CORTEX REAVER - A dangerous Ruby blog engine, with a photographic memory.

Cortex Reaver is a terrifying machine which stalks the corridors of Citadel Station, searching for humans to assimilate into Shodan’s perfection.

Cortex Reaver is also a blog engine for textual journals, photographs, and pages, written using Ramaze and Sequel. It supports tags, comments, and simple attachments. Features include canonically named URLs (/journals/show/a-post-title) with conflict resolution, useful formatting modules for content (HTML sanitizing, attachment references, BlueCloth, etc), and custom pages.

Why Cortex Reaver

  • Adaptible. Runs on every web server and database Ramaze and Sequel support.

  • Visual. Large, beautiful photographs, with concise EXIF support.

  • Comprehensive. Comments, file attachments, and tags on everything.

  • Devastating. The main page of aphyr.com scales to ~400 requests/second.

Getting Started

mkdir cortex_reaver # Create an empty directory for Cortex Reaver to lurk in. cd cortex_reaver cortex_reaver migrate # Initialize an SQLite database. cortex_reaver start # Start the daemon.

Then point your web browser at localhost:7000, hit control-x, and log in with username “shodan”, password “citadelstation”.

You can check Cortex Reaver’s status with status, and when you’re done, shut it down with cortex_reaver stop.

Configuration

You can use a YAML configuration file to define Cortex Reaver’s configuration. See proto/cortex_reaver.yaml. If no config file is specified on the command line with –config, Cortex Reaver assumes you mean cortex_reaver.yaml in the working directory.

Databases

Cortex Reaver uses Sequel (sequel.rubyforge.org/), so you can (in theory) use any database Sequel supports. By default, Cortex Reaver builds an SQLite database in the working directory, but you can set a different database in the config file, either as a Sequel string or with hash of options.

For MySQL:

create database cortex_reaver; grant all privileges on cortex_reaver.* to ‘cortex_reaver’@‘localhost’ 

identified by 'some-password' with grant option;

When you’ve set up a new database, run cortex_reaver migrate to update it to the latest version. If you need to wipe your database and start over, use cortex_reaver blank to drop the schema and rebuild it.

You can export your database to an SQLite file with cortex_reaver dump, and restore it with cortex_reaver load. This can help with migrating between database systems.

Customization

You may find development mode useful: it reloads source files when they’re changed, displays traces for errors instead of 404 and 500 responses, logs more, and runs in the console instead of detaching by default. Just set mode => :development in your configuration.

Cortex Reaver includes Erubis layouts and templates, static resources, CSS, and javascript files in lib/cortex_reaver/layout, lib/cortex_reaver/view, and lib/cortex_reaver/public. CR will override these resources with your local directory: if you request /images/foo.png, it will check <local_dir>/public/images/foo.png first, then <cortex_reaver_dir>/public/images/foo.png. CSS and JS files are compiled into single cached files from several smaller files. If you want, you can selectively replace parts of the default theme by creating css files with the same names in <local_dir>/public/css/. The same applies to views and layouts. You can also create new resources; they will be merged for you.

You can control the paths and search patterns in the configuration: see lib/cortex_reaver/config.rb.

Requests for static objects (images, css, javascript, etc.) is pulled from public/, which can be specified in your configuration or defaults to ‘cwd`/public. If a request for a file can’t be found there, it’s retrieved from Cortex Reaver’s own public directory in lib/public.

Credits

I am indebted to the good folks of #sequel and #ramaze for their many suggestions in writing Cortex Reaver, and also to Ryan Grove (wonko.com), the author of the excellent (and very similar) blog system Thoth, for invaluable advice and code. Cortex Reaver should have been a fork of Thoth, but my schema is just too complex for the changes to be maintainable going forward.