SlackReactionCounter

Tianpu Zhao* and André Vallières*

*Authors contributed equally

Brief description

Slack app for counting the number of reactions in a particular channel and sorting them based on a weighted average or total score. In our group, we use a channel to post relevant papers to our research. Group members then rate each paper for its relevancy, using reactions 1️⃣, 2️⃣, or 3️⃣ (corresponding to emojis :one:, :two: and :three:), with respective weights 1, 2, and 3. Then, to select a paper to present as part of our journal club, we look at previously posted papers and select from the most relevant ones.

Requiremenets

We haven't yet tested versions above/below the following specifications

beautifulsoup4==4.12.2
bs4==0.0.1
certifi==2022.12.7
charset-normalizer==3.1.0
idna==3.4
numpy==1.24.3
python==3.10
python-dateutil==2.8.2
python-dotenv==1.0.0
requests==2.30.0
six==1.16.0
slack-bolt==1.18.0
slack-sdk==3.21.3
soupsieve==2.4.1
tabulate==0.9.0
urllib3==2.0.2

How to install

Run

git clone https://github.com/ZhaoTianPu/SlackReactionCounter

in a desired directory to create a folder SlackReactionCounter with codes. Alternatively, download through GitHub and unzip.

Configuring a Slack bot in your Slack workspace

You need to create and configure a Slack bot in your Slack workspace first, before running the code.

1. Go to https://api.slack.com, after logging in, click Your Apps on top right corner.
2. Create the bot: Click Create New App and then click From scratch, then provide your app name and specify the workspace where the app will be installed.
3. After we have created the bot, we will be redirected to the bot’s configuration page. Under the Features section in the left panel, go to the OAuth & Permissions tab. Navigate down to Bot Token Scopes under Scopes and add the following scopes to grant necessary permissions for the bot:
   1. app_mentions:read
   2. channels:history
   3. channels:read
   4. channels:write
   5. chat:write
   6. commands
   7. reactions:read
4. Navigate down to User Token Scopes and add the following scopes:
   1. channels:history
   2. channels:read
   3. chat:write
   4. reactions:read
5. Enable slash commands: Under the Features section in the left panel, go to the Slash Commands tab. Create your own commands here. We use /papers in our group as the command.
6. Install the bot. Under the Settings section in the left panel, go to the Install App tab and install/reinstall to the workspace. Copy User OAuth Token and Bot User OAuth Token.
7. You will also need to get the ID of the Slack channel, from which threads will be ranked based on the reaction emojis. To obtain this, open your Slack workspace and click on the target channel, and the url reads: https://app.slack.com/client/ <workspace id>/<channel id>. Copy the <channel id>.

How to run locally

To run locally, simply execute python app.py, making sure you have all the required packages. Note that the environment variables SLACK_BOT_TOKEN (set the value to Bot User OAuth Token), SLACK_APP_TOKEN (set the value to User OAuth Token) and SLACK_CHANNEL_ID (set the value to <channel id>) should be set prior to running the app. This can easily be done by adding a .env file next to app.py. Next, run /papers help in an allowed channel to learn what can be done with this app.

How to host on AWS

To host on AWS, it is much more involved, but thankfully can be achieved with the free tier (max 1M requests per month). First, due to the 3 sec timeout set by Slack, the dispatcher dispatcher.py must be in its own AWS Lambda function that has publishing access to a SNS. Additionally, an AWS lambda API Gateway trigger must be added and set as the Slack app request URL.

Next, the actual app should reside in another AWS Lambda function that triggers on an SNS event. Since AWS Lambda doesn't have all Python packages, you need to package the virtual environment Python packages (specifically [virtualenv]/lib/python3.**/site-packages, here packaged as packages-python-3-10.zip) with the python files app.py, utils.py, and aws_app.py into a zip file (here app.zip, generated with package.sh) and upload it to AWS. The environment variables (SLACK_BOT_TOKEN, SLACK_APP_TOKEN and SLACK_CHANNEL_ID) can be set directly in the AWS dashboard. Note that the python version used in the virtual environment should match the one on AWS, as well as the operating system for packages such as numpy. Finally, because the app execution takes longer than 3 seconds, the timeout of the AWS Lambda function should be increased to a couple of minutes.

Example output

  Rank    Avg. Score    # voters  Link and title
------  ------------  ----------  -------------------------------------------------------------------------------
     1           3             4  http://arxiv.org/abs/2305.02907
                                  [2305.02907] Versatile parametric coupling between two statically decoupled
                                  transmon qubits
     2           3             6  http://arxiv.org/abs/2304.10401
                                  [2304.10401] The quartic Blochnium: an anharmonic quasicharge superconducting
                                  qubit
     3           3             4  http://arxiv.org/abs/2304.09345
                                  [2304.09345] Advancements in Superconducting Microwave Cavities and Qubits for
                                  Quantum Information Systems
     4           2.8           5  https://journals.aps.org/prx/abstract/10.1103/PhysRevX.13.021017
                                  Phys. Rev. X 13, 021017 (2023)  -  Consistent Quantization of Nearly Singular
                                  Superconducting Circuits
     5           2.8           5  https://arxiv.org/abs/2304.05601
                                  [2304.05601] Quantum control and noise protection of a Floquet $0-π$ qubit
     6           2.7           7  http://arxiv.org/abs/2304.08531
                                  [2304.08531] Symplectic geometry and circuit quantization
     7           2.6           5  https://journals.aps.org/prxquantum/abstract/10.1103/PRXQuantum.4.020319
                                  PRX Quantum 4, 020319 (2023)  -  Experimental Realization and Characterization
                                  of Stabilized Pair-Coherent States
     8           2.5           4  http://arxiv.org/abs/2304.13656
                                  [2304.13656] A diagrammatic method to compute the effective Hamiltonian of
                                  driven nonlinear oscillators
     9           2.5           2  https://arxiv.org/abs/2304.01741
                                  [2304.01741] An Intuitive Visualisation Method for Arbitrary Qutrit (Three
                                  Level) States
    10           2.3           3  http://link.aps.org/doi/10.1103/PhysRevB.107.195405
                                  Phys. Rev. B 107, 195405 (2023)  -  Quantum phase slips in a resonant Josephson
                                  junction