Remotecopy, two years later
It’s been over two years since I wrote remotecopy and I still use it every day.
The most recently added feature is the
-c option, which will remove the trailing newline from the copied data if it only contains one line. I found myself writing little scripts that would only output one line with the intent of using that output to build a command line on a different system, and the extra newline at the end often messed up the new command. The
-c solves this problem.
For instance, I have git-url, which outputs the origin url of the current git repository. This makes it easy to clone the repo on a new system (
rc is my alias for
firsthost:gitrepo$ git url | rc Input secret: rc-alaelifj3lij2ijli3ajfwl3iajselfiae
Now the clone url is in my clipboard, so I just type
git clone and then paste to clone on a different system:
secondhost:~$ git clone firstname.lastname@example.org:justone/gitrepo.git Cloning into 'gitrepo'... ...
No tmux pbcopy problems
Most OSX tmux users are familiar with the issues with pbcopy and the current workarounds.
Since remotecopy works by accessing the server over a tcp socket, it’s immune to these problems. I just use remotecopy on my local system as if I were on a remote system.
LA Perl Mongers
At the latest LA Perl Mongers meeting, the talks were lightning in nature, so I threw together a presentation about remotecopy. The interesting source bits are up on github, including a pdf copy of the slides.
For the presentation, I used the excellent js-sequence-diagrams to make this diagram, that hopefully helps show the data flow in a remotecopy interaction.