

2012-07-15 01:19:17

Change at time 2012-07-15 01:19:12

author

cowan

field

comment

newvalue

Reopening yet again for conspectus on `eqv?`

oldvalue

raw-time

1342289952744370

ticket

time

2012-07-15 01:19:12

Change at time 2012-07-15 01:19:12

author

cowan

field

resolution

newvalue

fixed

oldvalue

raw-time

1342289952744370

ticket

time

2012-07-15 01:19:12

Change at time 2012-07-15 01:19:12

author

cowan

field

status

newvalue

closed

oldvalue

writing

raw-time

1342289952744370

ticket

time

2012-07-15 01:19:12

Change at time 2012-04-09 22:27:10

author

cowan

field

comment

newvalue

oldvalue

raw-time

1333985230351401

ticket

time

2012-04-09 22:27:10

Change at time 2012-04-09 22:27:10

author

cowan

field

status

newvalue

writing

oldvalue

decided

raw-time

1333985230351401

ticket

time

2012-04-09 22:27:10

Change at time 2012-04-09 22:13:54

author

cowan

field

comment

newvalue

The WG voted to leave the behavior of `eqv?` where both arguments are `+nan.0` explicitly undefined.

oldvalue

raw-time

1333984434409019

ticket

time

2012-04-09 22:13:54

Change at time 2012-04-09 22:13:54

author

cowan

field

status

newvalue

decided

oldvalue

reopened

raw-time

1333984434409019

ticket

time

2012-04-09 22:13:54

Change at time 2012-02-21 03:49:28

author

cowan

field

comment

newvalue

Per Bothner points out that on his system (Java 1.7 on Fedora), nan and nan2 actually have the ''same'' bit patterns, rendering the above experiment mostly meaningless.  This makes me inclined to the R6RS semantics.

oldvalue

raw-time

1329767368871453

ticket

time

2012-02-21 03:49:28

Change at time 2011-10-08 04:28:12

author

cowan

field

comment

newvalue

We need to re-ballot this, because of the false claim that the "same" semantics that people voted for was equivalent to the R6RS semantics.  Here are the new ballot positions:

Let `nan` and `nan2` contain !NaNs, possibly but not necessarily with different bit patterns implying different origins.  In all cases, `(eqv? nan x)`, where x is not a NaN, is required to return `#f`, assuming that `(= nan x)` returns `#f` for any x even in R5RS.

The R5RS semantics requires `(eqv? nan nan)` and `(eqv? nan nan2)` to return `#f`.

The R6RS semantics allows `(eqv? nan nan)` and `(eqv? nan nan2)` to return either `#t` or `#f`.

The "same" semantics requires `(eqv? nan nan)` and `(eqv? nan nan2)` to return `#t`.  

The "same*" semantics requires `(eqv? nan nan)` to return `#t` and allows `(eqv? nan nan2`) to return either `#t` or `#f`.

I ran the following on my usual suite of Schemes on x86 Linux:

{{{
(define nan (/ 0.0 0.0))
(define inf (/ 1.0 0.0))
(define nan2 (- inf inf))
(list (eqv? nan nan) (eqv? nan nan2))
}}}

I confirmed that `nan` and `nan2` expand to different bit patterns using Java's `Double.doubleToRawLongBits` method on the same system.  

 * Racket, Gambit, Guile, Kawa, SISC, Chez, SCM, Larceny, Ypsilon, Mosh, !IronScheme, STklos returned `(#t #t)`, showing that they do not distinguish between `nan` and `nan2`.

 * Bigloo, Chicken, Chibi, Ikarus, KSi, Elk returned `(#t #f)`, showing that they do distinguish.  Chicken complains about division by 0.0, so I used `+nan.0` and `+inf.0` literals.

 * Gauche, UMB, and VX returned `(#f #f)`, following R5RS semantics.

 * MIT, Scheme48/scsh, Scheme 9, Scheme 7, !SigScheme, Oaklisp complained about the divisions by 0.0, and didn't provide any obvious workarounds.

oldvalue

raw-time

1318022892000000

ticket

time

2011-10-08 04:28:12

Change at time 2011-10-08 04:28:12

author

cowan

field

resolution

newvalue

oldvalue

invalid

raw-time

1318022892000000

ticket

time

2011-10-08 04:28:12

Change at time 2011-10-08 04:28:12

author

cowan

field

status

newvalue

reopened

oldvalue

closed

raw-time

1318022892000000

ticket

time

2011-10-08 04:28:12

Change at time 2011-10-08 04:16:40

author

cowan

field

comment

newvalue

oldvalue

raw-time

1318022200000000

ticket

time

2011-10-08 04:16:40

Change at time 2011-10-08 04:16:40

author

cowan

field

resolution

newvalue

invalid

oldvalue

raw-time

1318022200000000

ticket

time

2011-10-08 04:16:40

Change at time 2011-10-08 04:16:40

author

cowan

field

status

newvalue

closed

oldvalue

writing

raw-time

1318022200000000

ticket

time

2011-10-08 04:16:40

Change at time 2011-09-11 23:29:28

author

cowan

field

comment

newvalue

oldvalue

raw-time

1315758568000000

ticket

time

2011-09-11 23:29:28

Change at time 2011-09-11 23:29:28

author

cowan

field

owner

newvalue

cowan

oldvalue

alexshinn

raw-time

1315758568000000

ticket

time

2011-09-11 23:29:28

Change at time 2011-09-11 23:29:28

author

cowan

field

status

newvalue

writing

oldvalue

decided

raw-time

1315758568000000

ticket

time

2011-09-11 23:29:28

Change at time 2011-09-11 07:46:47

author

cowan

field

comment

newvalue

WG1 accepted this proposal.

oldvalue

raw-time

1315702007000000

ticket

time

2011-09-11 07:46:47

Change at time 2011-09-11 07:46:47

author

cowan

field

resolution

newvalue

oldvalue

raw-time

1315702007000000

ticket

time

2011-09-11 07:46:47

Change at time 2011-09-11 07:46:47

author

cowan

field

status

newvalue

decided

oldvalue

new

raw-time

1315702007000000

ticket

time

2011-09-11 07:46:47

Change at time 2011-08-18 18:27:32

author

cowan

field

comment

newvalue

R6RS already requires this behavior.

oldvalue

raw-time

1313666852000000

ticket

time

2011-08-18 18:27:32

Change at time 2011-07-11 13:24:56

author

aag

field

comment

newvalue

MIT Scheme does, in fact, have non-finite numbers, depending on the floating-point error mask that is set.  This isn't documented, however.

oldvalue

raw-time

1310365496000000

ticket

time

2011-07-11 13:24:56