Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/aidos-lab/donut
DONUT: Database of Original and Non-Theoretical Applications of Topology
https://github.com/aidos-lab/donut
flask-application frontend topological-data-analysis
Last synced: 5 days ago
JSON representation
DONUT: Database of Original and Non-Theoretical Applications of Topology
- Host: GitHub
- URL: https://github.com/aidos-lab/donut
- Owner: aidos-lab
- License: bsd-3-clause
- Created: 2022-09-17T15:02:44.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-10-31T01:01:35.000Z (16 days ago)
- Last Synced: 2024-10-31T02:16:37.372Z (16 days ago)
- Topics: flask-application, frontend, topological-data-analysis
- Language: TeX
- Homepage: https://donut.topology.rocks
- Size: 847 KB
- Stars: 10
- Watchers: 5
- Forks: 1
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# DONUT: Database of Original and Non-Theoretical Uses of Topology
[](https://zenodo.org/badge/latestdoi/537842772)   
This is the code for [DONUT](https://donut.topology.rocks), a search
engine for applications of topology. This repository contains the code
for the backend in case you want to run this search engine for yourself.
**Please use this repository to report any usability issues and
technical errors only. For missing articles, contact
[email protected].**
# Citing or acknowledging DONUT
Please use the following BibTeX citation when citing or acknowledging
DONUT in your own work:
```bibtex
@misc{DONUT,
author = {Giunti, Barbara and Lazovskis, J{\=a}nis and Rieck, Bastian},
title = {DONUT: Database of Original \& Non-Theoretical Uses of Topology},
url = {https://donut.topology.rocks},
year = {2022},
key = {DONUT},
}
```
# Setup
DONUT is relatively easy to set up. These notes are a work in progress.
## Installation
- Setup virtual environment
- Install `xapian` on system
- Download `xapian-bindings`
- Run `./configure --with-python3 --prefix=$VIRTUAL_ENV`
- Run `make install`
On Debian, this also works:
```console
$ sudo apt-get install python3-xapian
$ python3 -m venv --system-site-packages .venv
$ source .venv/bin/activate
$ pip install .
```
Afterwards, we need to set up a `systemd` service unit to start the
service automatically:
```bash
$ cat /etc/systemd/system/donut.service
[Unit]
Description=Gunicorn instance for serving "DONUT"
After=nginx.service
Wants=nginx.service
[Service]
User=bastian
Group=www-data
WorkingDirectory=/home/bastian/Projects/DONUT
Environment="PATH=/home/bastian/Projects/DONUT/.venv/bin:/usr/bin"
ExecStart=/home/bastian/Projects/DONUT/.venv/bin/gunicorn --workers 4 --bind unix:donut.sock -m 007 "donut:create()"
[Install]
WantedBy=multi-user.target
$ sudo systemctl enable donut.service
$ sudo systemctl start donut.service
```
Presto, we are done! Notice the use of `Environment` to specify the
environment variables required for running DONUT. Moreover, the
`WorkingDirectory` specification is necessary to set relative paths
for the service.
Finally, to make everything accessible from external hosts, you need to
configure a new virtual host and set your web server up to handle proxy
request. Here is how I did that for `nginx`:
```
$ cat /etc/nginx/sites-available/donut.topology.rocks
server
{
listen 80;
server_name donut.topology.rocks;
# enforce HTTPS
return 301 https://$server_name$request_uri;
}
server
{
listen 443 ssl;
server_name donut.topology.rocks;
ssl_certificate /etc/letsencrypt/live/donut.topology.rocks/fullchain.pem; # managed by Certbot
ssl_certificate_key /etc/letsencrypt/live/donut.topology.rocks/privkey.pem; # managed by Certbot
location /
{
include proxy_params;
proxy_pass http://unix:/home/bastian/Projects/DONUT/donut.sock;
}
}
```
The certificate information is of course optional, but it is good
practice to ensure that `http` requests are always forward to `https`.
## Updating the Database
Install a cron job that runs the following command:
python -m donut.zotero
## Running the app
flask --app donut:create run