Repology is a service which monitors a lot of package repositories and other sources and aggregates data on software package versions, reporting new releases and packaging problems.
This repository contains Repology updater code, a backend service which updates the repository information. See also the web application code.
- Python 3.8+
- Python module Jinja2
- Python module libversion (also requires libversion C library)
- Python module psycopg2
- Python module pyyaml
- Python module xxhash
- PostgreSQL 12.0+
- PostgreSQL extension libversion
Needed for fetching/parsing repository data:
- Python module requests
- Python module rubymarshal
- Python module lxml
- Python module rpm (comes with RPM package manager)
- Python module jsonslicer
- Python module pyparsing
- Python module protobuf
- Python module sqlite3 (part of Python, sometimes packaged separately)
- git
- rsync
- subversion
Optional, for doing HTML validation when running tests:
- Python module pytidylib and tidy-html5 library
Optional, for checking schemas of configuration files:
- Python module voluptuous
Optional, for python code linting:
- Python module flake8
- Python module flake8-builtins
- Python module flake8-import-order
- Python module flake8-quotes
- Python module mypy
Since repology rules live in separate repository you'll need to
clone it first. The location may be arbitrary, but rules.d
subdirectory is what default configuration file points to, so
using it is the most simple way.
git clone https://github.com/repology/repology-rules.git rules.d
First, you may need to tune settings which are shared by all repology
utilities, such as directory for storing downloaded repository state
or DSN (string which specifies how to connect to PostgreSQL database).
See repology.conf.default
for default values, create repology.conf
in the same directory to override them (don't edit repology.conf.default
!)
or specify path to alternative config via REPOLOGY_SETTINGS
environment variable, or override settings via command line.
By default, repology uses ./_state
directory for storing raw and parsed
repository data and repology/repology/repology
database/user/password
on localhost.
For the following steps you'll need to set up the database. Ensure PostgreSQL server is up and running, and execute the following commands to create the database for repology:
psql --username postgres -c "CREATE DATABASE repology"
psql --username postgres -c "CREATE USER repology WITH PASSWORD 'repology'"
psql --username postgres -c "GRANT ALL ON DATABASE repology TO repology"
psql --username postgres --dbname repology -c "CREATE EXTENSION pg_trgm"
psql --username postgres --dbname repology -c "CREATE EXTENSION libversion"
in the case you want to change the credentials, don't forget to add
actual ones to repology.conf
.
Next you can create database schema (tables, indexes etc.) and at the same time test that the database is accessible with the following command:
./repology-update.py --initdb
The database is now ready to be filled with data. Typical Repology update cycle consists of multiple steps, but in most cases you'll need to just run all of them:
./repology-update.py --fetch --fetch --parse --database --postupdate
--fetch
tells the utility to fetch raw repository data (download files, scrape websites, clone git repos) into state directory. Note that it won't refetch (update) data unless it's specified twice.--parse
enables parsing downloaded data into internal format which is also saved into state directory.--database
pushes processed package data into the database.--postupdate
runs additional database processing such as calculating summaries and updating feeds. It's separate from--database
because it can be ran in background, parallelly to the following fetch/update cycle.
- How to extend or fix rules for package matching
- How repology compares versions
GPLv3 or later, see COPYING.