Custom Parsers
JSON Schema $Ref Parser comes with built-in JSON, YAML, plain-text, and binary parsers, but you can add your own parsers to support additional file types, or even replace any of the built-in parsers with your own custom implementation.
You can see the source code for any of the built-in parsers right here.
Simple Example: Creating your own parser
The fastest way to learn is by example, so let’s do that. Here’s a simplistic parser that parses CSV files (comma-separated values):
let myParser = {
order: 1,
canParse: ".csv",
parse(file) {
let lines = file.data.toString().split("\n");
return lines.map((line) => {
return line.split(",");
});
},
};
$RefParser.dereference(mySchema, { parse: { csv: myParser } });
The order
property
All parsers have an order
property, even the built-in parsers. If you don’t specify an order
property, then your parser will run last. Specifying order: 1
, like we did in this example, will make your parser run first. Or you can squeeze your parser in-between some of the built-in parsers. For example, order: 201
would make it run after the JSON and YAML parsers, but before the plain-text and binary parsers. You can see the order of all the built-in parsers by looking at their source code.
The order
property and canParse
property are related to each other. For each file that JSON Schema $Ref Parser needs to parse, it first determines which parsers can parse that file by checking their canParse
property. If only one parser matches a file, then only that one parser is called, regardless of its order
. If multiple parsers match a file, then those parsers are tried in order until one of them successfully parses the file. Once a parser successfully parses the file, the rest of the parsers are skipped.
If none of the parsers match the file, then all of them are tried, in order, until one of them successfuly parses the file, or until they all fail. This is useful for situations where a file is a supported type, but it has an unrecognized file extension.
The canParse
property
The canParse
property tells JSON Schema $Ref Parser what kind of files your parser can handle. In this example, we’ve simply specified a file extension, but we could have used a simple boolean, an array of file extensions, a regular expression, or even a function with custom logic to determine which files to parse. Here are examples of each approach:
let myParser = {
// Parse ALL file types
canParse: true
// An array of file extensions (lowercased)
canParse: [".txt", ".csv"]
// A regular expression (matched against the FULL file path)
canParse: /\.(txt|csv)$/i
// A function that returns a truthy/falsy value
canParse(file) {
return file.extension === ".csv" || file.extension === ".txt";
}
};
When using the function form, the file
parameter is a file info object, which contains information about the file being parsed.
The parse
method
Obviously, this is where the real work of a parser happens. The parse
method accepts the same file info object as the canParse
function, but rather than returning a boolean value, the parse
method should return a JavaScript representation of the file contents. For our CSV parser, that is a two-dimensional array of lines and values. For your parser, it might be an object, a string, a custom class, or anything else.
Unlike the canParse
function, the parse
method can also be asynchronous. This might be important if your parser needs to retrieve data from a database or if it relies on an external HTTP service to return the parsed value. You can return your asynchronous value via a Promise or a Node.js-style error-first callback. Here are examples of both approaches:
let myCallbackParser = {
// Return the value in a callback function
parse(file, callback) {
doSomethingAsync(file.data, (data) => {
if (data) {
// Success !
callback(null, data);
} else {
// Error !
callback(new Error("No data!"));
}
});
},
};
let myPromiseParser = {
// Return the value in an ES6 Promise
async parse(file) {
let data = await doSomethingAsync(file.data);
if (data) {
// Success !
return data;
} else {
// Error !
throw new Error("No data!");
}
},
};