First off, I just want to say: Thank You to Alexandr! I've been developing Perl code for a while now without a real IDE, and it is really nice to be in an IDE again, and be able to step into a debugger.
Onto the topic. I have remote debugging working in my current IDE setup. With the way I have things configured I edit files on my laptop, and have a SFTP deployment push files to a remote test machine. When I want to debug:
- I launch my "debug" run configuration
- select my remote machine
- an external tool launches the perl process on the remote machine (ssh session that nohup's the perl process and returns without the terminal killing the process)
- my IDE connects to the perl process
At this point, instead of the debugger breaking into the (local) file that I have open, it opens the file from the remote machine and breaks there. And if I then step into code, that exists in a second file, it opens the second file from the remote machine (instead of using my local copy), and so on. This means that I have to wait for these files to open in order to set breakpoints. Obviously this is not as efficient as just setting breakpoints in local files and launching the debugger.
I am hoping that I have something wrong in my configuration. Is it possible, when remote debugging, to have the debugger use the local copies of files instead of opening the remote files?