New URL Sources & New DB integration

This commit is contained in:
MAHMUT YILMAZ
2021-11-26 14:42:50 +03:00
parent 055d6307f3
commit deab989d4a
2 changed files with 31 additions and 0 deletions

6
.dockerignore Normal file
View File

@@ -0,0 +1,6 @@
.git
.idea
__pycahce__/
venv/
custom/__pycahce__/
generic/__pycahce__/

25
Dockerfile-crawler Normal file
View File

@@ -0,0 +1,25 @@
FROM artifactory.turkcell.com.tr/local-docker-dist-dev/com/turkcell/ai-platform/aip/ubi8/python38:0.1.0
LABEL maintainer="Mahmut YILMAZ <yilmaz.mahmut@turkcell.com.tr>"
ENV AIP_APP_USER=appuser
ENV APP_DIR_HOME=/home/appuser
ENV IN_CONTAINER=1
ENV TZ=Europe/Istanbul
USER root
COPY requirements-crawler.txt ${APP_DIR_HOME}
WORKDIR ${APP_DIR_HOME}
RUN pip3 install --no-cache-dir --trusted-host=pypi.python.org --trusted-host=pypi.org --trusted-host=files.pythonhosted.org -r ${APP_DIR_HOME}/requirements_crawler.txt
COPY . ${APP_DIR_HOME}
RUN \
chgrp -R 0 $APP_DIR_HOME && \
chmod -R g+rx $APP_DIR_HOME
RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
CMD [ "python3", "-u", "./news_crawler.py" ]