Keep returnTo when using successReturnToOrRedirect

[grahamwhiteuk]

This is a small patch to fix the behaviour of the successReturnToOrRedirect option as illustrated in #919 and #281

As I understand it, the successReturnToOrRedirect option is supposed to, upon success, return or redirect to the string specified in session.returnTo or if session.returnTo is not specified then return/redirect to the string specified in the successReturnToOrRedirect option.

The problem with the situation as-is, is that session.returnTo is wiped out including session.returnTo and thus the user is always routed towards the default as specified in successReturnToOrRedirect. This behaviour is unexpected and one would expect that when you have set a returnTo value in the session and also specified the successReturnToOrRedirect option that the user would be directed towards the value specified in the session. As described in #919, there is a workaround by setting the keepSessionInfo option to true but this workaround involves keeping all of the previous session values rather than the minimal set required for successReturnToOrRedirect to operated as expected.

This patch ensures that if successReturnToOrRedirect is being used and the session contains a returnTo value that the session.returnTo value is maintained and thus the expected behaviour is implemented.

Checklist

• I have read the CONTRIBUTING guidelines.
• I have added test cases which verify the correct operation of this feature or patch.
• I have added documentation pertaining to this feature or patch.
• The automated test suite ($ make test) executes successfully.
• The automated code linting ($ make lint) executes successfully.

[sbsamaro]

@jaredhanson Can we get this fix in?

[shashank686]

@jaredhanson please fix it, so then there will be no need of a workaround by using keepSessionInfo flag as returnTo will be preserved.

[sbsamaro]

@shashank686 The work around did not work for me. Did it work for you? So right now the return too is not perserved and my app is failing with that behavior if i upgrade.

[florianwalther-private]

Is this going to be merged?

[grahamwhiteuk]

Is this going to be merged?

I don't think so. Seems like this repo is dead and no longer maintained.

[florianwalther-private]

Is this going to be merged?

I don't think so. Seems like this repo is dead and no longer maintained.

That sucks to hear. Do you know an alternative package that does the same job?

[jaredhanson]

Seems like this repo is dead and no longer maintained.

It's still maintained. As a general rule, I don't merge pull requests that lack corresponding tests. If this PR is updated with test cases, I will merge and publish. Otherwise, it'll wait until I have the time to write the tests.

Thanks.