@alalek The test data for this PR is in : opencv_extra/pull/1046.

@alalek @AleksandrPanov Is there something left to do in this PR from my side ?

@genciberisha I found this issue in the quirc library:
QR code ECC failure #88
support horizontally flipped QR-codes according to ISO 18004:2015 #89

Your solution to the problem is correct, thank you. Test images must be converted to jpg to take up less space.

The PR contains 2 tests that are dependent. It's not stable solution for testing. I propose to remove generator part and hardcode information as static array. There are only 2 images, not need to dump regression data to file.

@AleksandrPanov Converting these images to JPG decreases the size a little but also with a lot off quality loss.
I managed to compress these PNG file to under 3kb size, hope that counts.

@asmorkalov Agreed, I removed the generation test and replaced it with a simple validation test.
Please note because of the nature of the bug we need to validate the data at the end, so I saw it reasonable to
to pack the filename and expected result as a std::pair.

@alalek Seems there isn't anything left here to do, thank you.

There is build error:

/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c: In function 'quirc_flip':
/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c:923:9: warning: missing braces around initializer [-Wmissing-braces]
  struct quirc_code flipped = {0};
         ^
/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c:923:9: warning: (near initialization for 'flipped.corners') [-Wmissing-braces]
/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c:925:2: error: 'for' loop initial declarations are only allowed in C99 mode
  for (int y = 0; y < code->size; y++) {
  ^
/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c:925:2: note: use option -std=c99 or -std=gnu99 to compile your code
/build/4_x_ARMv7-lin/opencv/3rdparty/quirc/src/decode.c:926:3: error: 'for' loop initial declarations are only allowed in C99 mode
   for (int x = 0; x < code->size; x++) {
   ^

@AleksandrPanov Could you fix it?