migration

Author: guokan-shang

.envdefault

@@ -1,4 +1,13 @@
+# APPLICATION PARAMETERS
 APP_LANG=fr en
 ASSETS_PATH_ON_HOST=./assets
 ASSETS_PATH_IN_CONTAINER=/app/assets
-WORKER_NUMBER=1
+
+# SERVING PARAMETERS
+SERVICE_MODE=http
+CONCURRENCY=1
+
+# MICRO-SERVICE PARAMETERS
+SERVICE_NAME=extsumm
+SERVICES_BROKER=redis://172.17.0.1:6379
+BROKER_PASS=

Dockerfile

@@ -1,16 +1,20 @@
 FROM lintoai/linto-platform-nlp-core:latest
 LABEL maintainer="gshang@linagora.com"
 
-WORKDIR /app
+WORKDIR /usr/src/app
 
-COPY ./requirements.txt /app/
+COPY requirements.txt ./
 RUN pip install --no-cache-dir -r requirements.txt
 
-COPY .envdefault /app/
-COPY ./scripts /app/scripts
-COPY ./components /app/components
+COPY extsumm /usr/src/app/extsumm
+COPY components /usr/src/app/components
+COPY celery_app /usr/src/app/celery_app
+COPY http_server /usr/src/app/http_server
+COPY document /usr/src/app/document
+COPY docker-entrypoint.sh wait-for-it.sh healthcheck.sh ./
 
-HEALTHCHECK --interval=15s CMD curl -fs http://0.0.0.0/health || exit 1
+ENV PYTHONPATH="${PYTHONPATH}:/usr/src/app/extsumm"
 
-ENTRYPOINT ["/opt/conda/bin/gunicorn", "scripts.main:app", "--worker-class", "uvicorn.workers.UvicornWorker", "--bind", "0.0.0.0:80", "--access-logfile", "-", "--error-logfile", "-"]
-CMD ["--workers", "1"]
+HEALTHCHECK CMD ./healthcheck.sh
+
+ENTRYPOINT ["./docker-entrypoint.sh"]

LICENSE

@@ -1,3 +1,6 @@
+# 0.2.0
+- Migration to the [template]((https://github.com/linto-ai/linto-template-microservice)) of LinTO microservices.
+
 # 0.1.0
 - Initial commit.
 - Extractive Summarization.

RELEASE.md

@@ -0,0 +1,28 @@
+import os
+from celery import Celery
+
+from extsumm import logger
+
+celery = Celery(__name__, include=['celery_app.tasks'])
+service_name = os.environ.get("SERVICE_NAME", "extsumm")
+broker_url = os.environ.get("SERVICES_BROKER")
+if os.environ.get("BROKER_PASS", False):
+ components = broker_url.split('//')
+ broker_url = f'{components[0]}//:{os.environ.get("BROKER_PASS")}@{components[1]}'
+celery.conf.broker_url = "{}/0".format(broker_url)
+celery.conf.result_backend = "{}/1".format(broker_url)
+celery.conf.update(
+ result_expires=3600,
+ task_acks_late=True,
+ task_track_started = True)
+
+# Queues
+celery.conf.update(
+ {'task_routes': {
+ 'extsumm_task' : {'queue': 'extsumm'},}
+ }
+)
+
+logger.info(
+ f"Celery configured for broker located at {broker_url} with service name {service_name}"
+)

scripts/__init__.py renamed to celery_app/__init__.py

@@ -0,0 +1,36 @@
+import spacy
+import components
+
+from typing import Dict, List
+
+from celery_app.celeryapp import celery
+
+from extsumm import logger
+from extsumm.processing import LM_MAP, MODELS, get_model
+from extsumm.processing.utils import get_data
+
+
+@celery.task(name="extsumm_task")
+def extsumm_task(lang: str, texts: List[str], component_cfg: Dict = {}):
+ """Process a batch of articles and return the extractive summary predicted by the
+ given model. Each record in the data should have a key "text".
+ """
+ logger.info('ExtSumm task received')
+
+ # Check language availability
+ if lang in LM_MAP.keys():
+ model_name = LM_MAP[lang]
+ if model_name not in MODELS.keys():
+ raise RuntimeError(f"Model {model_name} for language {lang} is not loaded.")
+ nlp = spacy.blank(lang)
+ nlp.add_pipe("sentencizer", config={"punct_chars": ['|']})
+ nlp.add_pipe("extsumm", config={"model": {"@misc": "get_model", "name": model_name}})
+ else:
+ raise ValueError(f"Language {lang} is not supported.")
+
+ response_body = []
+ 
+ for doc in nlp.pipe(texts, component_cfg=component_cfg):
+ response_body.append(get_data(doc))
+
+ return {"extsumm": response_body}

celery_app/celeryapp.py

@@ -9,5 +9,4 @@ services:
  - .env
  volumes:
  - $ASSETS_PATH_ON_HOST:$ASSETS_PATH_IN_CONTAINER:ro
- command: ["--workers", $WORKER_NUMBER]
- runtime: nvidia
+ #runtime: nvidia

celery_app/tasks.py

@@ -0,0 +1,32 @@
+#!/bin/bash
+set -ea
+
+echo "RUNNING SERVICE"
+
+# Launch parameters, environement variables and dependencies check
+if [ -z "$SERVICE_MODE" ]
+then
+ echo "ERROR: Must specify a serving mode: [ http | task ]"
+ exit -1
+else
+ if [ "$SERVICE_MODE" = "http" ] 
+ then
+ echo "RUNNING HTTP SERVER"
+ python http_server/ingress.py --debug
+ elif [ "$SERVICE_MODE" == "task" ]
+ then
+ if [[ -z "$SERVICES_BROKER" ]]
+ then 
+ echo "ERROR: SERVICES_BROKER variable not specified, cannot start celery worker."
+ return -1
+ fi
+ /usr/src/app/wait-for-it.sh $(echo $SERVICES_BROKER | cut -d'/' -f 3) --timeout=20 --strict -- echo " $SERVICES_BROKER (Service Broker) is up"
+ echo "RUNNING CELERY WORKER"
+ celery --app=celery_app.celeryapp worker -Ofair -n nlp_${SERVICE_NAME}_worker@%h --queues=${SERVICE_NAME} -c ${CONCURRENCY}
+ else
+ echo "ERROR: Wrong serving command: $1"
+ exit -1
+ fi
+fi
+
+echo "Service stopped"