NewsPipe
This repository contains the complete pipeline for collecting online newspaper article. The articles are stored in a MongoDB. The whole pipeline is dockerized, thus the user does not need to worry about dependencies. Additionally, docker-compose is available to increase the useability for the user.
Requirement
To use this system, you need to create a .env file in which the MongoDB information is available:
MONGO_ROOT_USER=devroot
MONGO_ROOT_PASSWORD=devroot
MONGOEXPRESS_LOGIN=dev
MONGOEXPRESS_PASSWORD=dev
MONGO_CHART_USERNAME=dev
MONGO_CHART_PASSWORD=dev
POSTGRES_USER=airflow
POSTGRES_PASS=airflow
If you want to specify the number of threads then open airflow-newspipe-docker and adjust the sed command in airflow-docker/Dockerfile. If you want 4 threads per process:
&& sed -i'.orig' 's/max_threads = 2/max_threads = 4/g' ${AIRFLOW_HOME}/airflow.cfg \
Additionally, you can also specify the number of processes (2 processes in this case):
&& sed -i'.orig' 's/parallelism = 32/parallelism = 2/g' ${AIRFLOW_HOME}/airflow.cfg \
Getting Started
To start this application, run:
docker-compose up
- To see the database collections, mongo-express is in use and available on
localhost:8081. The MongoDB itself is available on port27017. - The airflow application should be available on
localhost:8083. You will see the airflow dashboard with the default examples. - For the mongo chart dashboard, open
localhost
Adding article sources
Each crawler is defined as DAG in 'dag'. To add a data source, you must therefore add DAGs in the dags folder. A DAG is a Python script that contains the settings for an entire crawling pipeline. Use the default example as a template. The DAGs are very simple and straightforward.
import os
import datetime
from dag_factory import create_dag
url = "taz.de" # url of newspaper source
# Defining the crawling intervals
airflow_config = {'schedule_interval': '@hourly', # set a interval, for continuous crawling
'start_date': datetime.datetime(2020, 6, 4, 21), # set a date, on which the dag will run
'end_date':datetime.datetime(2020, 6, 5, 6), # optinal, set if it is needed
}
# Create crawling DAG
DAG = create_dag(url=url,
airflow_config=airflow_config,
name=os.path.basename(__file__))
Options for schedule_interval:
| preset | meaning | cron |
|---|---|---|
@once |
Schedule once and only once | |
@hourly |
Run once an hour at the beginning of the hour | 0 * * * * |
@daily |
Run once a day at midnight | 0 0 * * * |
@weekly |
Run once a week at midnight on Sunday morning | 0 0 * * 0 |
@monthly |
Run once a month at midnight of the first day of the month | 0 0 1 * * |
@quarterly |
Run once a quarter at midnight on the first day | 0 0 1 */3 * |
@yearly |
Run once a year at midnight of January 1 | 0 0 1 1 * |
Mongo Charts
MongoDB Charts is a data visualization tool that is integrated within the MongoDB ecosystem. By default, there are no visualization available or shipped with NewsPipe. Therefore, you have to create dashboard on your needs. This involves following 3 steps:
- Setup data source
- Data aggregation
- Dashboard creation which are well documented on docs.mongodb.com.
Credentials:
The credentials for mongo charts are:
- E-Mail: [email protected]
- Password: MONGO_CHART_PASSWORD
Connection URI
- URI:
mongodb://MONGO_ROOT_USER:[email protected]:27017
12 Oct 22, 2022
1 Mar 28, 2022
7 Oct 23, 2022
1 Dec 7, 2021
1 Jan 3, 2022
273 Dec 31, 2022
20 Jan 6, 2023
1 Nov 6, 2021
1.5k Jan 4, 2023
11.9k Dec 30, 2022
45.5k Jan 7, 2023
2k Jan 5, 2023
677 Jan 7, 2023
1 Nov 3, 2021
1 Dec 26, 2021
4 Aug 30, 2022
3 Aug 12, 2022
15 Jul 28, 2021
4 Jan 09, 2022
1 Nov 07, 2021
3 Feb 27, 2022
3 Feb 17, 2022
2 Jun 06, 2022
11 Nov 19, 2022
5 Nov 29, 2022
0 Jan 07, 2022
15 Dec 05, 2021
1 Dec 20, 2021
692 Dec 22, 2022
3 Jan 03, 2022
3 Feb 13, 2022
1 Aug 27, 2022
0 Nov 22, 2021
87 Jan 06, 2023
1.1k Jan 06, 2023
90 Feb 09, 2021
1 Oct 29, 2021
6 Nov 21, 2022