peter-lam / climbr

A utility to log, visualize, and highlight climbing strengths, weaknesses, and trends. Built using Python, ElasticSearch, Kibana, and Docker

Climbr

Visualize your rock climbing progression and trends.
Website · Screenshots · Wiki · Report Bug · Request Feature

About

Climbr is a utility that allows users to visualize and highlight their climbing strengths, weaknesses, and trends. From professionals to beginners, trad climbers to indoor boulderers, users can log climbing sessions from a selection of templates and instantly view their trends, patterns, and statistics with pre-configured Kibana dashboards.

In addition, Climbr provides booking insights on local gyms (Altitude Kanata, Altitude Gatineau, Coyote Rock Gym), to help users find the most optimal climbing time according to location, time of day, and time of week. Climbers can utilize this information to make socially responsible decisions. Those looking to minimize their risk whilst still staying active can easily find the off-peak hours of their local gym.

Technologies

Contributors

Peter Lam

• Website
• Linkedin
• Resume

Features

• Log your climbing sessions and projects using yaml files
• Visualize climbing trends, patterns, and statistics with dashboards
  • Climbing Stats and Metrics
    • Map of visited locations
    • Most common climbing discipline (Lead, Bouldering, Trad)
    • Grade counter
  • Climbing Trends and Progression
    • Progression in grades over time
    • What's your flash grade? project grade?
    • Strongest/Weakest climbing style? What areas should I improve?
  • Bookings
    • Busiest time to climb
    • Best time to climb to avoid crowds
    • Average bookings based on gym
• Use data to influence your training regime and pinpoint weaknesses
• Make responsible decisions based on booking data

Host Requirements

• Docker 19+
• Python 3.6+ (Optional)
  • Linux: sudo apt-get install python3 && python3 -m pip install -U pip
  • Windows: Ensure 'Add Python to PATH' is selected

Note: Climbr can be run without Docker. The following must be running on your host machine:

• ElasticSearch 7.9.2
• Kibana 7.9.2
• Python 3.6+

Quick Start

Climbr

1. Open Command Prompt/Terminal
2. Change directories to this the root of this repository
3. Build and start up ElasticSearch, Kibana, and Climbr containers with:

   docker-compose up -d --build

   Note: This process may take a while to build
4. Confirm that all containers are running
   • Run docker ps, the containers climbr_cli, climbr_kibana, and climbr_es should be running
   • You should be able to establish a connection to http://localhost:5601
   • If the link is not reachable, wait a couple minutes and try again
5. Run the demo visualizations and data
   • If python is installed:
     1. Run the command python ./climbr.py demo
   • Otherwise:
     1. Copy the CONTAINER ID into your clipboard for the docker container of name climber_cli
     2. Run the following command to initialize the demo

        docker exec <CONTAINER_ID> bash -c "python ./climbr.py demo" 

6. View Climbr visualizations at: http://localhost:5601/app/dashboard

Bookings

1. Open Command Prompt/Terminal
2. Change directories <PATH_TO_REPO>/web_scraper
3. Run webscraper
   • Automated: start up bookings container by running

     docker-compose up -d --build

     Note: Confirm that the container climbr_bookings is running with the command docker ps
   • Manually: with Python

     python bookings.py ./bookings.py -l Altitude_Gatineau

4. View Climbr visualizations at: http://localhost:5601/app/dashboard

Getting Started

Please visit the wiki for more information regarding system requirements, set-up, and usage.

climbr's Issues

Use FastAPI to add an endpoint to create API calls

When session is full, need to parse result properly.

ValueError: invalid literal for int() with base 10: 'Full. Please make a different selection.'
Session info successfully retrieved! See information at '/workspace/data/output/bookings.json'

Also, if it's the last session of the day and it's full, then it will skip to the next day

Convert CRON jobs to python script

Helps with easily reading jobs and making edits when gyms change their schedules: https://stackabuse.com/scheduling-jobs-with-python-crontab/

Add climbers list to config to auto populate all templates

A list of climbers should be used to populate all templates. Maybe leave this in the config file

Set up poetry and nox

Replace requirements.txt and pip setup with poetry to manage dependencies. Set up nox to manage linters, saftey checks etc.

Also fix any linting issues

Provide better error messages for incorreclty formated yaml files

Currently hard to tell what exactly the error is

Visualize number of hours climbing

How many hours of climbing did I do in the past month? week? year?

FIX: Ability to send email via crontab inside docker container

Add workout to climbing log

Add fields and support for workouts when logging a climbing session

Create visualizations for cummalitive booking habits

As climbing gyms begin to remove capacity limits and reservation systems, create some new and update existing visualizations to give climbers a reference point when going to gyms in the future.

What already exists:

• Popularity by gym & time slot
• Relative to the current capacity
• Booking by weather

What could be helpful for the future:

• Popularity by gym & regular time intervals (every hour)
• Add a field to data points to show the % full based on the capacity of that time
  • For example: 50 bookings in March when the capacity was 50 ppl means 100% capacity, where as now it's like 50%
• Update jupyter graphs for weather trends
  • How do other variables influence how busy a gym is (weather, time of the year, events, lockdown)
• Bonus: Highlight and graph the differences in popularity based on lockdown rules
  • For example: 1st return from lockdown, 2nd ....
  • This would be more accurate if curious on data based on the limit but not 100% needed

Make directories and files not tracked by git

Since "Logs" and "Output" directories are ignored by git, the app needs to make these directories if they don't exists. Currently a fresh install results in an error like:

PS E:\GitHub-Projects\climbr-dev-work> poetry run cli export
Retrieving dashboards and related objects...
Exporting Kibana dashboard and objects...
Traceback (most recent call last):
  File "E:\GitHub-Projects\climbr-dev-work\common\common.py", line 383, in export_kibana
    with open(output, "w+") as file:
FileNotFoundError: [Errno 2] No such file or directory: 'E:\\GitHub-Projects\\climbr-dev-work\\data\\output\\climbr_2021-10-25_19-11-35.ndjson'

During handling of the above exception, another exception occurr

Automating booking log creation based on RGPro bookings

When a user books through RGPro, can a log file be automatically generated and filled with the location and date? Maybe something to read emails?

Visualize yearly wrap up

Create a dashboard to summarize the climbing year. Things that would be useful

• Climber count, who did you climb with the most this year?
• Breakeven: Did you break even for your membership? Did other climbers?
• Hardest grade
• Mostly climbed location

Restructure Climbr CLI

Move related files into a subfolder. For example. /common/args should be in the CLI subfolder.

Idea is to clean up the project into sub folders per program:

• Climbr CLI
• webscraper
• Climbr API

Poetry.lock not present in webscraper

On a fresh clone, poetry lock isn't present so there is a error when running docker-compose. Need to find a better solution than this and also duplicating the pyproject.toml....

Refactoring Bookings for Webscraper

Instead of having to change the input logic when something changes, like the booking from (capacity gauge VS reservation), convert the Booking dictionary to class and standardize the inputs/outputs

Fix bulk_api format for "injured" and "date" sections

Convert file paths to use Pathlib Path

More to enhance readability since os.path can be pretty long and less intuitive to write

[webscraper] Does not error when reservations are more than capacity

Need to send an alert that there are more reservations than the current capacity. Likely the coded capacity has changed

Add webscraping support for other gyms

Since other gyms in Montreal and Toronto use RGPro, try to see if the webscraper is robust enough to support other gyms without major maintenance.

Restructure Docker files

Combine webscraper and main project docker files and docker-compose files. These should sit at the top odf the directory,
To restart the webscraper specifically, use: docker-compose restart webscraper or to rebuild the 1 service use: docker-compose up --detach --build webscraper

Add a Github Actions Flow for Nox

Run Nox in Github Actions to check linters, safety on every commit. Prevent merge if doesn't pass tests

Warn users about demo data when running "climbr.py update"

If the user runs "demo" command first and then starts to populate their own data. Runing the "Update" command should either warn the user or auto delete demo files

Check for *demo.njson in the ES folder if user uses regular "climbr.py update". If present, should prompt user about deletion

Implement logging for errors, info, warning

Remove http://host.docker.internal from code

Since all containers now run off 1 docker-compose file, there is not longer a need to have a seperate env variable for docker vs non. Use localhost:xxx instead of http://host.docker.internal: