srfi-14 char-set<= problem

srfi-14 char-set<= problem

[Seth Alves]

(import (scheme base) (scheme write) (srfi 14))
(write (char-set<= (char-set #\u) (char-set #\u)))
(newline)
(write (char-set<= (char-set #\u) (char-set #\u #\a)))
(newline)

This prints:

#t
#f

     -seth

Re: srfi-14 char-set<= problem

[Jamison Hope]

[-- Attachment #1: Type: text/plain, Size: 490 bytes --]

On Nov 27, 2014, at 12:45 PM, Seth Alves <alves@hungry.com> wrote:

> (import (scheme base) (scheme write) (srfi 14))
> (write (char-set<= (char-set #\u) (char-set #\u)))
> (newline)
> (write (char-set<= (char-set #\u) (char-set #\u #\a)))
> (newline)
> 
> This prints:
> 
> #t
> #f
> 
>    -seth

Yup, there was a bug in the char-set<= logic.  The attached patch should fix it.  Per, if you would do the honors..

--
Jamison Hope
The PTR Group
www.theptrgroup.com



[-- Attachment #2: srfi14-fix.patch --]
[-- Type: application/octet-stream, Size: 2227 bytes --]

Index: gnu/kawa/slib/ChangeLog
===================================================================
--- gnu/kawa/slib/ChangeLog	(revision 8186)
+++ gnu/kawa/slib/ChangeLog	(working copy)
@@ -1,3 +1,7 @@
+2014-12-01  Jamison Hope  <jrh@theptrgroup.com>
+
+	* srfi14.scm (char-set<=): Fix bug reported by Seth Alves.
+
 2014-11-27  Seth Alves <alves@hungry.com>
 
 	* srfi60.scm: Add more aliases.
Index: gnu/kawa/slib/srfi14.scm
===================================================================
--- gnu/kawa/slib/srfi14.scm	(revision 8186)
+++ gnu/kawa/slib/srfi14.scm	(working copy)
@@ -483,6 +483,9 @@
            ((= (inversion-list (- ai 1))
                (cs:inversion-list (- bi 1)))
             (loop (- ai 2) (- bi 2)))
+           ((> (inversion-list ai)
+               (cs:inversion-list (- bi 1)))
+            (loop ai (- bi 2)))
            (else #f))))
 
   ((contains? (char ::character)) ::boolean
Index: testsuite/ChangeLog
===================================================================
--- testsuite/ChangeLog	(revision 8186)
+++ testsuite/ChangeLog	(working copy)
@@ -1,3 +1,8 @@
+2014-12-01  Jamison Hope  <jrh@theptrgroup.com>
+
+	* lib-test.scm: Add test for char-set<= bug reported by Seth
+	Alves.
+
 2014-11-26   Seth Alves <alves@hungry.com>
 	     Per Bothner  <per@bothner.com>
 
Index: testsuite/lib-test.scm
===================================================================
--- testsuite/lib-test.scm	(revision 8186)
+++ testsuite/lib-test.scm	(working copy)
@@ -1,6 +1,6 @@
 ;; -*- coding: utf-8 -*-
 
-(test-begin "libs" 257)
+(test-begin "libs" 259)
 
 (test-begin "vectors")
 (test-equal '(dah dah didah)
@@ -509,7 +509,7 @@
 
 (test-end)
 
-(test-begin "char-sets" 89)
+(test-begin "char-sets" 91)
 
 (import (srfi :14 char-sets))
 (import (rnrs sorting))
@@ -526,6 +526,8 @@
 (test-equal #t (char-set<= char-set:empty char-set:full))
 (test-equal #t (char-set<= char-set:empty char-set:lower-case
                            char-set:full))
+(test-equal #t (char-set<= (char-set #\u) (char-set #\u)))
+(test-equal #t (char-set<= (char-set #\u) (char-set #\u #\a)))
 ; char-set-hash
 (test-equal #t (= (char-set-hash char-set:empty)
                   (char-set-hash (char-set))))

Re: srfi-14 char-set<= problem

[Per Bothner]

On 11/30/2014 10:02 PM, Jamison Hope wrote:
> On Nov 27, 2014, at 12:45 PM, Seth Alves <alves@hungry.com> wrote:
>
>> (import (scheme base) (scheme write) (srfi 14))
>> (write (char-set<= (char-set #\u) (char-set #\u)))
>> (newline)
>> (write (char-set<= (char-set #\u) (char-set #\u #\a)))
>> (newline)
>>
>> This prints:
>>
>> #t
>> #f
>>
>>     -seth
>
> Yup, there was a bug in the char-set<= logic.  The attached patch should fix it.  Per, if you would do the honors..

Checked in.  Thanks.

-- 
	--Per Bothner
per@bothner.com   http://per.bothner.com/