Add dockerfile for backend, configure docker-compose, fix models (#11)

.gitignore

@@ -160,3 +160,6 @@ cython_debug/
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
 .DS_Store
+
+# Local env file
+.local.env

backend/Dockerfile

@@ -0,0 +1,45 @@
+# All directory paths for COPY commands are relative to the build context
+
+# Ensure this python version stays in sync with CI
+FROM python:3.11-slim as base
+WORKDIR /backend
+
+# set environment variables
+ENV PYTHONDONTWRITEBYTECODE 1
+ENV PYTHONUNBUFFERED 1
+ENV POETRY_HOME="/opt/poetry"
+ENV MYPYPATH="/app/src/stubs"
+
+# Use bash as the shell for the build
+# https://github.com/docker/for-linux/issues/408#issuecomment-414748815
+SHELL ["/bin/bash", "-o", "pipefail", "-c"]
+
+RUN set -eux && \
+    apt-get update && \
+    apt-get install -y \
+      build-essential \
+      # to install poetry
+      curl \
+      # to install psycopg2
+      libpq-dev python3-dev
+
+# https://python-poetry.org/docs/master/#installing-with-the-official-installer
+RUN curl -sSL https://install.python-poetry.org | python -
+ENV PATH="$POETRY_HOME/bin:$PATH"
+
+# install deps before copying project files so the cache is only invalidated
+# when the deps change
+COPY ./backend/pyproject.toml ./backend/poetry.lock .
+RUN poetry config virtualenvs.create false
+RUN poetry install --no-root --only main
+
+COPY ./backend .
+
+EXPOSE 8000
+
+###
+# development image
+###
+FROM base as development
+
+ENTRYPOINT ["bash", "./scripts/local_entry_point.sh"]

backend/db/models.py

@@ -66,16 +66,6 @@ class Extractor(TimestampedModel):
         server_default="",
         comment="The name of the extractor.",
     )
-    created_at = Column(
-        DateTime(timezone=True),
-        server_default=func.now(),
-        comment="Time when this extracted was originally created.",
-    )
-    modified_at = Column(
-        DateTime(timezone=True),
-        onupdate=func.now(),
-        comment="Last time this was modified.",
-    )
     schema = Column(
         JSONB,
         nullable=False,

backend/scripts/local_entry_point.sh

@@ -0,0 +1,10 @@
+# -e: fail on any nonzero exit status
+# -u: fail if any referenced variables are not set
+# -x: print commands before running them
+# -o pipefail: fail if a command in a pipe has a nonzero exit code
+set -euxo pipefail
+
+# For now just create the db if it doesn't exist
+python -m scripts.run_migrations create
+
+uvicorn server.main:app --host 0.0.0.0 --port 8000 --reload

backend/server/settings.py

@@ -1,5 +1,7 @@
 from __future__ import annotations
 
+import os
+
 from langchain_openai import ChatOpenAI
 from sqlalchemy.engine import URL
 
@@ -9,12 +11,12 @@ CHUNK_SIZE = int(4_096 * 0.8)
 MAX_CONCURRENCY = 1
 
 
-def get_postgres_url():
+def get_postgres_url() -> URL:
     url = URL.create(
         drivername="postgresql",
         username="langchain",
         password="langchain",
-        host="localhost",
+        host=os.environ.get("PG_HOST", "localhost"),
         database="langchain",
         port=5432,
     )

docker-compose.yml

@@ -1,4 +1,4 @@
-version: "3"
+version: "3.8"
 name: langchain-extract
 
 services:
@@ -7,24 +7,37 @@ services:
     # Make sure to keep in sync with CI 
     # version if being tested on CI.
     image: postgres:16
+    expose:
+      - "5432"
     ports:
       - "5432:5432"
     environment:
       POSTGRES_DB: langchain
       POSTGRES_USER: langchain
       POSTGRES_PASSWORD: langchain
+    healthcheck:
+        test: [ "CMD-SHELL", "pg_isready -U langchain -d langchain -W langchain" ]
+        interval: 10s
+        timeout: 5s
+        retries: 5
     volumes:
       - postgres_data:/var/lib/postgresql/data
 
-# For rely on docker compose to spin up postgres
-# but developer using docker
-# Add backend when we actually need it
-#  backend:
-#    build: ./backend
-#    ports:
-#      - "8000:8000"
-#    depends_on:
-#      - postgres
+  backend:
+    build:
+      context: .
+      dockerfile: ./backend/Dockerfile
+      target: development
+    env_file:
+      - .local.env
+    environment:
+      - PG_HOST=postgres
+    ports:
+      - "8000:8000" # Backend is accessible on localhost:8100
+    depends_on:
+      - postgres
+    volumes:
+      - ./backend:/backend
 
 volumes:
   postgres_data: