PLEASE VIEW THIS DOCUMENTATION ON WWW.ESPRUINO.COM - you can even Search. The markdown files in this repository are designed to be parsed by a build script - by viewing them on GitHub you may be missing useful links or information.
This project generates the website and tutorial code for Espruino.com. This also contains the modules that can be required by the Espruino Web IDE. You can contribute to the site and modules in this project.
- Find out how to create modules here
- Find out how to create tutorials here
Documentation files are written GitHub Markdown with additional metadata. Basically, a script that looks for the following (on the start of a line):
* KEYWORDS: Comma,Separated,List ; Defines keywords for this file
* APPEND_KEYWORD: Keyword ; Append a list of pages that match the keyword
* USES: Comma,Separated,List ; Defines parts that are used by the given tutorial
* APPEND_USES: part ; Append a list of pages that have this part in their USES_PARTS list
* APPEND_JSDOC: filename ; Append JavaScript documentation based on the JS in the given file
* APPEND_PINOUT: boardname ; Append Pinout for the given board
It also looks for a title (second line, after copyright notices) which it uses to create the title of the HTML page (and of links to it).
There are a few extra bits too:
[[My Page]]links to a page on the Espruino website[[http://youtu.be/VIDEOID]]puts a video on the pageAdds an image. Images should be in a directory named after the filename of the file referencing them (or the same directory as the file referencing them)
It then converts the Markdown to HTML and shoves it on the Espruino website. Lovely!
Any .js files in examples have a webpage created that uses the comments as markdown, and then adds the code as a code block right at the end.
All other .js files are treated as modules. They are minified using Google's online closure compiler and the SIMPLE_OPTIMISATIONS flag. To get advanced optimisations, you must add the exact text @compilation_level ADVANCED_OPTIMIZATIONS into the comments at the head of the file.
Checked out the Espruino source at the same same level as EspruinoDocs folder. Assuming you're in the EspruinoDocs folder...
$ cd ..
$ git clone git@github.com:espruino/Espruino.git`
$ cd EspruinoDocs
If you have Node.js and npm installed skip this step.
If you haven't got Node.js JavaScript runtime installed or the JavaScript Package managers installed do so via the installation guides below.
In order to generate the documentation and view it you require several JavaScript packages. To install them issue the following command:
$ npm install
This will install all JavaScript dependancies.
Currently they are two build scripts. One bash, one JavaScript.
The bash script does 3 things:
- Uses the
Espruinosource code to generate the pinout diagrams.pythonis required - Builds the production site at
~/workspace/espruinowebsite - Builds Espruino specific modules and minifies the JavaScript code
The JavaScript build process just builds the documentation in the html folder.
You will have to run build.sh at least once if you want the build.js to work.
Run:
$ ./build.sh
Note: You've had to have least ran the bash script once for this to build successfully.
Run:
$ npm run build
The output will be placed in the html directory.
You can load a development version of the website locally. It will not look exactly like the production site but you can test your build and links.
$ npm start
Then load up a page in a browser: http://localhost:3040/EspruinoBoard
On OSX, most likely the default amount of open files will be set too low. This may cause an error during the build, like: "Error: EMFILE, too many open files 'tasks/File Converter.md'"
Make sure you have at least 1024 for the value of open files.
$ ulimit -n # see current limit
Increase the limit:
$ ulimit -n 1024 # increase to 1024