suica / arxiver

arXivER: Just a Flask-based paper reading assistant!

arXivER

arXivER is a paper reading assistant using Flask and Vue.js, which aims at enhancing the experience when using arXiv.

Install

There are two ways to run arXivER, you can run from source code or choose the way from docker.

Run from source code

It is recommended that you create virtual environment and activate it first.

If you use anaconda, then you could use this.

conda create -n arxiver python=3.6

conda activate arxiver

If you don't, you can use your own virtual environment instead, it does not matter.

Now you should install all dependencies, use

pip install -r requirements.txt

Run Main Server

Then go to the project folder.

python app.py

to start the server.

Now, arXivER is running on http://0.0.0.0:5000.

Run Celery Worker

cd async_tasks

Change db_address in fetch_paper.py into an absolute address on your computer.

Then run

sh www.sh

to execute the celery worker.

Run from docker

The Dockerfile executes install.sh to initial database and install python3.6.

For convinience, we build image included both sqlite and python environment instead of using docker-compose.

You can check details in install.sh and Dockerfile

How to run

• build image docker build -t arxiver:0.1
• run image docker run -p 5000:5000 -t arxiver:0.1

Then arXiver is running on http://127.0.0.1:5000

Test

First, run the server first. And run cd tests.

Then use python tests.py to run the unit tests.

Deploy

Will be added soon.

Screen shots