PyBrowserID: a python library for the BrowserID Protocol

This is a python client library for the BrowserID protocol that underlies Mozilla Persona:

https://login.persona.org/

For the vast majority of deployments, you will simply want to call the module- level "verify" functon to verify a given assertion:

>>> data = browserid.verify(BROWSERIDASSERTION, "http://mysite.com")
>>> print data["email"]
"[email protected]"

The precise implementation of this function will change depending on the current recommendations of the BrowserID team. Currently it POSTs the assertion to the remote verifier service on persona.org.

Note that you must specify your site's root URL as the second argument to that function. This is the "expected audience" and is a key security feature of BrowserID.

If you are not able to determine the precise hostname by which your site is being accessed (e.g. due to virtual hosting) then you may specify one or more wildcard patterns like so:

>>> data = browserid.verify(BROWSERIDASSERTION, ["http://*.mysite.com"])
>>> print data["email"]
"[email protected]"

For finer control over the verification process, you can create an instance of a "Verifier" class and avoid having to specify the audience patterns over and over again:

>>> verifier = browserid.RemoteVerifier(["*.mysite.com"])
>>> data = verifier.verify(BROWSERIDASSERTION)
>>> print data["email"]
"[email protected]"

For improved performance, or if you just want to live on the bleeding edge, you can explicitly perform verification locally by using the LocalVerifier class like so:

>>> verifier = browserid.LocalVerifier(["*.mysite.com"])
>>> data = verifier.verify(BROWSERIDASSERTION)
>>> print data["email"]
"[email protected]"

Note that the details of the BrowserID Protocol are still in flux, so local verification might break due to incompatible changes. As things stabilise this will become the default implementation.