toddbirchard / blog-webhook-api

📑 🎛️ API to automate optimizations for self-hosted blogging platforms.

Personal	HackersAndSlackers

blog-webhook-api's Issues

Members: Create “/account/[ID]/“ GET endpoint for a single user ID

┆Issue is synchronized with this Jira Story

Add docstrings to all Pytest fixtures & unit tests

┆Issue is synchronized with this Jira Task

Historical searches: Save Algolia searches to a persistent table

GET /searches/historical: Create a historical database of all searches performed.

Comments: Zapier errors fail to persist commands to database

┆Issue is synchronized with this Jira Task

Data: post metadata endpoint reports success but has no effect

Comment creation fails spectacularly

• All new comments fail due to mailgun
• No meaningful messages are logged
• Every request returns inaccurate 200 response code along with incorrect description

┆Issue is synchronized with this Jira Bug

Netlify: Remove all deprecated netlify-related logic

┆Issue is synchronized with this Jira Story

README: Document endpoints that have been added, removed, or changed

┆Issue is synchronized with this Jira Story

Tags: Create GET endpoint to list all tags

┆Issue is synchronized with this Jira Story

Donations endpoint returns failed response

https://zapier.com/app/history?context=2359597&root_id=95737823&status=error&utm_source=error&utm_medium=email&utm_campaign=Error%20Alert#

┆Issue is synchronized with this Jira Bug

Top Searches: Truncate table before each query

GET /searches/top: Pull top searches from Algolia to generate search suggestions based on popularity.

Truncate existing table each time

Test coverage for ‘donation’ endpoints

┆Issue is synchronized with this Jira Story

Approach 80% test coverage

┆Issue is synchronized with this Jira Story

New Endpoint

Changes which require a new API endpoint to be created.

Dependency Dashboard

This issue provides visibility into Renovate updates and their statuses. Learn more

Open

These updates have all been created already. Click a checkbox below to force a retry/rebase of any.

• Update dependency numpy to v1.21.4
• Update github/super-linter Docker tag to v4.8.4
• Update dependency pyarrow to v6
• Click on this checkbox to rebase all open PRs at once

---

• Check this box to trigger a request for Renovate to run again on this repository

┆Issue is synchronized with this Jira Task

Testing: Update compatibility of unit & integration tests with recent FastAPI test syntax

Update FastAPI test syntax.

Also resolve underlying issues for tests with legitimate failures.

┆Issue is synchronized with this Jira Bug

Authors: Repurpose GET “/authors/“ to display read-only list of authors

┆Issue is synchronized with this Jira Story

Pydantic-settings: Migrate app config to utilize “pydantic-settings” standard library

• Github PR: https://github.com/toddbirchard/blog-webhook-api/pull/304

Note: current implementation throws a “pickle” error at runtime or via make test

┆Issue is synchronized with this Jira Story

Members: Utilize GeoIP to enrich member data with location tags

See brobot implementation

• Client: https://github.com/toddbirchard/broiestbot/blob/master/clients/geo.py
• Model: https://github.com/toddbirchard/broiestbot/blob/master/clients/geo.py

┆Issue is synchronized with this Jira Story

JWT encode deprecated

https://app.datadoghq.com/logs?query=-status%3Aok%20service%3A%28api%20OR%20pythonmyadmin%20OR%20hackersandslackers-app%20OR%20wp-admin-redirect%29&cols=host%2Cservice&event=AQAAAXrdHFDfa5GktQAAAABBWHJkSEZWTEFBQW94ODNOTE4weDlnQUE&index=&messageDisplay=inline&stream_sort=time%2Cdesc&from_ts=1627124939859&to_ts=1627211339859&live=true

┆Issue is synchronized with this Jira Bug

Create mailgun “welcome” endpoint

Comments fail on schema validation

• Zapier: https://zapier.com/app/editor/98969950/nodes/99000294/fields

┆Issue is synchronized with this Jira Bug
┆Attachments: IMG_1867.PNG

PSN: Reimplement PlayStation interactive commands

• Github: https://github.com/isFakeAccount/psnawp
• Docs: https://psnawp.readthedocs.io/en/latest/

┆Issue is synchronized with this Jira Story

Database: Utilize async ‘databases’ library

• databases in FastAPI: https://fastapi.tiangolo.com/advanced/async-sql-databases/
• Aiomysql: https://pypi.org/project/aiomysql/

┆Issue is synchronized with this Jira Story

Populate google storage /roundup directory with updated lynx images (include retina)

┆Issue is synchronized with this Jira Task

Database: Create SQLAlchemy data models for existing Ghost tables

┆Issue is synchronized with this Jira Story

Upgrade SQLAlchemy to v2.0

• Announcement: https://www.sqlalchemy.org/blog/2023/01/26/sqlalchemy-2.0.0-released/
• Migration guide: https://docs.sqlalchemy.org/en/20/changelog/migration_20.html

┆Issue is synchronized with this Jira Story

Images: Ensure image “moves” happen prior to transforms

┆Issue is synchronized with this Jira Story

DFS: SMS notification for EPL & UCL roster XI

DFS

• Fetch upcoming user rosters for UCL and EPL competitions
  **

┆Issue is synchronized with this Jira Major Functionality

Posts: Upon post publish, revert posts lacking HTML bodies to “drafts”

┆Issue is synchronized with this Jira Story

Action Required: Fix Renovate Configuration

There is an error with this repository's Renovate configuration that needs to be fixed. As a precaution, Renovate will stop PRs until it is resolved.

Error type: undefined. Note: this is a nested preset so please contact the preset author if you are unable to fix it yourself.

Posts: Return “/posts” responses as serialized models

Replace all instances where JSONResponse is returned by a /post endpoint with serialized Pydantic Models.

Random link?

https://roman.pt/posts/dont-let-dicts-spoil-your-code/

┆Issue is synchronized with this Jira Story

Lynx: Delay feature image update by 1 hour

Trigger on post publish: if feature image does not live in gcs /lynx directory, assign new image and corresponding social images

Images: Transform single post image

POST /images/transform/: Transform image for a single post passed in body.

Post: Update metadata upon individual post update

Authors: Create GET “/authors/profile“ to run profile update queries formerly handled by “/authors/“

┆Issue is synchronized with this Jira Story

Posts: Send SMS when author posts need feature images, etc

Analytics: Consume ‘Plausible’ API for site metrics

Fails as follows:

• Plausible API returns list of top-visited URLs
• Logic attempts to fetch corresponding post per URL from ghost admin api
  • “pages” and “posts” are considered different entities
  • Failure to filter page slugs results in error

┆Issue is synchronized with this Jira Bug

Bump Project Python version to 3.10

• Github PR: https://github.com/toddbirchard/blog-webhook-api/pull/295

​Known Issues

I. Running locally fails to start

Running locally via {{unicorn }}results in the following error:

TypeError: cannot pickle _cffi_backend.FFI object

The make run command previously bypassed the usage of ASGI in development by instantiating the app via Python directly: python -m main.py.

Reverting changes to make run will likely resolve the issue, as main.py already handles uvicorn instantiation via uvicorn.run().

II. Poetry Update fails from apparent timeouts

Poetry fails to resolve package via “pythonhosted.” Attempting to install or update packages via poetry results in infinite failed requests.

Potential resources:

• Reinstall poetry without brew install: https://stackoverflow.com/questions/73365780/why-is-not-recommended-to-install-poetry-with-homebrew
• Update Poetry’s venv version: https://github.com/python-poetry/poetry/issues/5490#issuecomment-1113409455
• Recreate venv via poetry: https://stackoverflow.com/questions/65783409/poetry-install-fails-with-envcommanderror-looks-for-version-2020-12-21-3-lambda
  • “We got a similar issue with corrupted artifact and solved it by removing the artifact itself from the cache system where it complained (in your case: /home/Library/Caches/pypoetry/artifacts/22/36/ca/c8aae41f80011be881aa337cc80abc25ffd8542471c325e4410afa86c8/authorities-deployer-2020.12.21.3-lambda.zip).”
• Ensure pyenv global is set to 3.10 & recreate the project’s poetry environment via poetry init

┆Issue is synchronized with this Jira Story

FastAPI: Utilize FastAPI's `SQLModel` & `Databases` Plugins

FastAPI Database Libraries

The implementation of FastAPI sister libraries SQLModel and Databases is highly optional. These libraries intend to improve developer experience; they are not required, nor do they offer significant performance benefits.

1. https://sqlmodel.tiangolo.com/

Extends SQLAlchemy’s “model” object with Pydantic syntax to avoid code duplication between Models and Schemas. Such duplication is otherwise nearly inevitable per FastAPI’s default pattern of defining SQLAlchemy https://fastapi.tiangolo.com/tutorial/sql-databases/?h=database#create-sqlalchemy-models-from-the-base-class alongside https://fastapi.tiangolo.com/tutorial/sql-databases/?h=database#create-sqlalchemy-models-from-the-base-class. Such duplication already exists in database/models.py and database/schemas.py, respectively.

2. https://fastapi.tiangolo.com/how-to/async-sql-encode-databases/?h=database#import-and-set-up-databases

FastAPI’s databases library allows predefined SQLAlchemy database models to build API responses defined solely via ORM queries. This is somewhat similar to the native Django behavior of defining model-based views (such as /list/[MODEL_NAME].

┆Issue is synchronized with this Jira Story

Newsletter: Create GET “/newsletter/“ endpoint to list all send newsletters

┆Issue is synchronized with this Jira Story

Lynx: Replace all hyperlinks with URL targets, not display names

Remove “/posts/backup” endpoint

┆Issue is synchronized with this Jira Task

Rename API project to be agnostic of client (no longer JAMStack-specific)

┆Issue is synchronized with this Jira Task

Analytics: Create data models for analytics fetched from Plausible & persist via ORM

• Refactor & cleanup
  • Handle “weekly” and “monthly” analytics as separate endpoints
    • /analytics/weekly
    • /analytics/monthly
    • Point Ghost webhooks to both endpoints
  • Ensure “null” results are never included
• Define new models & schemas
  • Create FastAPI Pydantic model representing each PageInsight
  • Create SQLAlchemy model representing each PageInsight
  • Modify SQL schema to handle additional insights (bounce rate, visitors, avg duration, etc.)
• Persist
  • Store results via DB session after fetching (handle in same endpoint)

┆Issue is synchronized with this Jira Story