My apologies for reviving a dead thread, but I wanted to see if anyone else had to do something similar, and how they approached the problem.
To reiterate, I am constructing a tool that uses CIL in its workflow: my (Python) code currently runs the "cilly" driver through a system call. I tried the first response that was provided, but I have since realized that "configure" and "make" hardcodes certain paths to my system, so I don't know/think that I will be merely able to include "cilly.asm.exe" or "cilly.byte.exe", or any executable, as part of my tool.
I am basically looking for a way to provide the executable without having to include all of the CIL source code, when distributing the tool. If it's not possible, I wouldn't mind having the user configure and make the source code, but I was wondering if it was possible to avoid this extra step for the user.
Thanks for any advice, and happy holidays!
On 28 February 2011 00:36, 林楠 <firstname.lastname@example.org>
run make and you will get an executable of your platform
On Feb 28, 2011 4:08 PM, "Jonathan Kotker" <email@example.com> wrote:
> Hello World.
> I am planning to use CIL as part of the research tool that I am writing.
> Currently, my code constructs and runs a system call that runs the cilly
> driver on my local machine, along with specific arguments. My question is
> this: if I were to release the tool as an open-source tool for all to see
> and use, would it be possible to only include the Cilly driver (or a small
> subset of the CIL files) as part of the files for the tool, or would a user
> have to install CIL (and thus perl and ocaml) before being able to run and
> use my tool?