Uncomment algorithms for jwt.

[kepkin]

Summary

Uncommented HS384 & HS512 algorithms for jwt

Full changelog

• Uncomment HS384 & HS512
• Add related tests

[kepkin]

Can somebody restart travis build? It looks like a random timeout error.

[ RUN      ] spec/03-plugins/01-tcp-log/01-tcp-log_spec.lua @ 106: Plugin: tcp-log (log) [#postgres] performs a TLS handshake on the remote TCP server
Error from thread: ./spec/helpers.lua:449: timeout
stack traceback:
	[C]: in function 'assert'
	./spec/helpers.lua:449: in function <./spec/helpers.lua:442>
spec/03-plugins/01-tcp-log/01-tcp-log_spec.lua:121: Expected objects to be the same.
Passed in:
(boolean) false
Expected:
(boolean) true
stack traceback:
	spec/03-plugins/01-tcp-log/01-tcp-log_spec.lua:121: in function <spec/03-plugins/01-tcp-log/01-tcp-log_spec.lua:106>
[  FAILED  ] spec/03-plugins/01-tcp-log/01-tcp-l

[thibaultcha]

Thank you @kepkin for the contribution! Looking forward to a few changes in the tests (thanks for providing them), and this should be good to go. Would you mind taking the time to address the comments? Thanks!

[thibaultcha]

Since 0.14.0 (now in the master branch), we only support OpenResty 1.13.2.6 and above, so we do not need these branches anymore. Our CI also runs with recent OpenResty version(s) as well: https://github.com/Kong/kong/blob/master/.travis.yml#L28-L29

[thibaultcha]

If the feature is supposedly not implemented on some older dependency version (although I am not sure why it wouldn't), then isn't the assertion doomed to fail?

Also, 'not' and 'implemented' clearly aren't equal, so the second assertion doesn't really make any sense? If anything, jwt_parser.encode() will throw an error, not return one.

This while branch goes away anyway since we only support OpenResty 1.13.6.2 and above from now on :)

[thibaultcha]

We consistently use double quotes throughout this codebase, will you please convert these single ones? There are other similar places in this patch. Thanks!

[thibaultcha]

@kepkin Hi there,

Any chance you could fix the few comments on this PR? Thanks!

[kepkin]

Hi @thibaultcha, sorry for the delay. Was busy with other tasks :-)

[thibaultcha]

@kepkin No worries. Will you also be able to provide a few integration tests? See how 4e2e1d7#diff-7169eb138e87d0233774868e385f8dcdR272 added a couple of tests - we should do the same for these new algorithms and make sure that their implementation works end-to-end. Will you have some time to do this? The tests will likely be very similar to the ones I linked you to, and can simply be adapted.

[kepkin]

@thibaultcha thank you for pointing out these tests. It required some fixes.

[thibaultcha]

@kepkin Very nice! Thank you for taking care of the tests. In order for us to merge your feature, we would also need an update to our documentation: https://github.com/Kong/docs.konghq.com

Would you be able to open a PR there (against the master branch is fine, and we'll rebase it appropriately for the next release).

Thank you!

[kepkin]

Looks like there is not much to add
Kong/docs.konghq.com#818

[thibaultcha]

@kepkin Thanks for the docs PR! This is merged as well to master with minor edits.

Feel free to reach out to us if you wish to receive a Contributor T-shirt 😉