ref: 2b07a5ae3d8342f171f99958225b523758ed35bc
parent: a50a045890a5a242b11fef0352bc88dcb6c7e876
author: Alexei Podtelezhnikov <[email protected]>
date: Tue Jul 28 19:14:30 EDT 2015
* src/base/ftcalc.c (FT_Vector_NormLen): Clarify.
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,7 @@
+2015-07-28 Alexei Podtelezhnikov <[email protected]>
+
+ * src/base/ftcalc.c (FT_Vector_NormLen): Clarify.
+
2015-07-27 Alexei Podtelezhnikov <[email protected]>
* src/base/ftcalc.c (FT_Vector_NormLen): Explicate type conversions.
--- a/src/base/ftcalc.c
+++ b/src/base/ftcalc.c
@@ -849,13 +849,15 @@
x_ = (FT_Int32)x;
y_ = (FT_Int32)y;
- /* Newton's iterations */
+ /* Newton's iterations */
do
{u = (FT_UInt32)( x_ + ( x_ * b >> 16 ) );
v = (FT_UInt32)( y_ + ( y_ * b >> 16 ) );
- /* converting to signed gives difference with 2^32 */
+ /* Normalized squared length in the parentheses approaches 2^32. */
+ /* On two's complement systems, converting to signed gives the */
+ /* difference with 2^32 even if the expression wraps around. */
z = -(FT_Int32)( u * u + v * v ) / 0x200;
z = z * ( ( 0x10000 + b ) >> 8 ) / 0x10000;
@@ -866,8 +868,10 @@
vector->x = sx < 0 ? -(FT_Pos)u : (FT_Pos)u;
vector->y = sy < 0 ? -(FT_Pos)v : (FT_Pos)v;
- /* true length, again taking advantage of signed difference with 2^32 */
- l = 0x10000 + (FT_Int32)( u * x + v * y ) / 0x10000;
+ /* Conversion to signed helps to recover from likely wrap around */
+ /* in calculating the prenormalized length, because it gives the */
+ /* correct difference with 2^32 on two's complement systems. */
+ l = (FT_UInt32)( 0x10000 + (FT_Int32)( u * x + v * y ) / 0x10000 );
if ( shift > 0 )
l = ( l + ( 1 << ( shift - 1 ) ) ) >> shift;
else