Approach: We implemented a REPL (Read-Eval-Print Loop) system to continuously accept user input, tokenize commands, and execute tasks based on user-defined operators like pipes and redirection.
Key Steps:
- Session Initialization: Initialization of the main session structure manages each command input's state, including tokens, commands, pipes, and operators.
- Processing the Input Line: Each input line is processed by tokenizing, applying operator rules, and managing pipes, redirections, and documents as needed.
- Task Execution: Commands and sub-commands are executed sequentially, with dependencies managed for pipes and conditional execution.
Approach: To interpret commands, we parse the input line for symbols such as $, |, ', or ".
We create, classify and manage tokens to represent each command segment.
Key Steps:
- Token Generation: The system splits the line into tokens, handles string expansions (e.g., environment variables), and contracts strings when quotes are closed.
- Error Handling: Syntax errors are detected early, with checks for unsupported or misplaced special characters.
Approach: We implemented dynamic pipe allocation and management to enable inter-process communication efficiently.
Key Steps:
- Pipe Allocation: Memory for pipes is allocated dynamically based on the number of tasks in a command sequence.
- Syntax and Logical Validation: Redirections (
<,>,|) are validated for syntax and logical correctness, with error handling for invalid configurations. - Pipeline Execution: Commands are executed sequentially in the pipeline, passing output from one command as input to the next.
Approach: Custom signal handlers manage interruptions gracefully, handling signals like SIGINT and SIGQUIT to prevent shell crashes.
Key Steps:
- Signal Ignoring: Specific signals are ignored during command execution to avoid unexpected termination.
- Custom Handlers: Custom actions for received signals help maintain the shell’s consistent state, even when interrupted.
Approach: Our execution model allows for both built-in commands and external binaries to be run in the shell.
Key Steps:
- Path Resolution: Paths for executable binaries are resolved, checking their existence in the system’s
PATH. - Error Messaging: Informative error messages are provided to the user when execution fails.
- Directory Detection: Directories are validated to prevent mistaken execution as binaries, enhancing reliability.