RFC: removing qasm2.inline

[Roger-luo]

now as we are moving towards squin as our high-level interface instead of qasm2.extended and given #150 and related issues where inline QASM seems causing confusions between Python linter and users. @weinbe58 proposed to remove the qasm2.inline command (+dialect) and instead for loading qasm files we alwayas recommend people to use QASM2 lowering transform to convert a QASM2 file into Method object then call the method object in a kernel (or execute directly).

I'm wondering if we are doing so, is there any other use case for a inline QASM2 command? I guess unlike inline ASM, QASM2 does not provide more low-level access to the hardware thus not very useful to have it inline?

[Roger-luo]

from @stefan-o

To me this sounds fine. The inline command does not seem crucial to me as long as we have an efficient way to translate QASM2 into a Method

[weinbe58]

I'm wondering if we are doing so, is there any other use case for a inline QASM2 command? I guess unlike inline ASM, QASM2 does not provide more low-level access to the hardware thus not very useful to have it inline?

I agree with this statement.