[PATCH] D46588: [LLDB][lldb-mi] Add possibility to set breakpoints without selecting a target.

[Pavel Labath via Phabricator via llvm-commits]

labath added a comment.

In https://reviews.llvm.org/D46588#1093267, @aprantl wrote:

> Can you give an example of what you mean by "specifying the lldb-mi commands up-front"? it's not immediately obvious to me how that would look like.


In your example `lldb-mi-demo.input` is static text, right? No command depends on the result of any previous commands. That's what I mean by "up-front".

The opposite of that would be if you used the output of one command to construct the next one. Something like:

  result = execute("first command")
  value = extract_something_interesting(result)
  result2 = execute(create_second_command(value))

My point is that lit+FileCheck is good if all your tests look like your example, but it's not if you need to do something like the above. I haven't looked at lldb-mi in much detail, so I can't which of these two options is correct, although it seems that at least the majority of the tests could be written that way.


Repository:
  rL LLVM

https://reviews.llvm.org/D46588