Merge pull request #44 from SAFEHR-data/paul/refactor-forwarding-api

Refactor forwarding API

Author: p-j-smith

docker-compose-postgres.yml

@@ -185,7 +185,7 @@ services:
 
   ocr-forwarding-api:
     build:
-      context: src/api
+      context: packages/api
       dockerfile: Dockerfile
       args:
         <<: *build-args-common
@@ -199,7 +199,6 @@ services:
     ports:
       - "${OCR_FORWARDING_API_PORT}:${OCR_FORWARDING_API_PORT}"
     volumes:
-      - ./src/api/app:/app
       - ${PWD}/${DATA_FOLDER}:/data
     networks:
       - pyonb_ocr_api

docker-compose.yml

@@ -0,0 +1,19 @@
+FROM ghcr.io/astral-sh/uv:python3.13-bookworm AS app
+
+SHELL ["/bin/bash", "-o", "pipefail", "-e", "-u", "-x", "-c"]
+
+WORKDIR /app
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+
+COPY ./pyproject.toml .
+COPY ./README.md .
+COPY ./src src/
+
+RUN uv venv
+RUN --mount=type=cache,target=/root/.cache/uv,sharing=locked uv sync --no-editable --no-dev
+
+# make uvicorn etc available
+ENV PATH="/app/.venv/bin:$PATH"
+
+CMD uvicorn pyonb_api.main:app --host 0.0.0.0 --port "$OCR_FORWARDING_API_PORT" --workers 4 --use-colors

packages/api/Dockerfile

@@ -0,0 +1,25 @@
+# `pyonb` forwarding API
+
+The forwarding API is used to send documents to an OCR service for processing. It provides a consistent
+interface for using the various OCR tools supported by `pyonb`.
+
+## Usage
+
+You will need to define the `OCR_FORWARDING_API_PORT` in a `.env` file.
+
+Then, spin up the `ocr-forwarding-api` servicer, along with the OCR service you wish to use.
+For example, if you would like to use `kreuzberg`, run the following from the top-level `pyonb` directory:
+
+```shell
+docker-compose --profile kreuzberg up --build --detach
+```
+
+You can then use `curl` to send a PDF to the forwarding API:
+
+```shell
+curl -v -X POST http://127.0.0.1:8110/kreuzberg-ocr/inference_single \
+  -F "file_upload=@document.pdf" \
+  -H "accept: application/json"
+```
+
+Note, this assumes you have set `OCR_FORWARDING_API_PORT` to `8110`.

packages/api/README.md

@@ -0,0 +1,16 @@
+[build-system]
+build-backend = "hatchling.build"
+requires = ["hatchling"]
+
+[project]
+dependencies = [
+    "aiohttp",
+    "fastapi[standard]",
+    "requests",
+    "uvicorn",
+]
+description = "pyonb forwarding API"
+name = "pyonb-api"
+readme = "README.md"
+requires-python = ">=3.11"
+version = "0.1.0"