Re: Surprise in Time#sec

Hi,

At Sat, 1 Jan 2005 09:40:34 +0900,
Steven Jenkins wrote in [ruby-core:04065]:
> Looking at the code in time.c, it's clear that time_sec() indirectly 
> calls either time_localtime() or time_gmtime(), both of which modify the 
> receiver. Is this side effect intended? If so, should the documentation 
> be updated? It was a bit of a suprise to me.

Do you mean the Time object returned by Time#getlocal should
always reset its timezone?


Index: time.c
===================================================================
RCS file: /cvs/ruby/src/ruby/time.c,v
retrieving revision 1.102
diff -p -U2 -r1.102 time.c
--- time.c	23 Jun 2004 12:59:01 -0000	1.102
+++ time.c	4 Jan 2005 10:55:44 -0000
@@ -1165,5 +1165,10 @@ time_getlocaltime(time)
     VALUE time;
 {
-    return time_localtime(time_dup(time));
+    struct time_object *tobj;
+
+    time = time_dup(time);
+    GetTimeval(time, tobj);
+    tobj->tm_got = 0;
+    return time_localtime(time);
 }
 


-- 
Nobu Nakada