add openEuler support for llm-docsum

Signed-off-by: zhihang <[email protected]>

Author: zhihangdeng

.github/workflows/docker/compose/llms-compose.yaml

@@ -23,6 +23,10 @@ services:
     build:
       dockerfile: comps/llms/src/doc-summarization/Dockerfile
     image: ${REGISTRY:-opea}/llm-docsum:${TAG:-latest}
+  llm-docsum-openeuler:
+    build:
+      dockerfile: comps/llms/src/doc-summarization/Dockerfile.openEuler
+    image: ${REGISTRY:-opea}/llm-docsum:${TAG:-latest}-openeuler
   llm-faqgen:
     build:
       dockerfile: comps/llms/src/faq-generation/Dockerfile

comps/llms/src/doc-summarization/Dockerfile.openEuler

@@ -0,0 +1,44 @@
+# Copyright (C) 2025 Huawei Technologies Co., Ltd. 
+# SPDX-License-Identifier: Apache-2.0
+
+FROM openeuler/python:3.11.13-oe2403lts
+
+ARG ARCH="cpu"
+
+RUN yum update -y && \
+    yum install -y \
+    shadow \
+    jemalloc-devel && \
+    yum clean all && rm -rf /var/cache/yum
+
+RUN useradd -m -s /bin/bash user && \
+    mkdir -p /home/user && \
+    chown -R user /home/user/
+
+COPY comps /home/user/comps
+
+ARG uvpip='uv pip install --system --no-cache-dir'
+RUN pip install --no-cache-dir --upgrade pip setuptools uv && \
+    if [ ${ARCH} = "cpu" ]; then \
+      $uvpip torch torchvision --index-url https://download.pytorch.org/whl/cpu; \
+      $uvpip -r /home/user/comps/llms/src/doc-summarization/requirements-cpu.txt; \
+    else \
+      $uvpip -r /home/user/comps/llms/src/doc-summarization/requirements-gpu.txt; \
+    fi
+
+ENV PYTHONPATH=$PYTHONPATH:/home/user
+
+# air gapped support: set model cache dir
+RUN mkdir -p /data && chown -R user /data
+ENV HF_HUB_CACHE=/data
+# air gapped support: pre-download tiktoken bpe files
+RUN mkdir -p /opt/tiktoken_cache
+ENV TIKTOKEN_CACHE_DIR=/opt/tiktoken_cache
+RUN python -c 'import tiktoken; tiktoken.get_encoding("cl100k_base");tiktoken.get_encoding("o200k_base");tiktoken.get_encoding("gpt2");tiktoken.get_encoding("r50k_base");tiktoken.get_encoding("p50k_base");tiktoken.get_encoding("p50k_edit");'
+RUN chown -R user /opt/tiktoken_cache
+
+USER user
+
+WORKDIR /home/user/comps/llms/src/doc-summarization
+
+ENTRYPOINT ["python", "opea_docsum_microservice.py"]

tests/llms/test_llms_doc-summarization_tgi.sh

@@ -21,7 +21,8 @@ service_name="docsum-tgi"
 
 function build_docker_images() {
     cd $WORKPATH
-    docker build --no-cache -t ${REGISTRY:-opea}/llm-docsum:${TAG:-latest} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/llms/src/doc-summarization/Dockerfile .
+    dockerfile_name="comps/llms/src/doc-summarization/$1"
+    docker build --no-cache -t ${REGISTRY:-opea}/llm-docsum:${TAG:-latest} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f "${dockerfile_name}" .
     if [ $? -ne 0 ]; then
         echo "opea/llm-docsum built fail"
         exit 1
@@ -148,29 +149,43 @@ function validate_microservices() {
         '{"messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5.", "max_tokens":32, "language":"en", "summary_type": "refine", "chunk_size": 2000, "timeout":200}'
 }
 
-function stop_docker() {
+function stop_service() {
     cd $WORKPATH/comps/llms/deployment/docker_compose
     docker compose -f compose_doc-summarization.yaml down --remove-orphans
 }
 
 function main() {
 
-    stop_docker
+    stop_service
 
-    build_docker_images
+    build_docker_images "Dockerfile"
 
-    trap stop_docker EXIT
+    trap stop_service EXIT
 
     echo "Test normal env ..."
     start_service
     validate_microservices
-    stop_docker
+    stop_service
 
     if [[ -n "${DATA_PATH}" ]]; then
         echo "Test air gapped env ..."
         start_service true
         validate_microservices
-        stop_docker
+        stop_service
+    fi
+
+    build_docker_images "Dockerfile.openEuler"
+
+    echo "Test with openEuler OS ..."
+    start_service
+    validate_microservices
+    stop_service
+
+    if [[ -n "${DATA_PATH}" ]]; then
+        echo "Test air gapped env ..."
+        start_service true
+        validate_microservices
+        stop_service
     fi
 
     echo y | docker system prune

tests/llms/test_llms_doc-summarization_vllm.sh

@@ -32,7 +32,8 @@ function build_docker_images() {
     fi
 
     cd $WORKPATH
-    docker build --no-cache -t ${REGISTRY:-opea}/llm-docsum:${TAG:-latest} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/llms/src/doc-summarization/Dockerfile .
+    dockerfile_name="comps/llms/src/doc-summarization/$1"
+    docker build --no-cache -t ${REGISTRY:-opea}/llm-docsum:${TAG:-latest} --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f "${dockerfile_name}" .
     if [ $? -ne 0 ]; then
         echo "opea/llm-docsum built fail"
         exit 1
@@ -159,30 +160,45 @@ function validate_microservices() {
         '{"messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5.", "max_tokens":32, "language":"en", "summary_type": "refine", "chunk_size": 2000, "timeout":200}'
 }
 
-function stop_docker() {
+function stop_service() {
     cd $WORKPATH/comps/llms/deployment/docker_compose
     docker compose -f compose_doc-summarization.yaml down --remove-orphans
 }
 
 function main() {
 
-    stop_docker
+    stop_service
 
-    build_docker_images
+    build_docker_images "Dockerfile"
 
-    trap stop_docker EXIT
+    trap stop_service EXIT
 
     echo "Test normal env ..."
     start_service
     validate_microservices
-    stop_docker
+    stop_service
 
     if [[ -n "${DATA_PATH}" ]]; then
         echo "Test air gapped env ..."
         start_service true
         validate_microservices
-        stop_docker
+        stop_service
     fi
+
+    build_docker_images "Dockerfile.openEuler"
+
+    echo "Test with openEuler OS ..."
+    start_service
+    validate_microservices
+    stop_service
+
+    if [[ -n "${DATA_PATH}" ]]; then
+        echo "Test air gapped env ..."
+        start_service true
+        validate_microservices
+        stop_service
+    fi
+
     echo y | docker system prune
 
 }