From: Nicholas Clark <nick@ccl4.org>
Date: Sat, 29 Nov 2003 14:44:22 +0000 (+0000)
Subject: Gconvert actually takes type NV, while nv may be either double
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=6b169c738013e37008468e17b428fca5b8142a66;p=p5sagit%2Fp5-mst-13.2.git

Gconvert actually takes type NV, while nv may be either double
or long double (depending on some conditional code)
Rafael and I think that this cast should work.

p4raw-id: //depot/perl@21804
---

diff --git a/sv.c b/sv.c
index 16c7bbc..3a05916 100644
--- a/sv.c
+++ b/sv.c
@@ -9361,7 +9361,7 @@ Perl_sv_vcatpvfn(pTHX_ SV *sv, const char *pat, STRLEN patlen, va_list *args, SV
 	    if ( !(width || left || plus || alt) && fill != '0'
 		 && has_precis && intsize != 'q' ) {	/* Shortcuts */
 		if ( c == 'g') {
-		    Gconvert(nv, precis, 0, PL_efloatbuf);
+		    Gconvert((NV)nv, precis, 0, PL_efloatbuf);
 		    if (*PL_efloatbuf)	/* May return an empty string for digits==0 */
 			goto float_converted;
 		} else if ( c == 'f' && !precis) {