gita / bhagavad-gita-api Goto Github PK

The scripts to add references is very unoptimised and slow because in each iteration it opens a new query for finding the ID of related tuple,
a much faster way is to have a Get All query once and then make a dict out of it and using in memory data to find IDs instead of making new query to server on each pass,

This need to be optimised for future users and contributors
Thanks
Hare Krishna

Logs https://github.com/gita/bhagavad-gita-api/runs/2961739738#step:5:109

pre-commit is using flake8 from local environment

🙁 Actual behavior

In this issue, it seems that the third line is missing in the Sanskrit version of the shloka. But it is completed in Romanized text.

Missing line

Complete shloka from holy-bhagavad-gita.org

🙂 Expected behavior

It should include the third line in the Sanskrit version of Shloka.

🔢 Steps to Reproduce the Problem

1. Go https://bhagavadgita.io/
2. Navigate to chapter 1 verse 20.
3. You can see the third line is missing in the Sanskrit version of the Verse.

Create a cron job that selects a random verse every day at 12 am IST and stores its ID in a new table called verse_of_the_day.

🙁 Actual behavior

So our instagram accont was taken down because of voilation of their policies, using a 3rd party library to post images and stuff
so this made our social media cron job to get deactivated,

🙂 Expected behavior

We need to disable instagram post request and then reactivate the cron job that we are hosting on cron-jobs.org

we need to comment this line

and modify the below if statement to remove instagram condition

if twitter_response == 200 :
                return Response(
                    status_code=201,
                    content="Verse of the day has been posted on Instagram and twitter",
                )

🔢 Steps to Reproduce the Problem

No response

Data from previous version should be collected and cleaned and the GitaVerse model needs to hold 2 more attributes
the word meanings and Transliterations

Once PR #3 is merged, I will start working on these

• #14
• #15
• #16
• reference once and update json data (the scripts that create references are too slow (why ?)) related to #11
• #6
• #7
• #17
• #18
• #19
• #20
• #21

we need to add this arguement automatically when using sqlite

connect_args={"check_same_thread": False}

We need to add Redis to the docker-compose files so that celery can make use of them for verse of the day feature.

What would you like to share?

There are quite a few grammatical errors and typos in the README file which should be corrected at the earliest. Ex: Gita should be preceded by the article 'the' i.e. 'The Gita' which sounds complete and correct.

I would be glad to fix these errors, if I am assigned to this issue.

Create a callback function that queries the vese_of_the_day table at let's say 6 am IST every day and notifies all the subscribed applications of the latest verse of the day.

We could later use this for our own bots. So, when the API notifies our bots of the latest "Verse of the Day", our bots will parse and post it to the relevant social media channel.

Reference guide: https://fastapi.tiangolo.com/advanced/openapi-callbacks/

Scrape and collect the media for all the 700 verses in separate audio files, with an easy naming convention,

Adding Sanskrit recitations from a reliable source to backend, making the field for that and deciding a cdn service for audio files,
as of now we can add the audio files along with JSON stored in https://github.com/gita/gita repo and it would work

Currently, the / route returns 404.

Instead, we should create an index page with a readme, API registration form, and links to docs & repo.

For simplicity, let's just use Jinja2 along with Bootstrap instead of creating a SPA using react/vue.

Can we add Mypy Precommit for static type checking ,
pydocstyle for numpy style documentation

we are manually creating the pydantic schemas. but the can be auto-generated from the sqlalchemy model