From d50b738df3a497eb3e50cc5d88afb198bc1451f9 Mon Sep 17 00:00:00 2001
From: Anil Vishnoi <vishnoianil@gmail.com>
Date: Fri, 6 Sep 2024 03:26:44 -0700
Subject: [PATCH] POC: Frontend docling library with fastapi and build the
 container image to deploy as a service

Signed-off-by: Anil Vishnoi <vishnoianil@gmail.com>
---
 Makefile                       |  5 +++++
 duckbill/Containerfile         | 34 ++++++++++++++++++++++++++++++++++
 duckbill/apiserver/__init__.py |  0
 duckbill/apiserver/main.py     | 24 ++++++++++++++++++++++++
 duckbill/requirements.txt      |  3 +++
 5 files changed, 66 insertions(+)
 create mode 100644 duckbill/Containerfile
 create mode 100644 duckbill/apiserver/__init__.py
 create mode 100644 duckbill/apiserver/main.py
 create mode 100644 duckbill/requirements.txt

diff --git a/Makefile b/Makefile
index 841d7919..e69f219d 100644
--- a/Makefile
+++ b/Makefile
@@ -47,6 +47,11 @@ ps-image: Containerfile.ps ## Build continaer image for the pathservice
 	$(CMD_PREFIX) docker build -f Containerfile.ps -t ghcr.io/instructlab/ui/pathservice:$(TAG) .
 	$(CMD_PREFIX) docker tag ghcr.io/instructlab/ui/pathservice:$(TAG) ghcr.io/instructlab/ui/pathservice:main
 
+duckbill-image: duckbill/Containerfile ## Build continaer image for the pathservice
+	$(ECHO_PREFIX) printf "  %-12s duckbill/Containerfile\n" "[docker]"
+	$(CMD_PREFIX) docker build -f duckbill/Containerfile --platform linux/amd64 -t ghcr.io/instructlab/ui/duckbill:$(TAG) ./duckbill
+	$(CMD_PREFIX) docker tag ghcr.io/instructlab/ui/duckbill:$(TAG) quay.io/instructlab-ui/docling:main
+
 ##@ Local Dev - Run the stack (UI and PathService) on your local machine
 .PHONY: stop-dev-local
 stop-dev-local:  ## Stop the npm and pathservice local instances
diff --git a/duckbill/Containerfile b/duckbill/Containerfile
new file mode 100644
index 00000000..281b7ba0
--- /dev/null
+++ b/duckbill/Containerfile
@@ -0,0 +1,34 @@
+FROM python:3.11-slim-bookworm
+
+WORKDIR /duckbill
+COPY ./requirements.txt /duckbill/requirements.txt
+
+RUN pip install --upgrade pip
+RUN pip install --no-cache-dir --upgrade -r /duckbill/requirements.txt
+
+ENV GIT_SSH_COMMAND="ssh -o StrictHostKeyChecking=no"
+
+RUN apt-get update \
+    && apt-get install -y libgl1 libglib2.0-0 curl wget git \
+    && apt-get clean
+
+# This will install torch with *only* cpu support
+# Remove the --extra-index-url part if you want to install all the gpu requirements
+# For more details in the different torch distribution visit https://pytorch.org/.
+#RUN pip install --no-cache-dir docling --extra-index-url https://download.pytorch.org/whl/cpu
+#RUN pip install --no-cache-dir docling fastapi
+
+ENV HF_HOME=/tmp/
+ENV TORCH_HOME=/tmp/
+
+RUN python -c 'from deepsearch_glm.utils.load_pretrained_models import load_pretrained_nlp_models; load_pretrained_nlp_models(verbose=True);'
+RUN python -c 'from docling.document_converter import DocumentConverter; artifacts_path = DocumentConverter.download_models_hf(force=True);'
+
+# On container environments, always set a thread budget to avoid undesired thread congestion.
+ENV OMP_NUM_THREADS=4
+
+COPY ./apiserver /duckbill/apiserver
+
+EXPOSE 5000
+
+CMD ["fastapi", "run", "apiserver/main.py", "--port", "5000"]
\ No newline at end of file
diff --git a/duckbill/apiserver/__init__.py b/duckbill/apiserver/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/duckbill/apiserver/main.py b/duckbill/apiserver/main.py
new file mode 100644
index 00000000..f388ee24
--- /dev/null
+++ b/duckbill/apiserver/main.py
@@ -0,0 +1,24 @@
+from typing import Union
+
+from fastapi import FastAPI
+from pydantic import BaseModel
+from docling.document_converter import DocumentConverter
+
+
+app = FastAPI()
+
+class UrlRequest(BaseModel):
+    url: str
+
+@app.get("/")
+def read_root():
+    return {"Docling Service is up and running."}
+
+
+@app.post("/simpleconvert")
+def simpleconvert(request: UrlRequest):
+    url = request.url
+    converter = DocumentConverter()
+    doc = converter.convert_single(url)
+    print(doc.render_as_markdown())
+    return {doc.render_as_markdown()}
\ No newline at end of file
diff --git a/duckbill/requirements.txt b/duckbill/requirements.txt
new file mode 100644
index 00000000..fec5b3ff
--- /dev/null
+++ b/duckbill/requirements.txt
@@ -0,0 +1,3 @@
+fastapi[standard]>=0.113.0,<0.114.0
+pydantic>=2.7.0,<3.0.0
+docling>=1.9.0