Generating the files

Requisites

• Your system must have yarn installed
• Access to the internet as files are downloaded on the fly

There are basically two steps

• Info step: Execute crawler/run.sh to gather IDL and MDN information into JSON files
• Generating step: Execute package_js/run.sh to generate the dart files from the generated JSON files

Steps detailed

Information gathering:

The main function calls 6 methods and most of them can be commented after running the first time:

• cloneMDN: will download the whole MDN website compressed from MDN github and unzip locally at tool/crawler/mdn.
• cloneIDLs: will recursively check the w3c repo for IDL files and download them at tool/webIDL/.
• mergeIDLs: will open, read and merge the IDLs files downloaded into one at tool/webIDL/merged.
• startMdn: will cd into the MDN dir and start their web server through yarn install and yarn start
• parseMDN: will fetch the index page of the MDN web site and then crawl all objects they expose while saving into the json at tool/webIDL/mdn.json
• addInfo: will read both the mdn.json and the merged IDL dir and then generate the dir tool/webIDL/info which is basically a parsed IDL with MDN info attached in the objects

Generating the dart files

The main function basically reads the contents of the last dir generated (tool/webIDL/info) in the first step and generates all bindings into the lib/bindings folder