Compare commits

...

2 Commits

Author SHA1 Message Date
SoniEx2 3483dbe9b5 Update README.md 2019-06-20 17:07:18 -03:00
SoniEx2 7a2b72bc8c Fix crash when loading multiple configs 2019-06-20 16:50:56 -03:00
2 changed files with 35 additions and 29 deletions

View File

@ -7,32 +7,31 @@ More importantly, it's a tool to help fight against BDFLs and other forms of cen
Basic Usage
-----------
Note: GAnarchy does not come with a main page. You need to provide one yourself. This should be in your website's `/` path (usually `index.html`).
The protocol handler uses the path `/?url=%s`. Additionally, your main page needs to include the `index.js` provided by GAnarchy
for the protocol handler to properly work.
First, initialize the database with `ganarchy.py initdb`. The database is stored in the XDG data home, as per XDG spec.
Then, set the project commit with `ganarchy.py set-commit COMMIT`, where `COMMIT` is the full commit hash.
The commit *must* start with `[Project]` followed by the project name, and may have an optional description.
(Note: This requirement isn't properly checked, but will be in the future. This is important for a future federation
protocol that allows for automatically discovering forks based on the project commit.)
Then create or edit the file `$XDG_CONFIG_HOME/ganarchy/config.toml`. It should contain the following items:
Once everything is initialized, add some repos with `ganarchy.py repo add URL`, and enable or disable them with `ganarchy.py repo enable URL` or
`ganarchy.py repo disable URL`. They currently come enabled by default. You are now ready to go.
Finally, add `ganarchy.py cron-target > path/to/page.html` to your cron. Optionally `scp page.html scp://server@example.org/page.html`.
Example shell session:
```text
$ ganarchy.py initdb
$ ganarchy.py set-commit 385e734a52e13949a7a5c71827f6de920dbfea43
$ ganarchy.py set-project-name GAnarchy
$ ganarchy.py repo add https://cybre.tech/SoniEx2/ganarchy
$ ganarchy.py cron-target > index.html
$ scp index.html scp://example.org/var/www/html/index.html
```
# Example GAnarchy config
# The base_url is the web address of the GAnarchy instance
# Restrictions: MUST be present. SHOULD be https.
base_url = "https://ganarchy.autistic.space/"
# The title is shown on the homepage. If not present, defaults to "GAnarchy on [base_url's hostname]"
title = "GAnarchy on autistic.space"
# The projects table is made up of "project commit" hashes (see below for project commit)
[projects.385e734a52e13949a7a5c71827f6de920dbfea43]
# Each project is made up of repos and branches
# HEAD is special and refers to the default branch
# Restrictions: active MUST be present, file URLs are disallowed
"https://cybre.tech/SoniEx2/ganarchy".HEAD = { active=true }
# repos/branches with active=false will not be shown or updated
"https://cybre.tech/SoniEx2/ganarchy".broken = { active=false }
```
A project commit is a commit that *must* start with `[Project]` followed by the project name, and may have an optional description.
Example project commit:
@ -43,13 +42,19 @@ A Project Page Generator written in Python, focused on giving forks of a
project the same visibility as the original repo.
```
To generate the output pages:
```
ganarchy.py cron-target index > index.html
ganarchy.py cron-target config > config.toml # if federation is desired
ganarchy.py cron-target $PROJECT_COMMIT > project/$PROJECT_COMMIT/index.html # for each project
```
It's recommended to use a wrapper script to generate the project pages. `ganarchy.py cron-target project-list` may be of use.
Advanced Usage
--------------
It's also possible to use the same GAnarchy DB with multiple projects. This is currently done by passing `--project COMMIT` to the `repo` commands,
and using `cron-target COMMIT`.
In addition to the basic configuration above, you can also place your own templates in `$XDG_CONFIG_HOME/ganarchy/templates`.
Note that the same repo may be part of multiple projects, so it is necessary to specify `--project COMMIT` also when enabling/disabling it.
Additionally, it's possible to set a non-default branch by passing `--branch BRANCHNAME` to the `repo` commands. Given that multiple branches of
the same repo may be part of the same project, it is also necessary to use `--branch BRANCHNAME` when enabling/disabling them.
The current templates are: `index.html`, `index.toml`, `project.html`.

View File

@ -447,7 +447,8 @@ class Config:
self.blocked_domain_suffixes = tuple(self.blocked_domain_suffixes) # MUST be tuple
# TODO re.compile("(^" + "|^".join(map(re.escape, domains)) + "|" + "|".join(map(re.escape, suffixes) + ")$")
if base:
self._update_projects(base.projects, sanitize=False) # already sanitized
# FIXME is remove=remove the right thing to do?
self._update_projects(base.projects, remove=remove, sanitize=False) # already sanitized
projects = config_data.get('projects', {})
self._update_projects(projects, remove=remove)