Merge branch 'master' into main

.gitignore

@@ -0,0 +1,3 @@
+venv
+.idea
+__pycache__

Dockerfile

@@ -0,0 +1,43 @@
+# STEP 1: build pdfact
+FROM ubuntu:20.04
+
+ENV DEBIAN_FRONTEND=noninteractive
+RUN apt-get update -y && apt-get upgrade -y && apt-get install -y maven git
+
+WORKDIR /
+# TODO freeze revision?
+RUN git clone https://github.com/ad-freiburg/pdfact.git
+WORKDIR pdfact
+
+RUN mvn install -DskipTests
+
+# STEP 2: run service
+FROM alpine:latest
+WORKDIR /
+
+# copy build artifact
+COPY --from=0 /pdfact/bin/pdfact.jar pdfact.jar
+
+# get java
+RUN apk update \
+    && apk upgrade \
+    && apk add --no-cache openjdk11-jre
+ENV JAVA_HOME="/usr/lib/jvm/java-11-openjdk"
+
+# get python
+RUN apk add --no-cache python3 \
+    && python3 -m ensurepip \
+    && pip3 install --upgrade pip setuptools
+
+# set environment variables
+ENV PYTHONDONTWRITEBYTECODE 1
+ENV PYTHONUNBUFFERED 1
+
+# install flask app
+WORKDIR /app
+COPY ./requirements.txt .
+RUN pip install -r requirements.txt
+
+COPY app.py gunicorn_conf.py ./
+
+ENTRYPOINT ["gunicorn", "--conf", "gunicorn_conf.py", "--bind", "0.0.0.0:80", "app:app"]

app.py

@@ -0,0 +1,56 @@
+from flask import Flask, request, flash, Response
+from werkzeug.utils import secure_filename
+import logging
+from os import path, remove
+import tempfile
+import subprocess as sp
+
+app = Flask(__name__)
+
+
+def get_format(req, fallback=("application/json", "json")):
+    accept = req.headers.get("accept").lower()
+    if accept == "application/json":
+        return accept, "json"
+    elif accept == "application/xml":
+        return accept, "xml"
+    elif accept == "text/plain":
+        return accept, "txt"
+    else:
+        return fallback
+
+
+@app.route('/analyze', methods=["POST"])
+def analyze():
+    if "file" not in request.files or len(request.files["file"].filename) == 0:
+        flash("No file.")
+        return
+
+    # save file to tmp folder
+    file = request.files["file"]
+    file_path = path.join(tempfile.gettempdir(), secure_filename(file.filename))
+    app.logger.debug(f"saving file as {file_path}")
+    file.save(file_path)
+
+
+
+    mime, response_format = get_format(request)
+    cmd = ["java", "-jar", "/pdfact.jar", file_path, "--format", response_format]
+    app.logger.debug(f"running {' '.join(cmd)}")
+
+    try:
+        result = sp.check_output(cmd)
+        return Response(result, mimetype=mime)
+    except sp.CalledProcessError:
+        return "Something went wrong", 500
+    finally:
+        remove(file_path)
+
+if __name__ == '__main__':
+    app.run()
+else:
+    # use gunicorns loglevel
+    gunicorn_logger = logging.getLogger("gunicorn.error")
+    app.logger.handlers = gunicorn_logger.handlers
+    app.logger.setLevel(gunicorn_logger.level)
+    app.logger.warn(gunicorn_logger.level)

gunicorn_conf.py

@@ -0,0 +1,8 @@
+loglevel = "info"
+errorlog = "-"  # stderr
+accesslog = "-"  # stdout
+worker_tmp_dir = "/dev/shm"
+graceful_timeout = 120
+timeout = 120
+keepalive = 5
+threads = 4

requirements.txt

@@ -0,0 +1,2 @@
+Flask==2.2.2
+gunicorn==20.1.0