Bridge: scripts, tools, and the terminal

You finished Parts 1–3 — that’s all the YAML you need to write configurations for chip flows. Part 4 is optional. It shows how a Python script — usually written by your CAD/tools team — actually reads the YAML you wrote and feeds it into an EDA tool.

You will not be expected to write Python after this. The goal is to recognise what is happening when something breaks, and to feel comfortable around the file paths, error messages, and one-line commands that appear in chip-design workflows every day.

The big picture

You write and edit the YAML. The Python script reads it and tells the tool what to do. The tool runs and produces results.

Five words you keep hearing

Before any code, here is what each of these actually means in plain English.

Script

A script is a plain text file containing instructions for a computer, written in a programming language. Common file extensions: .py (Python), .sh (shell). Unlike a compiled program, you can open a script in any text editor and read it.

A Python script is just a .py file you run by typing python script_name.py in the terminal.

Terminal (also called shell, command line, console)

The terminal is the black-or-white text window where you type commands instead of clicking. Same idea as Cadence’s Tcl prompt — text in, text out.

When this tutorial says “run this in the terminal”:

python my_script.py

…it means: open the terminal, type that line, press Enter.

File path

A file path is the address of a file on your computer.

• Relative path: chip_config.yaml — looks in the folder you are currently in
• Absolute path: /home/alice/projects/chip_a/chip_config.yaml — full address

When a script says “file not found”, it almost always means the path is wrong — usually because you ran the script from the wrong folder.

Library

A library is pre-written code somebody else made, packaged so you can use it. Python’s yaml library knows how to read and write YAML files — you just call it. Same idea as a .lib Liberty file in synthesis: ready-made standard cells you reuse.

pip

pip is the package installer for Python — it downloads and installs libraries. You normally run it once per machine:

pip install pyyaml

That command says: “fetch the pyyaml library and put it where Python can find it.” After that, any Python script can use it. In a corporate VLSI environment, your tools team usually runs this on the shared servers — you may never need to.

What does “import” mean?

When Python sees:

import yaml

…it loads the yaml library so the rest of the script can use functions like yaml.safe_load(). Same idea as include "stdlib.v" at the top of a Verilog module — bring in the building blocks before you use them.

The CAD team / engineer split

In nearly every semiconductor team:

This means: most error messages you see come from your YAML, not from broken Python. That is exactly why understanding YAML deeply (Parts 1–3) is far more valuable than learning Python — you will spend 95% of your time editing the config.

What you will see in the next three lessons

• Lesson 12 — three lines of Python that turn your YAML into something the script can use, plus how to look up specific values. You only need to recognise it, not write it.
• Lesson 13 — the reverse: how a script writes a YAML report file after a run.
• Lesson 14 — what happens when something is wrong: the error messages, what they mean, and how to read them back to fix your YAML.

Key takeaways

• A script is a text file with instructions; you run it from a terminal
• A library is reusable code; pip install fetches one; import loads it
• A file path is the address of a file — wrong paths are the #1 cause of “file not found”
• The CAD team writes the script; you write the YAML — error messages usually point back to your YAML
• Part 4 is for recognising what you see, not for writing Python yourself