From 78ed1bb287a1c8a12e89737a8eb8dca66c52df9c Mon Sep 17 00:00:00 2001 From: Elias Batek Date: Sun, 12 Jan 2025 02:27:53 +0100 Subject: [PATCH] Rename `UInt32p64` to `UDecimal` --- pixmappaint.d | 121 ++++++++++++++++++++++++++------------------------ 1 file changed, 62 insertions(+), 59 deletions(-) diff --git a/pixmappaint.d b/pixmappaint.d index b7188bc..c520613 100644 --- a/pixmappaint.d +++ b/pixmappaint.d @@ -375,9 +375,12 @@ static assert(Pixel.sizeof == uint.sizeof); } /++ - Unsigned 32-bit integer type with 64-bit precision for intermediate calculations + Unsigned 64-bit fixed-point decimal type + + Assigns 32 bits to the digits of the pre-decimal point portion + and the other 32 bits to the digits of the fractional part. +/ -struct UInt32p64 { +struct UDecimal { private { ulong _value = 0; } @@ -389,8 +392,8 @@ struct UInt32p64 { _value = (long(initialValue) << 32); } - private static UInt32p64 make(ulong internal) { - auto result = UInt32p64(); + private static UDecimal make(ulong internal) { + auto result = UDecimal(); result._value = internal; return result; } @@ -401,7 +404,7 @@ struct UInt32p64 { } /// - public UInt32p64 round() const { + public UDecimal round() const { const truncated = (_value & 0xFFFF_FFFF_0000_0000); const delta = _value - truncated; @@ -411,17 +414,17 @@ struct UInt32p64 { : truncated; // dfmt on - return UInt32p64.make(rounded); + return UDecimal.make(rounded); } /// - public UInt32p64 floor() const { + public UDecimal floor() const { const truncated = (_value & 0xFFFF_FFFF_0000_0000); - return UInt32p64.make(truncated); + return UDecimal.make(truncated); } /// - public UInt32p64 ceil() const { + public UDecimal ceil() const { const truncated = (_value & 0xFFFF_FFFF_0000_0000); // dfmt off @@ -430,50 +433,50 @@ struct UInt32p64 { : truncated; // dfmt on - return UInt32p64.make(ceiling); + return UDecimal.make(ceiling); } public { /// - UInt32p64 opBinary(string op : "+")(const uint rhs) const { - return UInt32p64.make(_value + (ulong(rhs) << 32)); + UDecimal opBinary(string op : "+")(const uint rhs) const { + return UDecimal.make(_value + (ulong(rhs) << 32)); } /// ditto - UInt32p64 opBinary(string op : "-")(const uint rhs) const { - return UInt32p64.make(_value - (ulong(rhs) << 32)); + UDecimal opBinary(string op : "-")(const uint rhs) const { + return UDecimal.make(_value - (ulong(rhs) << 32)); } /// ditto - UInt32p64 opBinary(string op : "*")(const uint rhs) const { - return UInt32p64.make(_value * rhs); + UDecimal opBinary(string op : "*")(const uint rhs) const { + return UDecimal.make(_value * rhs); } /// ditto - UInt32p64 opBinary(string op : "/")(const uint rhs) const { - return UInt32p64.make(_value / rhs); + UDecimal opBinary(string op : "/")(const uint rhs) const { + return UDecimal.make(_value / rhs); } } public { /// - UInt32p64 opBinaryRight(string op : "+")(const uint lhs) const { - return UInt32p64.make((ulong(lhs) << 32) + _value); + UDecimal opBinaryRight(string op : "+")(const uint lhs) const { + return UDecimal.make((ulong(lhs) << 32) + _value); } /// ditto - UInt32p64 opBinaryRight(string op : "-")(const uint lhs) const { - return UInt32p64.make((ulong(lhs) << 32) - _value); + UDecimal opBinaryRight(string op : "-")(const uint lhs) const { + return UDecimal.make((ulong(lhs) << 32) - _value); } /// ditto - UInt32p64 opBinaryRight(string op : "*")(const uint lhs) const { - return UInt32p64.make(lhs * _value); + UDecimal opBinaryRight(string op : "*")(const uint lhs) const { + return UDecimal.make(lhs * _value); } /// ditto - UInt32p64 opBinaryRight(string op : "/")(const uint) const { - static assert(false, "Use `int() / cast(int)(UInt32p64())` instead."); + UDecimal opBinaryRight(string op : "/")(const uint) const { + static assert(false, "Use `int() / cast(int)(UDecimal())` instead."); } } @@ -505,45 +508,45 @@ struct UInt32p64 { } @safe unittest { - assert(UInt32p64(uint.max).castTo!uint == uint.max); - assert(UInt32p64(uint.min).castTo!uint == uint.min); - assert(UInt32p64(1).castTo!uint == 1); - assert(UInt32p64(2).castTo!uint == 2); - assert(UInt32p64(1_991_007).castTo!uint == 1_991_007); + assert(UDecimal(uint.max).castTo!uint == uint.max); + assert(UDecimal(uint.min).castTo!uint == uint.min); + assert(UDecimal(1).castTo!uint == 1); + assert(UDecimal(2).castTo!uint == 2); + assert(UDecimal(1_991_007).castTo!uint == 1_991_007); - assert((UInt32p64(10) + 9).castTo!uint == 19); - assert((UInt32p64(10) - 9).castTo!uint == 1); - assert((UInt32p64(10) * 9).castTo!uint == 90); - assert((UInt32p64(99) / 9).castTo!uint == 11); + assert((UDecimal(10) + 9).castTo!uint == 19); + assert((UDecimal(10) - 9).castTo!uint == 1); + assert((UDecimal(10) * 9).castTo!uint == 90); + assert((UDecimal(99) / 9).castTo!uint == 11); - assert((4 + UInt32p64(4)).castTo!uint == 8); - assert((4 - UInt32p64(4)).castTo!uint == 0); - assert((4 * UInt32p64(4)).castTo!uint == 16); + assert((4 + UDecimal(4)).castTo!uint == 8); + assert((4 - UDecimal(4)).castTo!uint == 0); + assert((4 * UDecimal(4)).castTo!uint == 16); - assert((UInt32p64(uint.max) / 2).castTo!uint == 2_147_483_647); - assert((UInt32p64(uint.max) / 2).round().castTo!uint == 2_147_483_648); + assert((UDecimal(uint.max) / 2).castTo!uint == 2_147_483_647); + assert((UDecimal(uint.max) / 2).round().castTo!uint == 2_147_483_648); - assert((UInt32p64(10) / 8).round().castTo!uint == 1); - assert((UInt32p64(10) / 8).floor().castTo!uint == 1); - assert((UInt32p64(10) / 8).ceil().castTo!uint == 2); + assert((UDecimal(10) / 8).round().castTo!uint == 1); + assert((UDecimal(10) / 8).floor().castTo!uint == 1); + assert((UDecimal(10) / 8).ceil().castTo!uint == 2); - assert((UInt32p64(10) / 4).round().castTo!uint == 3); - assert((UInt32p64(10) / 4).floor().castTo!uint == 2); - assert((UInt32p64(10) / 4).ceil().castTo!uint == 3); + assert((UDecimal(10) / 4).round().castTo!uint == 3); + assert((UDecimal(10) / 4).floor().castTo!uint == 2); + assert((UDecimal(10) / 4).ceil().castTo!uint == 3); - assert((UInt32p64(10) / 5).round().castTo!uint == 2); - assert((UInt32p64(10) / 5).floor().castTo!uint == 2); - assert((UInt32p64(10) / 5).ceil().castTo!uint == 2); + assert((UDecimal(10) / 5).round().castTo!uint == 2); + assert((UDecimal(10) / 5).floor().castTo!uint == 2); + assert((UDecimal(10) / 5).ceil().castTo!uint == 2); } @safe unittest { - UInt32p64 val; + UDecimal val; - val = UInt32p64(10); + val = UDecimal(10); val += 12; assert(val.castTo!uint == 22); - val = UInt32p64(1024); + val = UDecimal(1024); val -= 24; assert(val.castTo!uint == 1000); val -= 100; @@ -551,20 +554,20 @@ struct UInt32p64 { val += 5; assert(val.castTo!uint == 905); - val = UInt32p64(256); + val = UDecimal(256); val *= 4; assert(val.castTo!uint == (256 * 4)); - val = UInt32p64(2048); + val = UDecimal(2048); val /= 10; val *= 10; assert(val.castTo!uint == 2047); } @safe unittest { - UInt32p64 val; + UDecimal val; - val = UInt32p64(9_000_000); + val = UDecimal(9_000_000); val /= 13; val *= 4; @@ -573,7 +576,7 @@ struct UInt32p64 { assert(val.round().castTo!uint == 2_769_231); // assert(uint(9_000_000) / uint(13) * uint(4) == 2_769_228); - val = UInt32p64(64); + val = UDecimal(64); val /= 31; val *= 30; val /= 29; @@ -2856,8 +2859,8 @@ private void scaleToImpl(ScalingFilter method)(const Pixmap source, Pixmap targe const ScalingDirection directionX = scalingDirectionFromDelta(delta.width); const ScalingDirection directionY = scalingDirectionFromDelta(delta.height); - const ratioX = (UInt32p64(source.width) / target.width); - const ratioY = (UInt32p64(source.height) / target.height); + const ratioX = (UDecimal(source.width) / target.width); + const ratioY = (UDecimal(source.height) / target.height); Point translate(const Point dstPos) { pragma(inline, true);