• Author: David Rodriguez
• Class: CS 452 Section 1
• Semester: Spring 2024

This project is a Shell implementation that allows users to run bash terminal commands. Users can also utilize various operators such as for background processes &, sequential processes ;, pipelines |, and I/O redirection < >.

The hardest part about this project was implementing the pipelines and background processes. Namely, child processes were difficult to work with as I could not find an effective way to debug them (at least not with the gdb debugger). This forced me to learn how the forking process works in bash. The most important lesson I learned from this is that all child processes carry a PID of 0. Knowing this, I knew what PIDs to wait for and how to exclude code just for the child processes and not for the parent process.

I look forward to learning more about how jobs play an important part in execution integrity in future projects. Even though I was able to create a data structure for these processes, I did not find myself using them much in this project. I understand that the jobs command in Shell allows users to monitor and manipulate tasks or processes started in the current shell. Perhaps my program could have used this resource more effectively.

To compile and run the Shell.c program on onyx, cd into the main directory if not done already. Make sure that the corresponding parent GNUmakefile is located in the parent directory (you can get it from ~jbuffenb/classes/452/pub/GNUmakefile). Then, run the following commands:

make try

Then

try

The program will prompt you to run a Bash command after each execution. To exit, use the exit command.

To run the regression Tester, run the following command (also from the main directory):

Test/run