Lessons Learned from Building Clipboard-to-Anki

1. Overview

• Project Name: Clipboard to Anki
• Purpose: lightweight command-line tool that automatically reads Japanese kanji from your clipboard, retrieves furigana (reading) and English translations, and saves them as a CSV file for easy Anki import. Just copy any Japanese text, run the program, and it generates ready-to-use flashcards—no manual input needed. Perfect for efficient language learning!
• Technologies Used:
  • Python
  • PyPl : pyperclip, pykakasi, deep-translation
• Source of Motivation:
  • Test Driven Development(TDD) -> The Clean Coder
  • UML -> personal interest

---

2. Key Skills and Knowledge Acquired

Programming/Technical Concepts

• Object-oriented design: Learned how to design program into independent classes that has its own responsibility, the relationship between classes.
• UML diagram: First Implication of UML's Object diagram (class components and relationship link)

Frameworks/Tools

• csv: Learned to use Python's csv module to export data.
• time: Learned to use Python's time module to slow down the loop
• 3rd party library: Gained confidence using library services that already built upon the real API call.

Debugging & Problem-Solving

• test code with external library: Use mock.patch() function/class in unittest that redirect that function and inject with designated result
• unrelated object between Deck and FileHandler: Fix __init__() function that takes parent class as an instance

---

3. Challenges Encountered

Technical Challenges

• testing tool & strategy: Struggled with mocking pykakasi instance methods and deep-translate instance class but resolved it by targeting the correct path.

Personal Challenges

• Over-Specification : Balancing between features and time/knowledges required to build the project
• Premature Quitting : Overcoming procrastination when debugging complex issues.

---

4. Improvements to Make in Future Projects

• Code Organization: Refactor the code into smaller, reusable modules for better maintainability.
• Testing: Write more comprehensive unit tests to cover edge cases.
• Performance: -Not for now-

---

5. Reflection

• What Went Well: [E.g., "Successfully implemented CSV export with tag formatting and avoided data inconsistencies."]
• What Could Have Gone Better: Spent too much time debugging due to lack of fundamental concept fluency
• Biggest Takeaway: Start with clear data flow designs to avoid shared state issues.

---

6. Additional Notes

• This project highlight importance of foundational programming concepts. (not only grasp but require the fluency)

---

7. Next Steps

• GCP translation service

---

8. Resources Used

• UML Diagrams: General knowledge of UML
• UML Relationship: Six different type of class relations
• Python| CSV doc: Guide on writing csv file with Python.
• Python| with statement: Review Context Manager and with statement