I use tmux on Linux and ConEMU on Windows. Both tools are some kind of terminal multiplexer. I often create a new terminal window, but the window just sends me back to my home directory. There are some ways around this, but I find this to be so annoying that I just created my own set of tools around it.
To solve this problem, I write my "working directory" to a file. This makes a lot of sense for me. When I want to work in a directory, I often wish to do this across reboots. Then when any new shell is created, it changes directory to that working directory. I wrote other tools that allow me to push and pop directories to a text file, allowing me to treat this directory store not just as a slot, but as a stack.
The tools are as follows:
ppushd <dir>: Push directory
<dir>to the persistent directory store, and change directories to it.
ppopd: Pop a directory off of the directory store, and change directories to it.
pchanged: Change directories to the current working directory.
plistd: List the contents of the persistent directory stack.
Finally, I put a
pchanged call at the end of my init shell script (
~/.zshrc for Mac and Linux).
Currently, I only have this implemented for windows, but plan on using the tools on Linux as well in the future.
The last idea -- putting
pchanged at the end of the init script -- feels
dirty, but it's changed my workflow for the better. I context switch a lot less
Hopefully this will prove useful to someone else out there as well :)