GithubHelp home page GithubHelp logo

ark's Introduction

ark Invest Tracking and Analytics

Ark Invest holdings tracking and analytics app built on GCP (Google Cloud Platform).

App link

Ark Holdings

Diagram

Future features

  • Daily trade price prediction based on open/high/low/close price, volume, action (Sell/Buy) and past Ark real execution prices
  • More analytical graphs, tables, figures etc.

Code Overview

csv_loader

a GCP Cloud function

  • Trigger: a new (including overrides) csv file uploaded to a specific cloud storage location (nw-msds498-ark-etf-analytics)
  • This function does one task: load the csv file (Ark Invest daily holding, specific format) to BigQuery table: ark.holdings
  • env.yaml - environment variables for cloud function
  • main.py - main logic
  • requirement.txt - python libraries required to run this function

pdf

One-off function to convert Ark Invest's PDF files to csv format, then load it to cloud storage/BigQuery.

  • docs - Store pdf files to be converted temporarily
  • csv - Store converted csv files temporarily
    • upload.sh - Upload converted csvs to cloud storage
  • main.py - convert holdings pdf to csv using tabula and pandas library
  • trade_price.py - Convert ark trade log (with real execution prices) pdf to csv format
  • merge_price.py - merged converted csvs, validates duplicates and inconsistencies as well

price.py

cli

  • ixe.py - core functions to get minute price and day price information using iexcloud API via iexfinance python library
  • config.py - configuration files used by multiple py file. Including reading/preparing API keys, dictionaries of commonly used fields like:
    • foreign symbols that can't be handled now
    • standard API errors
    • standard API request types
    • query all available trading days and symbols etc.
  • get_trade_days.py use pandas_market_calendars library to get trading days between date range
  • token.py - Retrieve iex API token, will deprecate soon

data and log

temporary folders for data and logging

cre_req.sh

Shell script to create python requirements file and remove pkg-resources==0.0.0, this is specific for develop in google cloud shell

get_price.py

A wrapper function to call cli/iex.py

load.sh

Upload files that with specific name pattern to cloud storage then load them from cloud storage to BigQuery

  • Delete table from BigQuery first if needed

mergedf.py

One-off function to merge mulitple csv file in the same format into one.

price.py

WIP python library to get data from iex API (potentially other APIs as well in the future), to solve the efficiency issue with iexfinance library

rename.sh

one off shell script to rename some files produced before

requirement.txt

python libraries required to run everything under price folder

upload.sh

bulk upload files to cloud storage

upload_n_load.sh

local --> cloud storage --> BigQuery

  • takes in two arguments
    • price type, min or day
    • file name to be loaded (on cloud storage)

pull

This folder contains all components of a full-fledged working cloud functions.

It pulls holdings data from Ark Invest website (csv format) and upload the csv files onto cloud storage. (Which would then trigger csv_loader function automatically)

  • env.yaml - environmental variables for cloud function
  • main.py code of the cloud function
  • requirements.txt - list of python libraries needed
  • sh folder - scripts to support deployment of the function etc.
    • deploy.sh - script to deploy this cloud function
    • log_today.sh to get logs of this function (ark-holdings-daily-pull) running on GCP for today.
      • Takes specific date and start time as well
    • list_raw.sh - simple script to list cloud storage where raw files are stored with a specific pattern
    • less_than_7_etfs.sh - old script to see whether load has failed for certainly day that resulted in not all 7 etfs are loaded
    • count_dup.sh - old command line version of BQ query to verify duplicate record counts in ark.holdings table

web

Web interface built with Dash framework on Google App Engine.

  • app.py - python code to create Dash app and server
  • apps folder
    • holdings.py page for ark holdings
    • changes.py page for daily changes in ark holdings
  • main.py - main page of web interface
  • core.py
    • core backend logic to support data presented on web interface
    • filling up data in dropdown of all pages
    • Once selection is made (including the default selection while opening the first page), retrieve data from BigQuery tables and return data to frontend
  • app.yaml - App Engine config file
    • app_flex.yaml - To deploy this app into flexible Google App Engine, override app.yaml using this file
    • app_stand.yaml - To deploy thsi app into standard Google App Engine, override app.yaml using this file.
      • By default app.yaml uses this standard config
  • assets - folder for css file and Ark Invest logo
  • requirements.txt - python libraries required to run this web interface

zWIP (Working in progress)

This fodler contains some deprecated functions, temporary fixes and functions that are not yet running in production

data (pls ignore for now)

Some temporary scripts and data mostly for data fixes

  • fixes - This folder contains data that's been fixed manually and some scripts related to it
  • tmp - trades table related describe, schema etc. info

dedup

A full-fledged cloud function that remove duplicates records from ark.holdings table

  • Still running on cloud function
  • No longer really needed as the problem that would cause duplicate records has been fixed

gmail and oauth (Not working yet)

for gmail authentication to retrieve daily trade notification emails from ARK, not working yet.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.