Theory of quantum software generation

Note: The following is an idea that has only just occurred to me and I feel the need to write as fast as possible.

Unless I am mistaken the core theory of quantum computing is that a Quantum Computer (Q. Computer) generates all possible (or at least billions) of conceivable answers to a problem and then repeatedly narrows it’s criteria until one or a small few answers are left.

In conventional software engineering the range of possible programs on a Digital Computer (DC) is constrained by the digital instruction set, available memory, application programming interface and need to avoid miscellaneous error conditions (segmentation faults, dividing by zero, etcetera). This leaves a non-zero set of programs that can be written.

Therefore a sufficiently large and sophisticated Q. Computer could be used to find the possible algorithms to be executed on a D. Computer and then select appropriate candidates by similar techniques to Unit Testing. These algorithms can then be added into (with some possible modifications such as variable names) into applications with conventionally created user interfaces.

Additionally these candidates will only have to generated once after which they can be stored to disk to see if they are appropriate for other projects.

Also the creation of a programming language with a difficult to obfuscate syntax (as opposed to C++ or Haskell) would be beneficial for this process.

Finally I believe that this process should put serious doubt on the eligibility of computer programs for patent and even copyright.