JJWT is a 'clean room' implementation based solely on the [JWT](https://tools.ietf.org/html/draft-ietf-oauth-json-web-token-25), [JWS](https://tools.ietf.org/html/draft-ietf-jose-json-web-signature-31), [JWE](https://tools.ietf.org/html/draft-ietf-jose-json-web-encryption-31) and [JWA](https://tools.ietf.org/html/draft-ietf-jose-json-web-algorithms-31) RFC draft specifications.
Most complexity is hidden behind a convenient and readable builder-based [fluent interface](http://en.wikipedia.org/wiki/Fluent_interface), great for relying on IDE auto-completion to write code quickly. Here's an example:
// the sub field was missing or did not have a 'jsmith' value
}
```
If it is important to react to a missing vs an incorrect value, instead of catching `InvalidClaimException`, you can catch either `MissingClaimException` or `IncorrectClaimException`:
If your JWT body is large and you have size restrictions (for example, if embedding a JWT in a URL and the URL must be under a certain length for legacy browsers or mail user agents), you may now compress the JWT body using a `CompressionCodec`:
.compressWith(CompressionCodecs.DEFLATE) // or CompressionCodecs.GZIP
.signWith(SignatureAlgorithm.HS256, key)
.compact();
```
This will set a new `calg` header with the name of the compression algorithm used so that parsers can see that value and decompress accordingly.
The default parser implementation will automatically decompress DEFLATE or GZIP compressed bodies, so you don't need to set anything on the parser - it looks like normal:
If the DEFLATE or GZIP algorithms are not sufficient for your needs, you can specify your own Compression algorithms by implementing the `CompressionCodec` interface and setting it on the parser:
You will then need to specify a `CompressionCodecResolver` on the parser, so you can inspect the `calg` header and return your custom codec when discovered:
*NOTE*: Because body compression is not JWT specification compliant, you should only enable compression if both your JWT builder and parser are JJWT versions >= 0.6.0, or if you're using another library that implements the exact same functionality. This feature is best reserved for your own use cases - where you both create and later parse the tokens. It will likely cause problems if you compressed a token and expected a 3rd party (who doesn't use JJWT) to parse the token.
- Minor [bug](https://github.com/jwtk/jjwt/issues/31) fix [release](https://github.com/jwtk/jjwt/issues?q=milestone%3A0.5.1+is%3Aclosed) that ensures correct Base64 padding in Android runtimes.
- Android support! Android's built-in Base64 codec will be used if JJWT detects it is running in an Android environment. Other than Base64, all other parts of JJWT were already Android-compliant. Now it is fully compliant.
- Elliptic Curve signature algorithms! `SignatureAlgorithm.ES256`, `ES384` and `ES512` are now supported.
- Super convenient key generation methods, so you don't have to worry how to do this safely:
The `generate`* methods that accept an `SignatureAlgorithm` argument know to generate a key of sufficient strength that reflects the specified algorithm strength.
This is a handy little feature. If you need to parse a signed JWT (a JWS) and you don't know which signing key was used to sign it, you can now use the new `SigningKeyResolver` concept.
A `SigningKeyresolver` can inspect the JWS header and body (Claims or String) _before_ the JWS signature is verified. By inspecting the data, you can find the key and return it, and the parser will use the returned key to validate the signature. For example:
The signature is still validated, and the JWT instance will still not be returned if the jwt string is invalid, as expected. You just get to 'see' the JWT data for key discovery before the parser validates. Nice.
This of course requires that you put some sort of information in the JWS when you create it so that your `SigningKeyResolver` implementation can look at it later and look up the key. The *standard* way to do this is to use the JWS `kid` ('key id') field, for example:
You could of course set any other header parameter or claims parameter instead of setting `kid` if you want - that's just the default field reserved for signing key identification. If you can locate the signing key based on other information in the header or claims, you don't need to set the `kid` field - just make sure your resolver implementation knows how to look up the key.
Finally, a nice `SigningKeyResolverAdapter` is provided to allow you to write quick and simple subclasses or anonymous classes instead of having to implement the `SigningKeyResolver` interface directly. For example:
- [Issue 6](https://github.com/jwtk/jjwt/issues/6): Parsing an expired Claims JWT or JWS (as determined by the `exp` claims field) will now throw an `ExpiredJwtException`.
- [Issue 7](https://github.com/jwtk/jjwt/issues/7): Parsing a premature Claims JWT or JWS (as determined by the `nbf` claims field) will now throw a `PrematureJwtException`.
This release adds convenience methods to the `JwtBuilder` interface so you can set claims directly on the builder without having to create a separate Claims instance/builder, reducing the amount of code you have to write. For example, this:
But you couldn't easily determine if the `jwt` was a `JWT` or `JWS` instance or if the body was a `Claims` instance or a plaintext `String` without resorting to a bunch of yucky `instanceof` checks. In 0.2, we introduce the `JwtHandler` when you don't know the exact format of the compact JWT string ahead of time, and parsing convenience methods when you do.
If you do not know the format of the compact JWT string at the time you try to parse it, you can determine what type it is after parsing by providing a `JwtHandler` instance to the `JwtParser` with the new `parse(String compactJwt, JwtHandler handler)` method. For example:
```java
T returnVal = Jwts.parser().setSigningKey(key).parse(compact, new JwtHandler<T>() {
//inspect it, then return an instance of T (see returnVal above)
}
});
```
Of course, if you know you'll only have to parse a subset of the above, you can use the `JwtHandlerAdapter` and implement only the methods you need. For example:
```java
T returnVal = Jwts.parser().setSigningKey(key).parse(plaintextJwt, new JwtHandlerAdapter<Jwt<Header,T>>() {
If, unlike above, you are confident of the compact string format and know which type of JWT or JWS it will produce, you can just use one of the 4 new convenience parsing methods to get exactly the type of JWT or JWS you know exists. For example:
JJWT depends on Jackson 2.4.x (or later). If you are already using a Jackson version in your own application less than 2.x, for example 1.9.x, you will likely see [runtime errors](https://github.com/jwtk/jjwt/issues/1). To avoid this, you should change your project build configuration to explicitly point to a 2.x version of Jackson. For example:
