Response Headers should be allowed to be null (or at least empty if not set)

[randomcoder]

Whilst running a Gatling test on a server that does not generate any headers at all I get the following exception:

java.lang.IllegalStateException: Response's headers hasn't been computed by your AsyncHandler.
        at com.ning.http.client.providers.netty.NettyResponse.getCookies(NettyResponse.java:184)
        at com.excilys.ebi.gatling.http.response.ExtendedResponse.getCookies(ExtendedResponse.scala:89)
        at com.excilys.ebi.gatling.http.ahc.GatlingAsyncHandlerActor.com$excilys$ebi$gatling$http$ahc$GatlingAsyncHandlerActor$$processResponse(GatlingAsyncHandlerActor.scala:198)
        at com.excilys.ebi.gatling.http.ahc.GatlingAsyncHandlerActor$$anonfun$receive$1.apply(GatlingAsyncHandlerActor.scala:107)
        at com.excilys.ebi.gatling.http.ahc.GatlingAsyncHandlerActor$$anonfun$receive$1.apply(GatlingAsyncHandlerActor.scala:83)
        at akka.actor.Actor$class.apply(Actor.scala:318)
        at com.excilys.ebi.gatling.core.action.BaseActor.apply(BaseActor.scala:23)
        at akka.actor.ActorCell.invoke(ActorCell.scala:626)
        at akka.dispatch.Mailbox.processMailbox(Mailbox.scala:197)
        at akka.dispatch.Mailbox.run(Mailbox.scala:179)
        at akka.dispatch.ForkJoinExecutorConfigurator$MailboxExecutionTask.exec(AbstractDispatcher.scala:516)
        at akka.jsr166y.ForkJoinTask.doExec(ForkJoinTask.java:259)
        at akka.jsr166y.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:975)
        at akka.jsr166y.ForkJoinPool.runWorker(ForkJoinPool.java:1479)
        at akka.jsr166y.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:104)

The problem is that the headers in the response are null which causes the exception to be thrown.

However, the HTTP Response RFC states that the headers are optional and that only the status line and a single CRLF are required.

The RFC is at http://www.w3.org/Protocols/rfc2616/rfc2616-sec6.html

with the relevant Augmented BNF being:

Response = Status-Line
           *(( general-header
           response-header
           entity-header ) CRLF)
           CRLF
           [ message-body ]

where the *((...)) indicates that 0 or more of the elements after the * are permitted.

Therefore it is not in keeping with the standard to throw an exception if there are no headers, simply to use empty headers if there are none.

[slandelle]

@jfarcand I can work on a fix for this. Do you have any objection if I replace everywhere throw IllegalStateException(HEADERS_NOT_COMPUTED) with return null/empty list/ empty FluentCaseInsensitiveStringsMap?

[jfarcand]

@slandelle +1 !

[slandelle]

OK, will do.

[rlubke]

Thanks for the submission!

[slandelle]

You're welcome. Thanks for this great project!

[slandelle]

@jfarcand @rlubke This bug was originally reported against Gatling and I'm about to cut the next release. Do you think it would be possible to release 1.7.9, please? Otherwise, I can ship a custom AHC build.

Cheers,

Stéphane

[slandelle]

I found a bug with what I've done, since I didn't realize that AsyncHttpProviderUtils.parseCharset could return null.
I'll submit a patch in a few minutes, sorry.

[jfarcand]

Salut, no problem, when do you need it? I can do it anytime this week. A+

[slandelle]

I plan to release on Thursday (France time). Could you release tomorrow, please?
Merci beaucoup (les bières vont commencer à s'amasser)!

[jfarcand]

Parfait! I wil cut the release tomorrow morning.