arXiv update bot

arxiv update bot is a simple python script that scraps the arXiv, search for interesting paper and send a message on telegram if any was found.

Usage

The package comes with a command line script arxiv-update-bot. Here the help message :

usage: arxiv-update-bot [-h] [-c CONFIG_PATH] [-q]

Scrap the arXiv

optional arguments:
  -h, --help            show this help message and exit
  -c CONFIG_PATH, --config-path CONFIG_PATH
                        Path for configuration path. Replace default of
                        /etc/arxiv-update-bot/config.ini
  -q, --quiet           If quiet is set, then the bot will not send message if
                        no article are found.

Installation

The package can be installed either via the pypi repository :

or using the sources :

Configuration file

In order to work, the script needs a configuration file. It will by default search for the configuration file in /etc/arxiv-update-bot/config.ini. Note that you have to manually create the folder and give the good permissions.

You can override the default behavior with the -c option on the command line and by giving the path as argument.

An example configuration can be found at arxiv_update_bot/config.example.ini and is also included in the package. Here is the example:

[bot]
token = 0000000000:AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

[quant-ph]
chat_id = 0
category = quant-ph
buzzwords = cvqkd,continuous variable,continuous-variable,qkd,quantum key distribution,rfsoc,fpga

The [bot] section is here to parametrize the bot. It must have the token value (with the "bot" word).

Then for each update, you need to define a section. The name of the section is not really important (it must be unique and not "bot").

• The chat_id corresponds to the id of the individual or group where the notification must be sent. For now you can only configure 1 recipient per update.
• The category is the name of the arxiv category. It will be used to determinate which RSS flux will be scraped.
• The buzzwords are a list of words, separated by comas (without spaces) and in lowercase. The articles that will be keeped will be the ones with one of the buzwwords in the title.

Cron configuration

It is advised to use a cron to execute the script periodically :

0 10 * * * arxiv-update-bot

to run the script every day at 10 am.

How it works

For each update, the script get the RSS flux, goes through the article and try to match the articles titles with the buzzwords. If there is match, a notification is sent.