README.md

Ludum Dare

This is the source code for the Ludum Dare event website(s)! It's currently live here:

https://ldjam.com/

Want to help out? Setup Instructions are here:

https://github.com/ludumdare/dairybox

Development discussion (site-dev only please):

• Web: https://gitter.im/ludumdare/ludumdare (Slack-like)
• IRC: https://irc.gitter.im/

Structure

Many folder are documented. Just browse the tree to learn about the contents.

Source Code

Source code is found here:

• /src - Source Code (PHP, JavaScript, CSS, etc)

Live sites

These folders contain the live sites that are served. They tend to be simple PHP files that include things from the /src folder. They also contain the output of the toolchain (in the /-/ subfolder).

• /public-ludumdare.com - Ludum Dare focused version of the common site -- https://ldjam.com
  • This will be moving to ludumdare.com eventually.
• /public-url.shortener - A variety of URL shortening services. -- https://ldj.am, https://jam.mr
  • At the moment only https://ldj.am/$id is supported, where $id is the node_id to redirect to. e.g. https://ldj.am/$11

These are not currently in active development.

• /public-jammer.vg - Jammer, the general version of the common site -- https://jammer.vg
• /public-jammer.bio - Jammer Bio's, a variant focused on user pages. i.e. jammer.bio/your-user-name

Other

• /public-api - Restful API calls. https://api.ldjam.com
  • NOTE: /vx/ is experimental and may change. We are working towards a /v1/ launch soon.
• /public-static - Where static files go. Images, etc. -- https://static.jam.vg
• /sandbox - Old code, experiments, and debugging tools. Very chaotic here.

The Makefile

The Makefile is the core build script. It is used to build the project. Invoke it with the make command from inside the VM (or outside with appropriate config).

Usage:

• make - Compile all changed files and build the target
• make clean - Delete all intermediate files. NOTE: When you make again, everything needs to be recompiled
• make lint - Run all code through the linter. NOTE: make only lints files that have changed. This lints everything
• Advanced build options
  • make all - Build all targets (default, until you set TARGET in config.mk)
  • make TARGET=public-ludumdare.com - Make a specific target (in this case, public-ludumdare.com)
  • make mini - Like make, but refreshes the UID (used to bypass caching proxies, etc)
  • make clean-some - Clean, but don't delete the output files (useful when clean building on live)
  • make clean-all - Clean all targets
  • make clean-svg, make clean-css, make clean-js - Clean specifically the SVG, CSS, or JS
  • make clean-all-svg, make clean-all-css, make clean-all-js - Clean specifically the SVG, CSS, or JS of all targets
  • make lint-all - Lint all code for all targets
  • make lint-css, make lint-js, make lint-php - Lint specifically the CSS, JS and PHP code
  • make lint-all-css, make lint-all-js, make lint-all-php - Lint specifically the CSS, JS, or PHP code of all targets
  • make clean-lint - Force re-linting of everything on build
• config.mk - Create this file and you can hardcode makefile settings
  • TARGET=public-ludumdare.com to build just Ludum Dare
  • DEBUG=true to enable debug builds (append ?debug to the URL)
  • SOURCEMAPS=true to enable Source Maps. NOTE: Requires a full rebuild when changed
  • JOBS=2 to compile using 2 threads (parallel build)
    • NOTE: VirtualBox is really bad at threads. Beyond 2, you start seeing serious diminishing returns, to the point where it actually gets worse than 2 the higher you go.
    • If you actually want to try values other than 1 and 2, it you'll need to change the number of CPUs in Vagrantfile, as well as set JOBS.
    • Below is a list of benchmarks, testing different combinations of CPUs and Jobs. Testing on an AMD FX-8350 (8 core, which is not Hyperthreaded, but rather has 2 cores per die that share caches)

C,J  Sec   Change Diff
-----------------------
1,1: 901 = 100% = 0%
2,2: 521 = 173% = 73%
3,3: 427 = 211% = 38%
4,4: 401 = 225% = 14%
8,8: 824 = 109% = -116%