Player is loading...
ScalaWebTest - integration testing made easy
Dani Rey, Unic AG
Thursday, June 13, 2019 · 4:47 p.m. · 42m 17s
ScalaWebTests helps you to set up integration tests in no time and provides useful functions in order to reduce the effort needed to write integration tests for web applications. I will give a short introduction on what an integration test is and why it is important. I will then show you how to use ScalaWebTest and why it makes your life easier. Together we will add ScalaWebTest to a Play application and setup everything needed to write and execute integration tests. I will also cover topics like mocking of thirdparty systems and integration testing in the context of containerized application. Required Knowledge: If you've ever written a web application and posses some basic Scala knowledge, you're prepared to follow this talk. Nowadays Daniel is mostly involved in service integration and web development on the JVM, but he has a strong background as database administrator. He currently works at Unic AG, located in Berne, Switzerland. He takes a huge interest in programming languages and persistence options. He is more opinionated in one area then the other, chat him up to find out which.
Embed
Transcriptions
Note: this content has been automatically generated.
00:00:00
hello everyone i like to start with a short story on
00:00:05
why he creates collect asked some and i'm standing and um
00:00:12
what sort of this so once upon a time
00:00:16
i like is is softer redesign went well
00:00:21
white and and i then test cost something as strange so that code base had
00:00:28
huge amounts of tests so that it was uh uh not just two thousand it was
00:00:33
comparable to do with that amount of code is is a lot so
00:00:39
um what did even redesign so do you have there isn't
00:00:42
and the price yeah mass at the quite complex lincoln things so
00:00:47
what it is um you would have to external as those resource cost you um
00:00:54
you are you are all stiff domains and a tough multiple domains and also it should
00:01:00
um have a featured of users could just copy the address for and then create a link to
00:01:05
fund the c. m. s. and then would internalise the linking creator resource profit of it so that if
00:01:10
some of that we named other resource that no links would be broken
00:01:14
and uh it it also have central asked a asset management but those
00:01:19
assets should appear on different domains depending on where you would use them
00:01:23
and so this was the only because it talks style and
00:01:29
bit confusing so link and link was down the fin template sending back around
00:01:35
shops and in models ending controllers and on the rewriting of pipeline and so on
00:01:41
i mean it they re factoring under the under the above was holding factory very could put
00:01:47
a swing to enriches detect what what's that and we created according model of
00:01:53
an instance of different according type of this and this with and take em
00:01:58
care of uh telling yeah okay you want an external s. version of
00:02:02
me and then internal whatever i can say if you look at this
00:02:09
it's a quite quite hard to to match like
00:02:14
this with that and if we now think of unit tests
00:02:20
how many unit tests could v. t. how many unit tests would help our redesign
00:02:26
not many i mean let's be honest uh templates were not tessa double
00:02:32
if unit test spectrum chops are not really tessa models and folders were tested
00:02:37
and they didn't really know and if the new models so we could get some inspiration
00:02:42
from those tests that we could not take them over we just had to rewrite them
00:02:48
um why did this go well
00:02:51
yeah we had a huge amount of integration tests as well there's
00:02:55
integration tests were not on in the clever but mostly there or
00:02:59
even just a lot of tests came to page and just looking if there's a certain length there and the the look to certain white
00:03:05
there's enough to cover up all the box which he created during the redesign handy
00:03:11
no hand about other features which we should have implemented during the
00:03:15
redesign but we didn't and you could fix that before shipping this so
00:03:22
it's safe ever x. basically and that's when i started
00:03:26
to embrace integration tests and look looking more into this topic
00:03:31
so what they had to even mean if i talk about integration tests
00:03:37
um in the context of web applications that means after deployment
00:03:43
the deployment can be on your c. i. options or on or on your local nope of
00:03:47
so this doesn't have to be a test server this can be somewhere but it has to be
00:03:53
are running complete system with everything and then you got you
00:03:58
got a a system with the browser over h. t. t. p.
00:04:03
that's for me this is an integration s. o. if people
00:04:07
call this and when tests browse it s. u. i. test
00:04:11
now there's the famous test my roommate and it's uh s. that i'm wrong
00:04:19
i said the plan it is wrong so what the parents as is you have to have a lot
00:04:25
of unit asked some service tessa maybe a very few
00:04:28
u. i. test because they're too slow and too cumbersome
00:04:33
i say when designing software is very key to
00:04:37
what we do so we need tests that embrace this
00:04:41
and then we need tests for for like having a good coverage of of a core domain logic so
00:04:48
for core domain logic unit tests are the best
00:04:51
is very easy to create a lot of difference equations
00:04:55
we can also use probably based testing there as a variant of unit testing but still you need
00:05:00
a lot of integration tests and integration testing can even uh i hope you if you
00:05:07
do a complete rebuild if your system you can rebuild the system in a different language
00:05:12
and a different framework they still work because they check with your user expects from you
00:05:17
and now that i'm not internally so i thought what's
00:05:22
next to apply remit to quite close to some camels
00:05:29
and how likely they have two homes a year i think
00:05:34
can say there should be a lot of unit has some of integration tests and if you want to have code quality
00:05:40
you also need a lot of oil tools so you need some static analysis you probably have
00:05:44
some packages servers test you will have your regression tests but never forget to use your brain
00:05:53
if so why do people not like integration tests what larger problems
00:06:02
how they can be tedious to right
00:06:06
that can be difficult to really huge
00:06:11
i can be on reliable and this is this is definitely not
00:06:15
what do you want before test it that's another chico on there
00:06:19
so let's let's have a look at what i mean so this here it's it's got test
00:06:25
if someone yum that's where you get out of the box there and it's very good at
00:06:31
certain stuff so you can easily find
00:06:34
for example single elements and unverified them
00:06:39
nothing to be down there you can also click on stuff from fill out forms it's all fine
00:06:45
but
00:06:47
when you want you check more interest one element that you want to check
00:06:51
attributes an element and being that they're correctly knesset and all that stuff then
00:06:58
it says you can very cumbersome because i guess no one
00:07:01
of you can just immediately see what's going on here so
00:07:08
let's take some inspiration from other fields are they doing in manufacturing
00:07:14
manufacturing you also want to verify that would you create
00:07:19
is correct and i mean just for ball since you use
00:07:26
the list of information which you could weigh five four it's endless
00:07:31
so probably not work from this so that the uses that had this they have gauges
00:07:38
and maybe someone already handed one to you so we can have a look at those so there's the for
00:07:44
example or set the so called collector cage and let it pass yes and asked you if this collector's working
00:07:51
try to fit stuff in and if it fits on the right side then it is too small
00:07:57
and if it fits on the other side then it's good but if it doesn't fit
00:08:02
on the other side and it's too big and that's how you can very easily test this
00:08:08
and i'm o'clock h. works just the opposite way if the rat part fits then it
00:08:13
the whole is too big and if what the pork doesn't fit then it's too small
00:08:19
so that's all the right so can be transformed it somehow
00:08:24
because i was really into gazes testing that was a smooth let's take this so i
00:08:29
can attest h. t. m. l. with h. t. m. l. and in fact we can
00:08:35
so the test which we saw before which probably non
00:08:39
if you could really read that's doing exactly the same thing
00:08:45
so we just say we want our current page to fit this
00:08:50
cage and the cage definition is just some h. t. m. l.
00:08:56
so high step down
00:08:59
okay so on the left side you see an h. t. m. l. document
00:09:03
and on the right side you see all the code needed
00:09:06
to verify this so there's really no line missing here and
00:09:12
what happens when that
00:09:14
so yeah one but he's uh it's not h. m. l. it's dollar
00:09:19
so we use x. m. l. because this is just native to support it
00:09:23
but the cool thing is then you you don't need any additional
00:09:26
frameworks and your ideal even how you making sure that's correct so
00:09:34
what is your grass to do with h. t. m. l. it transforms it to treat them three
00:09:41
what this colour do we comply with x. m. l. transforms it which really
00:09:47
so now they really look alike so let's see what's going on
00:09:51
if we can try to apply this cage if you try to fit
00:09:56
the document to the gauge so the driver now it's you're on the right side
00:10:02
that's a gauge so the first thing the cage cares for its caring for it if
00:10:07
so it's searching the document for a day if it's finding want
00:10:11
sushi for you l. annex searching for an ally below this so
00:10:15
we just go down that cage and you proceed for john to the link
00:10:20
and then mister to control the uh the attributes and what we see as
00:10:25
they're not the same so this is not the element to looking for so we go back
00:10:31
up until we have a sibling which you could take into consideration and they go again down
00:10:38
and the very file the attributes and if they all aligned than this document fits our gauge
00:10:45
that's it will allow now
00:10:50
test should not be brittle and then she will allow us to have a lot of additional
00:10:55
stuff in between which you don't really care for so it can be transformed this to our trees
00:11:04
and then but you can see you know is is a bit to the spurs three on and
00:11:10
if in the cage because there's you well if
00:11:12
no well i below as directed link so um
00:11:18
doesn't make too much sense as h. mobile should just show you how we can work so
00:11:27
from the well which just look for it show element which is a
00:11:31
link and it doesn't matter if this is a child of a child
00:11:34
we can just search down the tree and that's what we do and then we checked the attributes and also here we can see that
00:11:42
we only checked attributes which are part of the cage we don't
00:11:45
care about additional attributes because that's how h. m. l. works you can
00:11:48
have as many attributes as you want there's only certain part which meant
00:11:52
that uh at that point in time so you only better for those
00:11:58
now what about element order
00:12:00
comment or did definitely matters we don't want to mix
00:12:04
elements up because that's have their position on the page
00:12:07
often there your about ranking so there's something you really want to very fine when you to integration tests
00:12:14
so again you go from the right side you
00:12:17
go down and we compare the first two parliaments
00:12:22
and you see that within the document to document is wrong the document search of second
00:12:28
and then this coming first so that that's not the elementary looking for
00:12:34
so we going back up and now when you go again when we go down here
00:12:39
you can see in the cage we still care for the first element we haven't found
00:12:43
anything for the first element so we don't go to the second one is still with the first one
00:12:49
and we found one so that's good so we go to the next l. i.
00:12:54
and this has to be a so one so we find one that's good
00:12:58
and then it very friday order and we can see now that's not good
00:13:03
that's not what we have been looking for so um
00:13:07
this is cowardice part of of the of the framework so let's leave very fortunately more
00:13:18
so um
00:13:20
let's go over here huh
00:13:23
didn't seem right i'll never see ah have to i'm sorry
00:13:29
i didn't check is a um i have to mirror my screen
00:13:36
mm
00:13:42
mm
00:13:54
so sort of for that so nice you should see my screen
00:13:59
i'll just delete everything rent which you have here so that you
00:14:02
can really follow me on the way to what we need so
00:14:07
basically you just sort of a class and then you have to extend
00:14:13
one of my penis traits and and
00:14:19
usually you would extend the fiscal attested
00:14:21
maybe extent flats that free spec and you know it's just integration free stick
00:14:28
so not much different the next thing you have to provide is
00:14:33
you have to tell me what you base your i should be
00:14:37
because you have to know where to fetch the document
00:14:43
and then another thing which you have to provide is the power off because we have to fetch
00:14:49
that from from uh from up off so we separates the base your
00:14:53
eye from the pop off because usually you want to be able to just
00:14:58
be changed to base your eye so you can maybe test your production system
00:15:03
so i'm thinking of us go a a rest home perch
00:15:10
and now
00:15:12
we can start out with test um and that's missing so that's it
00:15:20
so let's go to this web page you want to wire phi something from
00:15:27
you so what you can do is just take to inspect your goes somewhere
00:15:34
we say we care for this news item here and then it just saying you carpool
00:15:41
after h. t. m. l. and you go back here and you say uh current page
00:15:48
fits and you could pretty much more here and you say so this is
00:15:53
a it's a new site or something so that's it and we run it
00:16:12
program or something then the line um
00:16:20
i guess it's something with this text here well let's just remove that one
00:16:29
i guess it's something about being multi line ah
00:16:35
it should haves uh not to to specify maybe but different things of just sending it hoping it over
00:16:41
so would you can also do is um for example here we can say okay this should
00:16:48
actually contains cages and so you can only have
00:16:54
parts of it and you can also use reg x.
00:16:57
and you can use that for text and you can use that
00:17:01
for um uh attributes as well and if you if you break it
00:17:08
then you get some nice reports also on this
00:17:12
uh_huh
00:17:19
no oh yeah of course if a half contains then uh it it's still contain that
00:17:29
okay
00:17:36
and there you can see that it's uh s.
00:17:39
and texas not fitting and it'll than disc right um
00:17:45
that simple catches didn't equal oh this h.'s whatever
00:17:52
and um yeah so another thing which you
00:17:56
can do is um see there's an element
00:18:04
which objects but um i think that's also quite interesting because often times
00:18:10
you just you do not want to try this on the whole page
00:18:14
you want to try this on a certain element maybe because you just um
00:18:19
mm spawn something which would create a certain element or which
00:18:23
should be modified so that you can do is you can reuse
00:18:28
um so you can you can use that value very can
00:18:31
use find all from from scott tests to find images here
00:18:36
and then you have again h. which is just um putting it into a variable and
00:18:41
then use for all so use for all image in images the image should fit that cage
00:18:49
and every but whenever you do that make sure that you verify that
00:18:53
you received the correct amount of images because if that's empty then although fit
00:19:02
it's uh
00:19:05
uh_huh oh that's not huh okay let's right so that was my back and
00:19:13
said yeah other thing which probably a lot of you care for is jason
00:19:18
that's to those publications is angela chase them basically so um b.
00:19:25
we also have try some gauges and third quite similar to
00:19:29
to h. m. o. k. which is but a lot simpler um
00:19:34
just make sure you to write files uh
00:19:39
um because with choice on you cannot
00:19:43
have intermediate um elements to hire keep matters
00:19:47
so we know exactly which level to expect about so that's a lot easier to test
00:19:52
and nevertheless i mean there's good means to where if i have any
00:19:57
of those properties but it takes more time so rather half this lazy test
00:20:05
i mean there's no that that's this is not the clean solution
00:20:08
to televise things are going wrong i'd rather have this uh nothing
00:20:13
because that's often what's what's going on under pressure it's this or nothing
00:20:19
so which is copied in one example resort and make sure it's there
00:20:26
and um we can also see um so i think for
00:20:32
no
00:20:34
i can i can i can tell you get so i think for tess it's usually more interesting to see them fail then
00:20:40
to see them just having green so um what's most missed
00:20:45
all the extra uh_huh and yeah put which you will then get
00:20:51
and
00:20:54
something like this like so the minutes equal the chaney x.
00:20:59
or something so um so they provide also meaningful output if
00:21:06
they fail but they don't provide unfortunately because you can't really
00:21:10
is context by this certain property was wrong but still um
00:21:17
so usually have to combine them with more explicit tests for single properties
00:21:23
and the other thing which we have is the
00:21:28
alright
00:21:31
so what you can do here is instead of providing values
00:21:37
you just remove all the values and your targets to just check for the types
00:21:42
so this is very useful if you have was amounts of what you have each listen you're
00:21:47
just to make sure that every element it almost has all attributes which are expected the correct type
00:21:52
you could use schemas for that but is a lot easier to write that much just jason scheme
00:21:58
if you are now okay jason scheme that's no fun to write you do you can generate it or you probably not to so
00:22:07
that's our way to do it
00:22:09
and there's some other options so you can decide if you want to care for a right order
00:22:14
or not and if you want to cry for size of or else uh not a a so
00:22:22
let's have a look at the concepts what target
00:22:24
concepts the hats got its license colour test and selenium
00:22:30
it takes care of the life cycle so that's where you had to you
00:22:34
problem when i had to provide to the base you're right the parser can
00:22:38
look this stuff up for us we don't have to care for days uprights
00:22:42
you defaults for for the browser for example and promotes convenience functions to access stuff
00:22:48
and it follows decoding style of stock test so if you want
00:22:52
to other feature just probably trade for that which you can extend
00:22:57
nat testing website cool um
00:23:01
as with every testing framework so that's what he just inherit from scott has fifty for
00:23:06
all before each and then you have your tests and then after each and so on
00:23:11
but we do is really do you two things in the before all
00:23:17
but you faster go back for your
00:23:21
so go one back for the so um so
00:23:26
in the before all there's no need to prepare the web driver so especially riff cool
00:23:32
this is important because crawl masses a service which you want to start
00:23:36
just ones so this your test sufficient so that's what we do here and
00:23:42
yeah we make sure that it's raining it's good although it quite
00:23:45
configuration then you provided hook for the user to do stuff before logan
00:23:51
then in tennis you half looking stuff um defined
00:23:55
the dual okay so before we did the law can be applied the
00:23:58
configuration which is used for the log ins or slogan convict reapply that
00:24:03
we did a lot can then be applied the normal convict and then we handle back to the user with the off to looking oh okay
00:24:10
and then the before each and then every test is going on then before each test
00:24:16
we navigated what y'all which you it's basically a combination of the base your hand
00:24:22
the prof we navigate they're just make sure the browser is sitting there if it
00:24:26
if it went away because you clicked on stuff read during one test we go back to this
00:24:31
and if you didn't quite can stuff you still there so you don't
00:24:35
make it there so you can speed up test because this is important
00:24:39
and i can tweak that because maybe i you don't change the page but you still
00:24:43
have side effecting stuff so you want to every for so you can configure that's no problem
00:24:51
browser options the fall this actual unit this is j. v. m. biased
00:24:55
um nothing to install on your computer does works you can on the website you can
00:25:01
uh click on links they go to discuss the rounds in circuses so many nothing need it
00:25:07
um the channel script support is a bit mediocre um so if you have transcript heavy
00:25:15
stuff h. wall unit is probably not your friend get power is it exposes h. t.
00:25:20
t. p. status codes response either something so many um does not so if you have
00:25:25
a channel unit we can extend a an excess of trying to get access to this stuff
00:25:31
if you use chrome which is another option you don't get to use that
00:25:36
i'm really support every so i'm up driver
00:25:40
but you've both think wrong um we tried the same with firefox there we
00:25:46
currently have some issues and travis it's travis related not that not test framework related
00:25:52
uh if you can't possibly will be part of a framework as
00:25:54
well but i don't want to ship stuff which is not rocks all
00:26:00
and yep all requests are are are of all can also
00:26:03
for browsers but uh please help me make that run on travis
00:26:08
so play framework let's let's have a look at how you do at this too
00:26:14
something to work framework so um play it's quite simple
00:26:20
uh usually you have this um sort of just enable
00:26:24
parkinson they just have to uh that uh integration test conflicts
00:26:29
unlike integration test complex basically just unheard stuff from
00:26:32
test and you have you been selected for integration tests
00:26:37
so this is uh oh one in important part uh you can
00:26:41
find this all on and get help that's part of the project
00:26:44
you have an example subfolder we can find all this stuff also
00:26:48
before read me describing how to use this time you need one dependency
00:26:53
should be or c. to buy now um you can add some additional depends if
00:27:01
you want to make sure that you you don't have like mismatches with with others but
00:27:06
they would otherwise speech and stiffly be pulled in so it just depends on how you work
00:27:12
and this is the whole script that the the ten important is the stage your play
00:27:18
framework so that you get this um said farm of of of the compiled code in
00:27:25
and then we just are run this we have to provide
00:27:29
the secret and then we can from s. b. t. can say
00:27:33
s. p. t. i. t. tassel around all our tests and it
00:27:35
finished except was just stopping play again that's all you don't anymore
00:27:43
now we'll divide external dependencies so in
00:27:47
when a now external depends is what i
00:27:49
mean it's external dependencies for your about publication so
00:27:54
this is a p. s. you call it
00:27:57
opposes to use because this matters for integration testing
00:28:03
because stable test so she so you have to bring
00:28:07
those external dependencies on the control when you do testing
00:28:14
and i think uh i truly believe there's just one way
00:28:17
to do that and that's more king all your external dependencies
00:28:22
um that might seem like a lot of work this is the low birth weight not part of scroll over test
00:28:27
because there are already many good frameworks and there should
00:28:31
match your problem so we can not provide this for you
00:28:36
and so if you want to more can help yeah i uh i i can
00:28:40
suggest why mark i think mark server looks also very good i'm never use this
00:28:46
and if you have craft you'll there's fate feel that you could
00:28:49
use and also an option to always consider is a small but framework
00:28:54
because of the small that framework you thing again can code those responses
00:29:00
and maybe you can even have a mix and match so some responses you just pretty record
00:29:05
and replace it with something like where mark and for other stuff you use your but framework
00:29:12
now what's most doing all this stuff so if you want
00:29:17
to look this up this call the test harness a test harness
00:29:22
has the responsibility to start and stop your marks and face it starts and
00:29:26
stops your application on the test to make sure the applications that is correct
00:29:31
make sure your mark and fakes to half the correct state
00:29:36
and this can often be done if if you lines of code
00:29:40
within your tests so this can just be part of your integration tests
00:29:47
sometimes combined with a bit of fashion or something and so
00:29:52
yeah one important tracts you you can't uh at test
00:29:56
harness running is team you like all the object trick
00:30:00
'cause if you have an object and you import this as part of your
00:30:05
by straight or class which you use from all your past
00:30:10
then this will always be executed before the first test so that's very useful because
00:30:17
um what i usually do in in this is um
00:30:21
i switch my application to the to the market place
00:30:25
it's usually like to double up against a real view of mineral my tests i want to run against the marks
00:30:31
and i don't want to fill this by hand or something and because of this trait can it's always running it doesn't matter if
00:30:37
used to just start a single test the from your i. d. e. or if you start all your test it will always be forced
00:30:44
and thanks to the shot unhook it can also clean up after itself
00:30:49
at least as long as you don't collect is we call the process it cannot quite now
00:30:56
another thing which i consider very important is adjusting dates and times if in
00:31:01
your market data so you record your market data at a certain point in time
00:31:06
record to write down when you did record that because this gives
00:31:10
you the ability to calculate the difference between now and that reference time
00:31:16
and then you can have an adjuster and then you can use this
00:31:19
to just although it's because oftentimes
00:31:22
things like last modified or created it's
00:31:28
they are of relevant to have friends appear or not if they're shown or not
00:31:34
and maybe you need some additional fiddling there maybe you have
00:31:37
to make sure that you don't just birthday it's an so
00:31:42
that's just something you which is really depending on you too may have you sort of that
00:31:49
i know there are contain arrest applications so that's like seems to be the way everyone is
00:31:56
doing it now that contain rice everything and and just put it on companies and everything is good
00:32:02
yeah so it depends um if you just feel the containers part if your normal built and
00:32:09
you have local build tools like s. d. t. and all the machines or miller you name it
00:32:15
then just makes colour test part of your normal build as we just seen with the play framework that's
00:32:23
nothing special need and then you can even i mean you can access than the container from that's fine
00:32:30
if you blow out uh if you do build a fan a container and
00:32:34
no one has still tools on their machines because they're part of the container
00:32:38
uh no not you you need a container to build your tests as well
00:32:44
so um for that's it way she we do have a solution um so
00:32:51
use the doc file i will not go into details but
00:32:54
the cool thing is cycle provided us gillespie t. doc file
00:32:59
and uh um based on this the problem that we're on
00:33:03
doctor container which you can just got from doctor hop which and
00:33:10
let's see yeah that's the right one so you just for why it it's the volume
00:33:16
our um which provides a test source and then you can just start it
00:33:21
and they will start as dirty little compiler sources the little around that
00:33:26
and if you want to keep that session open just change to
00:33:29
command from from s. p. t. test s. p. t. and you're done
00:33:34
because then you will just say within the s. p.
00:33:36
t. session and you can just reiterate with a running tests
00:33:42
and as i said before i like failing test because i'm more interesting so
00:33:47
you make the one fail as well you can see the applet here
00:33:51
um would you probably would want to do a small if we have something
00:33:55
like this is also use a volume couldn't get a test output which
00:33:59
was written down out of the container again um now the other situation is
00:34:08
if you already have that can we not know the container which
00:34:12
contains pretty compile out tests which you can then ship reaching and then
00:34:17
run the cancer production or stuff like that so so why not
00:34:21
because you can pack tests and you can tack tests which do not
00:34:25
you can tell like those tests on change state and they don't depend on data so that
00:34:31
then you can run them as we question tests which is just the party for cassie
00:34:35
why not do that so they stop profiles the more complex and here we can not
00:34:41
it cannot ship your container to do that because isn't this contains your composite tests so you have
00:34:49
to wrap that taco far from our repository maybe it up to the bit and and we use
00:34:55
basically when it does it composite and then it's um as it's a multi search boats open
00:35:00
again sets of uh from alpine next and ten different most uh you have to test char
00:35:08
and then it uses this contest runner to run your tests
00:35:13
so if you look at uh how this is going um
00:35:23
so those um doctor just verify the nothing changed so nothing to do and you
00:35:28
can see that it's also all up fast because we just have to run the test
00:35:32
and we can also provide an environment variable just change for example the base your eye
00:35:39
and of course it's falling all over the place because it's completely wrong a u. r. l.
00:35:44
and of course is not all match our gauge but this is how you can then
00:35:49
provide information into that container so um all the configurations um
00:35:56
but let's say like the ones which i think
00:35:59
are necessary can be configured with environment variables around arguments
00:36:09
yeah so we just saw that so let's let's have a look at some some take away messages
00:36:17
so one thing which i consider right important it's make ups verbal what isn't
00:36:23
is often times you can not write an integration test
00:36:26
because you cannot observe what you would like to test
00:36:30
but if that's the case probably you're also a bit wind during
00:36:35
uh when you execute your application and in productions of make that stuff experience uh
00:36:41
uh observable expose it protected also ah remember about this if you make stuff observable
00:36:48
but that's a good way to do and uh usually just
00:36:50
hopped our applications also locked do not modify on important details
00:36:58
control your application state
00:37:02
and mark your external dependencies
00:37:05
and uh our last on a loose integration tests are your friends
00:37:11
so there's a new release 'em score baptists rio
00:37:15
it's around the corner and just waiting for the final release of why try some
00:37:21
and uh it supports two thirteen two twelve into heaven it supports any sullen younger driver
00:37:27
and that's stock from this we just saw because you can configure it if environment but it off
00:37:33
so thank here and if you want to become my future comic you're hiring
00:37:41
so i'm happy to take questions
00:37:45
hi
00:37:50
hi
00:38:00
you'll attend will the uh and uh
00:38:05
oppose oppose the for him would you then take a c.
00:38:09
email morris there are also some simple but yes so um
00:38:17
let's delete your way around selenium is part of scott baptist
00:38:22
so i you would you would use like if you don't really even
00:38:27
realise what's going on if you if you go because it's it's seamless
00:38:33
it's the same in him but browser reaches control where it sets so what you would do receiver
00:38:38
to use how it's document that on this colour
00:38:42
test website how to interactive selenium riffs colour test
00:38:47
and there you fill out the form says so you really use
00:38:50
what what's provided by scott test for then you sent the form
00:38:55
and then usually after sending the form that's when you can use the gauges
00:39:01
and what i would also recommend this that you use a cage to well okay to form
00:39:06
it's not because same test but if you're if you're for test fails it's nice to know that why
00:39:12
it failed and and the coach can help you with this because it will tell you it's missing that field
00:39:19
while i'm filling out the forms and just uh it didn't find out on and that's maybe a bit more obscure
00:39:27
hope that answers your question
00:39:32
yeah it did you also test sums in based on and or something and a new page application of too
00:39:38
late to the javascript to finish and all this stuff yes yes so that's a that's a very good question
00:39:44
and that's a nice the support that thanks to scott tests uh eventually so
00:39:50
pro uh basically nothing you have to do they're just put it in turn eventually and all just checking checking check
00:39:56
and as soon as it's green it's green and if it isn't a complete different a certain amount of time that's right
00:40:03
so that works very well but you have to remember every
00:40:06
time you interactive stuff it's going on it's got and javascript
00:40:11
you have to put in an intern eventually sometimes you got lucky but shouldn't but on that
00:40:20
maybe on how would you how would he uses it it had a testicle work flow
00:40:25
like you have to hit model and points lying and you have to pass you know
00:40:30
how about a look um so i would i would approach it the same way as we did with selenium
00:40:39
but just have immediate okay just to verify if what you see there is what you would expect
00:40:46
so if you had like it's like this so you i like on like uh for me had like
00:40:50
three steps and uh_huh so you get us the first episode you test the third set you have to
00:40:55
pass through the first two steps place you have little like three different tests
00:41:00
testing like step one step two and walls that one in wanted to it and it is true and
00:41:06
that now now i think i see where you're going
00:41:08
the cool thing about scott test us it is ordered
00:41:14
so you can still have separate tests but within the same class
00:41:18
and so what this nifty gives you better reporting or where it
00:41:22
where it failed you don't have to look into detail to seaward furrowed
00:41:27
um you don't have to look at the exception message would you require c. which test failed
00:41:32
of course all tests which can later on will most probably fail but
00:41:37
i mean it doesn't matter anyway half to fix something and you see which was the first step
00:41:42
that fails so it make one class and i just ordered oh steps a separate test thank you
00:41:58
so it seems like they're no more questions um if
00:42:02
if you have questions i'm i'm here at uh at um
00:42:06
the till the end of the day and um otherwise which
00:42:09
we helped conditioned trailer or you can reach me by mail
Conference Program
A Tour of Scala 3
Martin Odersky, Professor EPFL, Co-founder Lightbend
June 11, 2019 · 5:15 p.m.
8335 views
A story of unification: from Apache Spark to MLflow
Reynold Xin, Databricks
June 12, 2019 · 9:15 a.m.
1267 views
In Types We Trust
Bill Venners, Artima, Inc
June 12, 2019 · 10:15 a.m.
1569 views
Creating Native iOS and Android Apps in Scala without tears
Zahari Dichev, Bullet.io
June 12, 2019 · 10:16 a.m.
2231 views
Techniques for Teaching Scala
Noel Welsh, Inner Product and Underscore
June 12, 2019 · 10:17 a.m.
1295 views
Future-proofing Scala: the TASTY intermediate representation
Guillaume Martres, student at EPFL
June 12, 2019 · 10:18 a.m.
1156 views
Metals: rich code editing for Scala in VS Code, Vim, Emacs and beyond
Ólafur Páll Geirsson, Scala Center
June 12, 2019 · 11:15 a.m.
4695 views
Akka Streams to the Extreme
Heiko Seeberger, independent consultant
June 12, 2019 · 11:16 a.m.
1552 views
Scala First: Lessons from 3 student generations
Bjorn Regnell, Lund Univ., Sweden.
June 12, 2019 · 11:17 a.m.
577 views
Cellular Automata: How to become an artist with a few lines
Maciej Gorywoda, Wire, Berlin
June 12, 2019 · 11:18 a.m.
386 views
Why Netflix ❤'s Scala for Machine Learning
Jeremy Smith & Aish, Netflix
June 12, 2019 · 12:15 p.m.
5026 views
Massively Parallel Distributed Scala Compilation... And You!
Stu Hood, Twitter
June 12, 2019 · 12:16 p.m.
958 views
Polymorphism in Scala
Petra Bierleutgeb
June 12, 2019 · 12:17 p.m.
1113 views
sbt core concepts
Eugene Yokota, Scala Team at Lightbend
June 12, 2019 · 12:18 p.m.
1655 views
Double your performance: Scala's missing optimizing compiler
Li Haoyi, author Ammonite, Mill, FastParse, uPickle, and many more.
June 12, 2019 · 2:30 p.m.
837 views
Making Our Future Better
Viktor Klang, Lightbend
June 12, 2019 · 2:31 p.m.
1682 views
Testing in the postapocalyptic future
Daniel Westheide, INNOQ
June 12, 2019 · 2:32 p.m.
498 views
Context Buddy: the tool that knows your code better than you
Krzysztof Romanowski, sphere.it conference
June 12, 2019 · 2:33 p.m.
393 views
The Shape(less) of Type Class Derivation in Scala 3
Miles Sabin, Underscore Consulting
June 12, 2019 · 3:30 p.m.
2321 views
Refactor all the things!
Daniela Sfregola, organizer of the London Scala User Group meetup
June 12, 2019 · 3:31 p.m.
514 views
Integrating Developer Experiences - Build Server Protocol
Justin Kaeser, IntelliJ Scala
June 12, 2019 · 3:32 p.m.
551 views
Managing an Akka Cluster on Kubernetes
Markus Jura, MOIA
June 12, 2019 · 3:33 p.m.
735 views
Serverless Scala - Functions as SuperDuperMicroServices
Josh Suereth, Donna Malayeri & James Ward, Author of Scala In Depth; Google ; Google
June 12, 2019 · 4:45 p.m.
936 views
How are we going to migrate to Scala 3.0, aka Dotty?
Lukas Rytz, Lightbend
June 12, 2019 · 4:46 p.m.
709 views
Concurrent programming in 2019: Akka, Monix or ZIO?
Adam Warski, co-founders of SoftwareMill
June 12, 2019 · 4:47 p.m.
1974 views
ScalaJS and Typescript: an unlikely romance
Jeremy Hughes, Lightbend
June 12, 2019 · 4:48 p.m.
1377 views
Pure Functional Database Programming‚ without JDBC
Rob Norris
June 12, 2019 · 5:45 p.m.
6374 views
Why you need to be reviewing open source code
Gris Cuevas Zambrano & Holden Karau, Google Cloud;
June 12, 2019 · 5:46 p.m.
484 views
Develop seamless web services with Mu
Oli Makhasoeva, 47 Degrees
June 12, 2019 · 5:47 p.m.
785 views
Implementing the Scala 2.13 collections
Stefan Zeiger, Lightbend
June 12, 2019 · 5:48 p.m.
810 views
Introduction to day 2
June 13, 2019 · 9:10 a.m.
250 views
Sustaining open source digital infrastructure
Bogdan Vasilescu, Assistant Professor at Carnegie Mellon University's School of Computer Science, USA
June 13, 2019 · 9:16 a.m.
374 views
Building a Better Scala Community
Kelley Robinson, Developer Evangelist at Twilio
June 13, 2019 · 10:15 a.m.
245 views
Run Scala Faster with GraalVM on any Platform
Vojin Jovanovic, Oracle
June 13, 2019 · 10:16 a.m.
1340 views
ScalaClean - full program static analysis at scale
Rory Graves
June 13, 2019 · 10:17 a.m.
463 views
Flare & Lantern: Accelerators for Spark and Deep Learning
Tiark Rompf, Assistant Professor at Purdue University
June 13, 2019 · 10:18 a.m.
380 views
Metaprogramming in Dotty
Nicolas Stucki, Ph.D. student at LAMP
June 13, 2019 · 11:15 a.m.
1250 views
Fast, Simple Concurrency with Scala Native
Richard Whaling, data engineer based in Chicago
June 13, 2019 · 11:16 a.m.
624 views
Pick your number type with Spire
Denis Rosset, postdoctoral researcher at Perimeter Institute
June 13, 2019 · 11:17 a.m.
245 views
Scala.js and WebAssembly, a tale of the dangers of the sea
Sébastien Doeraene, Executive director of the Scala Center
June 13, 2019 · 11:18 a.m.
661 views
Performance tuning Twitter services with Graal and ML
Chris Thalinger, Twitter
June 13, 2019 · 12:15 p.m.
2003 views
Supporting the Scala Ecosystem: Stories from the Line
Justin Pihony, Lightbend
June 13, 2019 · 12:16 p.m.
163 views
Compiling to preserve our privacy
Manohar Jonnalagedda and Jakob Odersky, Inpher
June 13, 2019 · 12:17 p.m.
301 views
Building Scala with Bazel
Natan Silnitsky, wix.com
June 13, 2019 · 12:18 p.m.
565 views
Diversity and Inclusion Panel: Bring the Thunder!
Panel
June 13, 2019 · 2:30 p.m.
244 views
Asynchronous streams in direct style with and without macros
Philipp Haller, KTH Royal Institute of Technology in Stockholm
June 13, 2019 · 3:45 p.m.
304 views
Interactive Computing with Jupyter and Almond
Sören Brunk, USU Software AG
June 13, 2019 · 3:46 p.m.
681 views
Scala best practices I wish someone'd told me about
Nicolas Rinaudo, CTO of Besedo
June 13, 2019 · 3:47 p.m.
2702 views
High performance Privacy By Design using Matryoshka & Spark
Wiem Zine El Abidine and Olivier Girardot, Scala Backend Developer at MOIA / co-founder of Lateral Thoughts
June 13, 2019 · 3:48 p.m.
753 views
Immutable Sequential Maps – Keeping order while hashed
Odd Möller
June 13, 2019 · 4:45 p.m.
276 views
All the fancy things flexible dependency management can do
Alexandre Archambault, engineer at the Scala Center
June 13, 2019 · 4:46 p.m.
389 views
ScalaWebTest - integration testing made easy
Dani Rey, Unic AG
June 13, 2019 · 4:47 p.m.
468 views
Mellite: An Integrated Development Environment for Sound
Hanns Holger Rutz, Institute of Electronic Music and Acoustics (IEM), Graz
June 13, 2019 · 4:48 p.m.
213 views
Closing panel
Panel
June 13, 2019 · 5:54 p.m.
400 views