Fix alpha blending with with offset; should help with #1155.

diff --cc src/lib/image.cc
index 33a0077db8e40d49bc40902a6c5879b4bba69e3d,a6354f2d16793c8790930f7eb4f622bbfd1db7d0..b85451fb32f11ef4101363f27544eba9961c04e4
--- 1/src/lib/image.cc
--- 2/src/lib/image.cc
+++ b/src/lib/image.cc
@@@ -436,40 -430,36 +436,6 @@@ Image::make_transparent (
        memset (data()[0], 0, sample_size(0).height * stride()[0]);
  }
  
--template <class T>
--void
--component (
--      int n,
--      Image* base,
--      shared_ptr<const Image> other,
--      shared_ptr<const Image> rgba,
--      int start_base_x, int start_base_y,
--      int start_other_x, int start_other_y
--      )
--{
--      dcp::Size const base_size = base->sample_size(n);
--      dcp::Size const other_size = other->sample_size(n);
-       int const bhf = base->horizontal_factor(n);
-       int const bvf = base->vertical_factor(n);
-       int const ohf = other->horizontal_factor(n);
-       int const ovf = other->vertical_factor(n);
-       for (int by = start_base_y / bvf, oy = start_other_y / ovf, ry = start_other_y; by < base_size.height && oy < other_size.height; ++by, ++oy, ry += ovf) {
 -      for (int by = start_base_y, oy = start_other_y; by < base_size.height && oy < other_size.height; ++by, ++oy) {
--              /* base image */
-               T* bp = ((T*) (base->data()[n] + by * base->stride()[n])) + start_base_x / bhf;
 -              T* bp = ((T*) (base->data()[n] + by * base->stride()[n])) + start_base_x;
--              /* overlay image */
--              T* op = ((T*) (other->data()[n] + oy * other->stride()[n]));
--              /* original RGBA for alpha channel */
-               uint8_t* rp = rgba->data()[0] + ry * rgba->stride()[0];
-               for (int bx = start_base_x / bhf, ox = start_other_x / ohf; bx < base_size.width && ox < other_size.width; ++bx, ++ox) {
 -              uint8_t* rp = rgba->data()[0] + oy * rgba->stride()[0];
 -              for (int bx = start_base_x, ox = start_other_x; bx < base_size.width && ox < other_size.width; ++bx, ++ox) {
--                      float const alpha = float (rp[3]) / 255;
--                      *bp = *op * alpha + *bp * (1 - alpha);
--                      ++bp;
--                      ++op;
-                       rp += 4 * ohf;
 -                      rp += 4;
--              }
--      }
--}
--
  void
  Image::alpha_blend (shared_ptr<const Image> other, Position<int> position)
  {
@@@ -589,19 -579,19 +555,67 @@@
        }
        case AV_PIX_FMT_YUV420P:
        {
 -              shared_ptr<Image> yuv = other->scale (other->size(), dcp::YUV_TO_RGB_REC709, _pixel_format, false, false);
 -              component<uint8_t> (0, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 -              component<uint8_t> (1, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 -              component<uint8_t> (2, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 +              shared_ptr<Image> yuv = other->convert_pixel_format (dcp::YUV_TO_RGB_REC709, _pixel_format, false, false);
-               component<uint8_t> (0, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
-               component<uint8_t> (1, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
-               component<uint8_t> (2, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
++              dcp::Size const ts = size();
++              dcp::Size const os = yuv->size();
++              for (int ty = start_ty, oy = start_oy; ty < ts.height && oy < os.height; ++ty, ++oy) {
++                      int const hty = ty / 2;
++                      int const hoy = oy / 2;
++                      uint8_t* tY = data()[0] + (ty * stride()[0]) + start_tx;
++                      uint8_t* tU = data()[1] + (hty * stride()[1]) + start_tx / 2;
++                      uint8_t* tV = data()[2] + (hty * stride()[2]) + start_tx / 2;
++                      uint8_t* oY = yuv->data()[0] + (oy * yuv->stride()[0]) + start_ox;
++                      uint8_t* oU = yuv->data()[1] + (hoy * yuv->stride()[1]) + start_ox / 2;
++                      uint8_t* oV = yuv->data()[2] + (hoy * yuv->stride()[2]) + start_ox / 2;
++                      uint8_t* alpha = other->data()[0] + (oy * other->stride()[0]) + start_ox * 4;
++                      for (int tx = start_tx, ox = start_ox; tx < ts.width && ox < os.width; ++tx, ++ox) {
++                              float const a = float(alpha[3]) / 255;
++                              *tY = *oY * a + *tY * (1 - a);
++                              *tU = *oU * a + *tU * (1 - a);
++                              *tV = *oV * a + *tV * (1 - a);
++                              ++tY;
++                              ++oY;
++                              if (tx % 2) {
++                                      ++tU;
++                                      ++tV;
++                              }
++                              if (ox % 2) {
++                                      ++oU;
++                                      ++oV;
++                              }
++                              alpha += 4;
++                      }
++              }
                break;
        }
        case AV_PIX_FMT_YUV420P10:
        case AV_PIX_FMT_YUV422P10LE:
        {
 -              shared_ptr<Image> yuv = other->scale (other->size(), dcp::YUV_TO_RGB_REC709, _pixel_format, false, false);
 -              component<uint16_t> (0, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 -              component<uint8_t>  (1, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 -              component<uint8_t>  (2, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
 +              shared_ptr<Image> yuv = other->convert_pixel_format (dcp::YUV_TO_RGB_REC709, _pixel_format, false, false);
-               component<uint16_t> (0, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
-               component<uint8_t>  (1, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
-               component<uint8_t>  (2, this, yuv, other, start_tx, start_ty, start_ox, start_oy);
++              dcp::Size const ts = size();
++              dcp::Size const os = yuv->size();
++              for (int ty = start_ty, oy = start_oy; ty < ts.height && oy < os.height; ++ty, ++oy) {
++                      uint16_t* tY = (uint16_t *) (data()[0] + (ty * stride()[0])) + start_tx;
++                      uint8_t* tU = data()[1] + (ty * stride()[1]) + start_tx;
++                      uint8_t* tV = data()[2] + (ty * stride()[2]) + start_tx;
++                      uint16_t* oY = (uint16_t *) (yuv->data()[0] + (oy * yuv->stride()[0])) + start_ox;
++                      uint8_t* oU = yuv->data()[1] + (oy * yuv->stride()[1]) + start_ox;
++                      uint8_t* oV = yuv->data()[2] + (oy * yuv->stride()[2]) + start_ox;
++                      uint8_t* alpha = other->data()[0] + (oy * other->stride()[0]) + start_ox * 4;
++                      for (int tx = start_tx, ox = start_ox; tx < ts.width && ox < os.width; ++tx, ++ox) {
++                              float const a = float(alpha[3]) / 255;
++                              *tY = *oY * a + *tY * (1 - a);
++                              *tU = *oU * a + *tU * (1 - a);
++                              *tV = *oV * a + *tV * (1 - a);
++                              ++tY;
++                              ++tU;
++                              ++tV;
++                              ++oY;
++                              ++oU;
++                              ++oV;
++                              alpha += 4;
++                      }
++              }
                break;
        }
        default: