martimors

repos:
  - repo: https://github.com/psf/black
    rev: 23.1.0
    hooks:
      - id: black
        args: [--config=./pyproject.toml]
  - repo: https://github.com/python-poetry/poetry
    rev: 1.3.0
    hooks:
      - id: poetry-check

martimors

---
apiVersion: v1
kind: Service
metadata:
  name: cockroachdb
spec:
  selector:
    app: cockroachdb
  type: ClusterIP
  ports:

martimors

---
apiVersion: v1
kind: Service
metadata:
  name: postgres
spec:
  selector:
    app: postgres
  ports:
  - port: 5432

martimors

# Builder container uses pdm to install dependencies
FROM python:3.10 AS builder

ENV PDM_VERSION=1.13.4

WORKDIR /build
COPY pdm.lock pyproject.toml ./
RUN python -m pip install --upgrade pip setuptools wheel
RUN pip install pdm && pdm install --prod --no-lock --no-editable

martimors

[tool.black]
line-length = 100
include = '\.pyi?$'
exclude = '''
/(
    \.git
  | \.mypy_cache
  | \.tox
  | \.venv
  | venv

martimors

[flake8]
# Specify a list of codes to ignore.
ignore =
    E203, # whitespace before ‘:’ (conflicts with black)
    E501, # line too long (black will fix this)
    W503, # line break before binary operator (conflicts with black)
    F821, # undefined name error, (bug in pyflakes)
max-line-length = 100
max-complexity = 12
# Specify the list of error codes you wish Flake8 to report.

martimors

# Apply to all files without committing:
#   pre-commit run --all-files
# Update this file:
#   pre-commit autoupdate
repos:

# format markdown files
-   repo: https://github.com/pre-commit/mirrors-prettier
    rev: "v2.6.1"
    hooks:

martimors

prefixes = ["", "SCHEMA_REGISTRY", "SCHEMA_REGISTRY_SCHEMA_REGISTRY"]
middle = ["CONFLUENT", "KAFKASTORE", "ACL", "TOPIC"]
suffix = ["GROUP_ID", "GROUPID", "CONSUMER_GROUP_ID", "CONSUMER_GROUPID", "GROUP"]


from itertools import permutations

middle_choices = []

for a in range(len(middle)+1):

martimors

# Significantly reduces the size of the repository
# The scripts can be converted back into notebooks, but the cache will be lost
# Excludes the directory ./venv and any directory with a . prefix such as .git
# Converts all jupyter notebooks to python scripts recursively starting from $(pwd)
find . -path ./venv -prune -false -o -name '*.ipynb' | xargs -0 jupyter nbconvert --to script

martimors

def datetime_to_string(df):
    """
    Convert all datetimes to strings in a dataframe.
    Assumes dates are all timezone aware.
    """
    datetime_columns = df.select_dtypes(include="datetimetz").columns.to_list()
    for col in datetime_columns:
        df[col] = df[col].dt.strftime(date_format="%Y-%m-%dT%H:%M:%S%z")
    return df