Fix the koffice build.

PR:		40342
Submitted by:	Peter Avalos <pavalos@theshell.com>
Forgotten by:	petef :)

1 files changed, 26 insertions, 0 deletions

diff --git a/editors/koffice-kde3/files/patch-kspread_interpreter.cc b/editors/koffice-kde3/files/patch-kspread_interpreter.cc
new file mode 100644
index 000000000000..3b71bb0a64b0
--- /dev/null
+++ b/editors/koffice-kde3/files/patch-kspread_interpreter.cc
@@ -0,0 +1,26 @@
+$FreeBSD$
+
+--- ./kspread/kspread_interpreter.cc.orig	Tue Aug 13 18:08:16 2002
++++ ./kspread/kspread_interpreter.cc	Tue Aug 13 18:09:55 2002
+@@ -2547,10 +2547,10 @@
+         }
+   // This is not correct solution for problem with floating point numbers and probably
+   // will fail in platforms where float and double lenghts are same.
+-  if (approx_equal(floor(args[0]->doubleValue()*pow(10,digits)), args[0]->doubleValue()*pow(10,digits)))
++  if (approx_equal(floor(args[0]->doubleValue()*::pow(10,digits)), args[0]->doubleValue()*::pow(10,digits)))
+       result = args[0]->doubleValue();
+   else
+-      result=floor(args[0]->doubleValue()*pow(10,digits)+1)/pow(10,digits);
++      result=floor(args[0]->doubleValue()*::pow(10,digits)+1)/::pow(10,digits);
+   context.setValue( new KSValue( result) );
+ 
+   return true;
+@@ -3404,7 +3404,7 @@
+         return false;
+         }
+ 
+-  double arg=pow(sqrt(pow(imag,2)+pow(real,2)),args[1]->intValue());
++  double arg=::pow(sqrt(pow(imag,2)+pow(real,2)),args[1]->intValue());
+   double angle=atan(imag/real);
+ 
+   double real_res=arg*cos(angle*args[1]->intValue());