leetcode-problem-picker

Note: This project is vibe coded for my personal use. I don't use Python and have little knowledge of what's happening in the code. All changes are vibe coded using Windsurf AI.

LeetCode-Problem-Picker uses your history to select problems catered to your goals and needs.

Its purpose is to: Track progress in order to identify and improve on areas of weakness Choose problems that are most beneficial: frequently asked by the top tech companies Ensure balance between all subjects with weighted rotation.

During my leetcode journey, I discovered the answer to "which problems should I do next?" depend on goals and progress:

1. Topic Focus: Narrow down to 1+ subjects e.g. "trees, graphs or DP". Intended for learning and focusing on weaknesses.
2. Frequently Asked: Questions from a list, e.g. ones asked by companies or list of Blind's Curated 75. The default.
3. Level Up (WIP): Deduces user's "skill range" for each topic in order to challenge appropriately.
4. Weighted random (TODO): Weighted towards questions with high like count, greater like/dislike ratio, etc.

Setup:

git clone https://github.com/shubhamverma1811/leetcode-problem-picker.git

Usage:

Interactive Helper Menu (Recommended for beginners)

python main.py -m

This launches an interactive wizard that guides you through building your command. It will:

• Show all available topics in an organized layout
• Let you select topics by number instead of typing names
• Allow you to exclude specific topics
• Help you filter by difficulty level
• Generate and optionally execute the final command

Command Line Options

python main.py [-t stack trie graph dp] [-e graph dp] [--list airbnb google blind75 skipped] [-d Easy Medium Hard] [-k 5] [-i]

--help-menu -m      launches interactive helper menu to build your command (recommended for beginners)
--topic_list -t     selects from a pool of problems associated with a subject (e.g. trie, greedy, graph)
--exclude_topics -e excludes problems that belong to specified topics (e.g. dp, graph)
--list -l           chooses problems from one or more text files (comma-delimited)
                    use 'all' to include all problems from all_problems.json
--difficulty -d     filters problems by difficulty level(s): Easy, Medium, Hard
--num_problems -k   number of problems to get
--interactive -i    interactive mode. Preferred way to input data. See section below for more info.
note: no topic or list will result in a problem randomly being selected

Examples:

# Launch the interactive helper menu (easiest way for beginners)
python main.py -m

# Get 5 array problems
python main.py -t arr

# Get 10 linked list problems (using the alias)
python main.py -t ll -k 10

# Get 3 medium difficulty problems from the blind75 list
python main.py -d Medium -k 3

# Get hard topological sort problems from all problems
python main.py -t "top sort" -d Hard -l all

# Get both easy and medium problems from dynamic programming
python main.py -t dp -d Easy Medium -l all

# Get array problems but exclude those that are also graph problems
python main.py -t arr -e graph -l all

# Get linked list problems but exclude those that also involve recursion
python main.py -t ll -e recursion -l all

python main.py --info 91 Displays information about a specific problem: Name, difficulty, Acceptance rate, and companies that ask it

Interactive Mode:

This mode selects and displays a single problem and waits for input:

info                displays details about problem: problem name, difficulty, acceptance rate
hint                displays topics related to a solution
y/n,num_errs,time   Enter data regarding attempt. See next section for details
easy                mark as completed (quickly), then selects a different problem
hard                adds to a hard/skipped list. selects a less similar, less challenging problem
revisit [ID]        mark problem as one to revisit later
refresh [ID]        mark problem as one to "refresh" on later
pause               pause the timer
break               take a break. restarts the timer
quit                stop the program

When No Problems Match All Criteria

If no problems match all your specified criteria (e.g., topics + difficulty + exclusions), the program will:

1. Display clear feedback about why no problems were found
2. Check if there are any uncompleted problems in your selected topics
3. If uncompleted problems exist, offer you a choice to:
   • Show uncompleted problems from your selected topics (ignoring other filters)
   • Continue with the original behavior (try problems from all topics)
4. If all problems in your selected topics are completed, it will notify you and try alternative selections

Completed Problems (optional/recommended):

To maximize this program, you need to maintain a list of completed problems. This can be done in the following ways:

1. interactive mode: After completion of each assigned problem, the result is appended to completed.csv.
2. completed.txt: Best if you have a history of solving leetcode problems. The file expects a comma-delimited list of problem numbers e.g. 78,5,42,1337 To quickly populate with leetcode data:

• Option 1 (Manual): Login to leetcode and visit your list of solved problems. Select "All" rows per page in the bottom-left dropdown.

  • Open up Developer Tools (F12) and run this line in your Console tab: console.log(Array.from(document.querySelectorAll('.reactable-data tr td[label="#"]')).map(x => x.textContent).toString())
  • Copy the list and save into completed.txt

• Option 2 (Automated): Use the provided script to automatically fetch your completed problems via LeetCode API:

  Using a .env file (recommended):

  1. Create a .env file in the project root (copy from .env.example)
  2. Add your LeetCode credentials to the file:

     LEETCODE_USERNAME=your_username
     LEETCODE_SESSION=your_session_cookie
     

  3. Run the script:

     python -m src.utils.fetch_completed

  Using environment variables:

  # Set environment variables with your LeetCode credentials
  export LEETCODE_USERNAME="your_username"
  export LEETCODE_SESSION="your_session_cookie"
  
  # Run the script to update completed.csv
  python -m src.utils.fetch_completed

  To get your LeetCode session cookie:

  1. Log in to LeetCode in your browser
  2. Open Developer Tools (F12) → Application tab → Cookies
  3. Find and copy the value of the LEETCODE_SESSION cookie

3. completed.csv: The file that interactive mode writes to. Provides details about attempt. One problem per line Expected format: LC number,was solved,[num_errors],[time],[date]

LC number           integer
was solved          string. valid inputs: y/n (or yes/no)
num_errors          number of mistakes made when solving problem
time                amount of time spent on the problem (minutes)
date                DateTime. date completed