Token parsing is not consistent between ExtractJwt.fromHeader and the default token parser about passport-jwt HOT 3 CLOSED

extractors.fromHeader('Authorization') should be interchangeable with extractors.fromAuthHeader(). This is not the case.

The bug responsible for the inconsistency is due to the fact that fromHeader() does not account for the four first characters in JWT ${token} as fromAuthHeader() does.

from passport-jwt.

There are multiple extractors that read the JWT from headers, you may just be missing the differences.

extractors.fromHeader reads the JWT from any header, allowing you to specify the header name.

extractors.fromAuthHeader works just like the default token parser and reads the JWT from the authorization header if and only if the authentication scheme is 'JWT'

extractors.fromAuthHeaderWithScheme reads the JWT from the Authorization header and allows the user to specify the scheme they expect so they can use something different from JWT.

Each of these extractors was added in response to multiple feature requests. I don't think its a matter of inconsistent implementation as much as it is enabling as much flexibility as possible.

from passport-jwt.

Closing but if you want to suggest some documentation improvements feel free to submit a pull request.

from passport-jwt.