I was tasked with a project to create profiles for all 50 companies in the NZX50. This included creating my own weighted index based on the NZX50 Smartshares fund. This is the master file. In order to keep this up to date I create a new sheet and update the share prices of each company every Friday. This is done in the NZX50.xlsx file. I was further given the task of creating a series of 5 charts charts in drupal 8. This requires me to create and update 5 csvs for each company to upload to the company server. The 5 charts for each company that would be uploaded to the website are created by the 3 .py files you see.
-
Create_share_price.py
-
Create_rank.py
-
Create_cap.py
In addition to other manual tasks I do for this job, this meant I had to update an additional 250 csv files every friday. As such, with no prior coding experience I decided to try and automate this process using Python to save myself time.
Share price:
- Pulled values from csv link on Yahoo Finance.
- Sorted to pull only share prices that fall on Fridays
- Creates csv
Rank:
- Based on my own master file called 'NZX 50.xlsx'
- Pulls the rank of each ticker
- Creates csv
Capitalisation:
- Based on my own master file called 'NZX 50.xlsx'
- Pulls capitalisation of each ticker
- Creates csv
Capitalisation change (week's % change):
- Based on the capitalisation file created under Capitalisation
- Finds the % change in capitalisation between a week and the week before
- Creates csv
Capitalisation change (week's $ change)
- Based on the capitalisation file created under Capitalisation
- Finds the $ change in capitalisation between a week and the week before
- Creates csv
Overall this code has saved me almost 5 hours per week of work.
The way I plan to improve on the code is as follows:
- Combining the capitalisation files into one (completed)
- Calculating the % and $ changes using a loop (rather than the inbuild pct_change function in python).
- Automate the creation and updating of the master NZX 50.xlsx file.
- Updated the project to combine all 5 of the .py files into one large file called "Run.py".
- Automated the process of FTP'ing the files to the website server via FileZilla.
- Created an .exe file and a batch file to run the .exe file. Scheduled it to run every Friday via Task Scheduler in Windows.
General Comments
I took a break from coding for a couple of months and recently got back into it. I've been learning a lot of Java predominantly and now that I've come back to automating parts of my job in Python it's much easier and has taken me only a fraction of time to write new scripts than it took when I wrote my first Python script. I suppose that means I'm getting better.
You will notice the BKBMScraper repo which is the newest addition to my automation scripts in python. I wrote this completely by myself, and it has saved me even more time at work which I can now dedicate to learning more coding. It's my first project where it runs completely hands-free. All I have to do is check that it ran properly. True automation!
My next steps will be:
- Continuing to automate more tasks at work
- Learn CSS/HTML/JavaScript
- Learn Solidity