FlixsterPartOneDebuggingChallange

Debugging strategies

Basic debugging workflow

• Understand how to reproduce the issue
  • Retrace steps, focus on the input or sequence of inputs that caused the issue (toasts)
• Figure out the most likely reasons the problem occurred
  • Think about the tasks that would be happening as the app crashes and which ones you suspect might be causing this kind of crash
• Review each suspect and narrow down until you have a culprit
  • Use logcat/toasts/debugger to track down the exact cause of the issue
• Brainstorm potential solutions
  • Think about what you can change to fix this issue
• Resolve the issue
  • Implement the changes and test the app

3 Main ways to debug

• Toasts : small in app notifications
• Logging with logcat : sending data to the log ( viewable from logcat)
• Debugger witgh Breakpoints : Set a breakpoint so you can pause app execution and look at variable states at specific points

Toasts : best as a preventative

Good

• Very fast to set up
• Easy to see while using app normally

Bad

• Can not stack, some may not show if multiple try to play at the same time
• They can not be copied from and do not persist

Best times to use

• Checking if things that fail gracefully (no crash/error ) are working during normal use
• Examples Listeners, and Asynchronous callbacks

Logging : with Logcat

Good

• Easy to check multiple things at the same time
• Can be reviewed after crash
• Most crashes will show up in logcat with a Line #
• Can filter based on options

Bad

• It can be hard to use while using the app, so you may be forced to review after crash or after run.

Best times to use

• First place to check if you get an unexpected crash
• If you are unsure if an API call is working you can print the response URL and check in browser.
• If you are looking for the cause of a crash that is coming form a code you have not written (example glide or room)

Breakpoints : with the Debugger

Good

• Gives you the absolute most information about the state of variables.
• Requires no extra lines of code
• Allows for line by line code execution

Bad

• Line by line code exploration can take a lot of time
• Can make Async functions (API calls) break

Best times to use

• When you know the issue is related to a variable assigned an unexpected value (null, or junk), but you are unsure where
• When you are unable to find the cause via Logging