Georgia Tech's CS 3600 TAs created this tutorial to introduce PyCharm to students taking CS 3600. Two main reasons for using PyCharm over a text editor:
- Dynamic error checking and warning messages
- A built-in debugger
We will go over the debugger towards the end of the tutorial. If you already have PyCharm set up and working, you can skip right to that part.
Note: Some of the pictures may look a little different, we used an Early Access Preview version of PyCharm.
JetBrains is a company that creates development tools for programmers. A full list of IDEs and tools can be found here. For our course, we will be using their Python IDE called PyCharm. If you used IntelliJ for CS 1332, this should feel very similar.
You can download PyCharm here. There are two different versions:
- Community: Free for everyone to use and has all the features needed for our course. We recommend this version.
- Professional: Developed for corporations, has lots of enterprise-level features. Requires a license.
JetBrains offers all their professional tools for free to students! You can create a student account here. We won't go into much more detail on this, as this is not a required step.
After downloading and installing PyCharm, run the application. You will have to choose a few options for the initial setup, but this should only take a few seconds.
We recommend that you stick to all the default settings. Feel free to change the theme: Dracula looks similar to Atom and Sublime Text.
This is pretty self explanatory. Make sure the CS 3600 project is downloaded and you know where it is.
Click on the Open button in the welcome screen, select the project folder in the navigation view, and click Ok.
Usually, PyCharm is able to find your default Python installation path. However, if your laptop is set up incorrectly, you will need to explicitly specify the path. Follow these steps to check:
File->Settings- Select
Project->Project Interpreteron the left of the settings menu - Make sure
Python 2.7.*is selected in the dropdown menu (the * can be any number)
If you see python in the dropdown menu, you can skip to the next section. Otherwise:
- Select
Show Allin the dropdown menu. - Click on the green plus button.
- Add the directory to
python.exe. Depending on your operating system, you might have to Google where you can find the executable on your computer. For Windows users, it's in the Python2.7 folder either inC:/Program FilesorC:/Program Files (x86).
If you have trouble with these steps, please check stackoverflow or ask on piazza!
Instead of using the terminal to run the autograder, we will create a Run Configuration. With this, you will be able to re-run the command (and later use the debugger) without having to re-type the entire command.
We wish to run the autograder with no extra parameters (no -q q1). Here is how we can do that:
- Open
autograder.pyby double-clicking on it on theProject Menutoward the left - Scroll down until you see:
if __name__ == "__main__":Click the green run button to left, then press Run. This will run the autograder, which is the equivalent of typing in python autograder.py into the terminal.
Notice how this has created a Run Configuration on the top left. Any files you run will show up in this dropdown menu. To re-run any of these, you can select it in the menu and press on the green arrow to the right.
We can edit the Run Configuration to pass in parameters to the autograder. For example, we can run python pacman.py -q q1 with the following:
- Click on the dropdown menu in the top right, select
Edit Configurations - In the pop-up menu, select the
autograderrun configuration from before. Click on theCopybutton on the top (should be to the right of a red minus). - We have now made a copy of the run configuration that we can change. In the
Namefield, type in something descriptive, like "autograder q1". - In the
script parametersfield, type in the parameters like "-q q1". - Press
Ok
Now when you run "autograder q1", it will run the script with the specified parameters.
This same method can be extended to run other files. For example, to run python pacman.py -l mediumMaze -p SearchAgent -a fn=bfs, just make a new run configuration for "pacman.py" with script parameters "-l mediumMaze -p SearchAgent -a fn=bfs".
The main reason we recommend PyCharm is for the debugger! Before continuing, double check that you understand the Running the Autograder section above. You will need the autograder Run Configuration to run the debugger.
You might be wondering what the debugger is. Basically, it is a tool that lets you set specific Breakpoints where your program will pause execution. Everytime the program pauses, you can:
- Check the values assigned to all your variables
- Step into function calls and check what they return
- Make changes to the code while still "running" the program
- And a lot more...
To use the debugger, we need to set Breakpoints, which are lines of code where the program will pause execution every time it hits that line. To set a breakpoint, single-click on the bar to the right of the line numbers. This creates a red circle, which is a breakpoint. Later, you can click on it again to remove it.
Go ahead and click on the debug button (to the right of the run button at the top). Make sure you selected a Run Configuration that will actually get to that line of code, such as the autograder.
As you can see in the gif, the program paused at that line. The line of code has not been executed yet, and you can view all the variables currently assigned before executing the line. Note that objects are nested, so you can expand each object to see what attributes it has.
This is extremely useful for debugging, certainly better than printing out all your variables!
We can do much more than just pause at a specific line. Here are some more useful features of the debugger:
After pausing the program at a breakpoint, you can have it continue like normal until it reaches another breakpoint. This breakpoint could be the same one being called in a loop, or a completely different one.
Runs the next line of code in the program and then pauses again. It will just go line by line. Note that if you have any function calls, it won't go into the function call stack.
Runs the next line of code in a function call. Basically, it will step into any function that you are currently paused at, unlike step over.
Of course, make sure to hit the stop button after you're done with the debugger.
We'll let you play around with the debugger on your code. Try it out!
Don't run your code in debugger mode if you aren't actually debugging. It's a lot slower.
To show line numbers, right-click on the bar where you set debug breakpoints, then select the show line numbers option.
PyCharm has git and GitHub integration built right into it. You can find out more about it here. No more complaining that your laptop crashed and you lost all your work please...
Of course, we don't want you guys sharing code over GitHub, so please make them private repos. Note that posting your code publicly is an honor code violation. You can set up the repoistory on Gatech's GitHub, as that comes with unlimited free private repos.
You can get unlimited free private repos on the regular GitHub site if you are a student (normally it costs $7 / month for 5 private repos). You need to sign up for a GitHub Student Account. It comes with a bunch of other free products not needed, such as AWS credit, custom domain names, continuous integration build hours, cloud-based solutions, etc.