[28s] so
[42s] do
[1m 38s] so
[2m 6s] so
[2m 34s] do
[6m 18s] hello
[6m 22s] can anyone can anyone hear me
[6m 32s] absolutely no sound or only luke again
[6m 51s] hey daniel hey super hey steve
[6m 55s] and cool that you can hear me
[7m 2s] hello oh yeah i think hello
[7m 6s] all right so cool said you were online
[7m 9s] for seven minutes without any audio
[7m 12s] well to be fair we were playing a video
[7m 14s] before i think that worked
[7m 16s] okay hopefully no terminator voice
[7m 21s] yeah uh marion is absolutely right you
[7m 24s] should write a pest test
[7m 25s] before stream that checks if your
[7m 27s] microphone works
[7m 30s] that's the next level isn't it
[7m 33s] yes steve you're uh you're a thief
[7m 37s] how are you this evening tom what
[7m 41s] how are you i'm fine uh
[7m 44s] stressy day but uh still fine so
[7m 53s] yeah i had to eat something because it
[7m 54s] was my first meal today
[7m 56s] so that's why i'm a bit late no worries
[8m 1s] i was only sorting my audio out anyway
[8m 2s] so you've missed nothing
[8m 6s] your audio is coupled to to my presence
[8m 9s] in your stream
[8m 10s] that's it exactly i'm forever tied and
[8m 13s] indebted to you
[8m 15s] yeah all right let's jump in
[8m 19s] so we've had some some goodies added to
[8m 23s] pest
[8m 23s] since we last played with the stream two
[8m 25s] weeks ago it's amazing
[8m 27s] how quickly time flies really um
[8m 31s] so i kind of wanted to start by going
[8m 33s] over the tests that we
[8m 35s] started writing last week and see if we
[8m 36s] can chuck a few of these goodies in
[8m 40s] because i think so you've added
[8m 44s] a lot of things to pests since then so
[8m 51s] you are only noise
[8m 56s] i have absolutely no idea what you are
[8m 58s] doing sets
[9m 8s] but steve if your daughter is still
[9m 10s] watching the hey you
[9m 29s] should we probably just have a phone
[9m 30s] call and i
[9m 32s] put you on on loudspeaker and
[9m 36s] everyone hears you via my microphone
[9m 47s] steve tell me what have you done said it
[9m 50s] worked yesterday
[9m 51s] for you and luke so
[9m 57s] said we can repeat it
[10m] yeah three four
[10m 3s] that was um i don't know what's going on
[10m 6s] because
[10m 9s] actually i worked yesterday right worked
[10m 11s] on the street worked on the past meetup
[10m 13s] it worked yesterday maybe it's something
[10m 14s] to do with when i host uh
[10m 17s] when i host or maybe restream
[10m 21s] yeah maybe let's blame restream steve
[10m 23s] just for the fun
[10m 25s] we could also blame wednesday or
[10m 34s] all right i'm gonna crack on you just
[10m 36s] like say something if my
[10m 38s] microphone stops working and i just have
[10m 40s] to unplug and replug again
[10m 42s] yeah um so
[10m 46s] all things added to pest first of all i
[10m 49s] wanted to turn
[10m 49s] this test that we did last time into a
[10m 52s] higher order test
[10m 56s] um first of all we already refactored
[10m 59s] this right
[11m] so i'm pretty sure we can say now this
[11m 3s] user
[11m 7s] and when we do that we obviously get the
[11m 12s] just a bit of a nicer syntax there let's
[11m 15s] see if that works
[11m 21s] it does um
[11m 25s] and basically we only use this once
[11m 27s] right and then the only other time we
[11m 29s] use it is to refresh
[11m 31s] yeah so
[11m 35s] perhaps if we were to inline
[11m 41s] this how would we go about doing it
[11m 44s] close this off first of all we need to
[11m 48s] maybe tap right the tap is a new helper
[11m 52s] that was added to
[11m 55s] um just over a week ago
[11m 59s] basically i think the pull request was
[12m 2s] opened
[12m 3s] during our stream but it wasn't watched
[12m 5s] yet yes
[12m 7s] um i think it did have a couple of
[12m 10s] updates so which you'll see
[12m 11s] in a moment um after the stream
[12m 15s] so there's some ideas that came from
[12m 16s] that stream that kind of went into it as
[12m 18s] well but basically what tap allows you
[12m 19s] to do is
[12m 21s] dive into the test case to do something
[12m 24s] to perform some action right but then
[12m 26s] have the
[12m 27s] test case returned to you to carry on
[12m 29s] your higher order expectation chain
[12m 32s] or your higher order test chain i should
[12m 33s] say um
[12m 36s] so in this case we want to do this
[12m 38s] action
[12m 41s] and i think we probably want to set the
[12m 43s] user
[12m 45s] to this user
[12m 48s] so i'm just setting a property
[12m 52s] on the test object itself here
[12m 55s] called user which will be equal to an
[12m 57s] instance of a github user
[12m 59s] and then obviously dispatching the job
[13m 1s] there
[13m 4s] let's bring all of this out and get rid
[13m 6s] of this at the bottom
[13m 9s] and we
[13m 12s] have one two you know one few uh
[13m 16s] uh bracket that we need there we go so
[13m 19s] this is our higher
[13m 20s] order test um obviously we've got a
[13m 23s] little bit of a syntax problem here
[13m 24s] which
[13m 25s] we can fix with the other thing that was
[13m 27s] in that same pr
[13m 29s] which is passing a callable or a closure
[13m 32s] to the expect method
[13m 38s] and that will just become this user
[13m 41s] refresh
[13m 42s] because we want to access this object
[13m 44s] here right
[13m 48s] other than that let's run this test
[13m 51s] again
[14m 1s] ah there we go happy days
[14m 4s] so yeah that's that's the higher order
[14m 6s] version of the test we wrote last week
[14m 9s] so we are done for today and can go
[14m 11s] offline that's it
[14m 12s] thanks guys i i really enjoyed
[14m 16s] really enjoyed having you so that was
[14m 19s] that's pretty cool and i did
[14m 20s] i did kind of perhaps explain these two
[14m 23s] methods in a bit more detail on the past
[14m 25s] meetup
[14m 26s] so if you've not watched that yet go
[14m 27s] check that out um
[14m 31s] thought about that is we can remove that
[14m 32s] use soon as we never actually deal with
[14m 34s] the user model anymore
[14m 37s] and that's that's that test done isn't
[14m 40s] it let's move on to our user test
[14m 42s] because there are some other sweet
[14m 44s] treats from
[14m 46s] uh since two weeks ago right so again we
[14m 50s] can higher order test here
[14m 52s] and that's as simple as making this
[14m 56s] expect a chained method
[15m] and passing a closure in because we're
[15m 3s] obviously accessing the test case
[15m 4s] anytime you access the test case you
[15m 5s] need to pass a closure
[15m 7s] in higher order tests and i think we're
[15m 11s] done there right
[15m 12s] so test see if it passes
[15m 17s] it does i think steve uh
[15m 20s] asked a pretty good question how you
[15m 23s] would use things like bus fake you fake
[15m 26s] mail fake notification fake and all of
[15m 28s] these
[15m 29s] level fake things uh would you use the
[15m 32s] tap method
[15m 33s] or um i think we actually
[15m 36s] so we did touch on this very very
[15m 38s] slightly in steve's stream yesterday
[15m 42s] i i think if you're doing something over
[15m 44s] and over on each test you should
[15m 45s] probably
[15m 46s] move it into the test case as a method
[15m 49s] on the test case because you can call
[15m 51s] any method on the test case
[15m 53s] in higher order tests so anything
[15m 55s] available on the test case is available
[15m 56s] in a higher order test
[15m 58s] um i think tap is more for those
[16m 1s] times where you it's a one-time use
[16m 3s] right you don't need it in any other
[16m 4s] test
[16m 5s] you need to fake some end point and once
[16m 7s] you've faked it that's it you're not
[16m 9s] going to be doing anything else in
[16m 10s] another test with that same data
[16m 14s] so just to go back to this file here
[16m 18s] if we end up doing this over and over
[16m 20s] again
[16m 21s] it probably will be an indication that
[16m 25s] we need to
[16m 26s] to move the functionality into the test
[16m 27s] case itself rather than
[16m 29s] have it in a tap but we'll see
[16m 34s] we'll see okay
[16m 37s] i i do need an m1 steve yeah i'm i'm
[16m 41s] kind of holding off
[16m 41s] until the next one's come out but i
[16m 43s] guess i could be saying that forever
[16m 44s] right
[16m 45s] but i also have to get it past the wife
[16m 50s] something i it i think it's only
[16m 54s] like an air layers for tap but something
[16m 57s] i would really love
[16m 58s] to see is like a prepare method or
[17m 1s] something like this
[17m 2s] or before set up however you want to
[17m 5s] name it but
[17m 7s] like and not
[17m 11s] not the surreal setup method you have in
[17m 13s] php units that runs for
[17m 15s] all cases in this class
[17m 18s] but to to do some preparation
[17m 21s] for this specific test
[17m 24s] so like for example
[17m 27s] if you go back to the first test case we
[17m 31s] um defining this uh cis users this user
[17m 35s] coming beer
[17m 36s] it doesn't feel right to do it in the
[17m 39s] tab method
[17m 40s] yeah because it's some preparation
[17m 44s] that's not part of the test logic in
[17m 47s] that case but it's
[17m 49s] it has to be done to be able to run the
[17m 53s] test
[17m 54s] so i know this before each but it could
[17m 57s] be something like that yeah
[17m 58s] i certainly see what you're saying i i
[18m] agree to be fair
[18m 2s] um if we would add a second test case
[18m 4s] here that shouldn't use gummy bear but
[18m 7s] uh luke for example yeah we would have
[18m 10s] to do something else
[18m 11s] it would be specul to have a
[18m 14s] attainable setup before prepare whatever
[18m 18s] method
[18m 19s] so it can sets does
[18m 22s] exactly the same like the tab method i
[18m 24s] think but
[18m 26s] one step before i mean i guess you could
[18m 29s] you can double tap yeah
[18m 32s] so you could do something like that
[18m 39s] but because te past is all about this
[18m 41s] readability
[18m 42s] and yeah so on yeah i don't care as a
[18m 46s] user even if it's just
[18m 49s] putting into the same stack with an
[18m 51s] array
[18m 53s] array not push but opposite to push it
[18m 57s] to the front
[18m 58s] uh i don't care how it works behind the
[19m 1s] scenes but
[19m 2s] having an explicit before method
[19m 5s] yeah yeah would make it much more
[19m 7s] readable
[19m 8s] definitely definitely and i tell you
[19m 11s] what we'll do actually
[19m 12s] i'm going to leave this like this for
[19m 13s] now maybe come back and refactor that in
[19m 16s] a bit
[19m 16s] in a bit but there's something that kind
[19m 18s] of touches on this point of
[19m 20s] wanting to add things to uh pest
[19m 25s] that you can actually do in your
[19m 26s] projects and it's really cool and it's
[19m 28s] it's a little bit underutilized in my
[19m 29s] opinion
[19m 30s] um so we'll i'm actually tell you what
[19m 32s] let's look at that now right
[19m 34s] um it's these postgres sql tests right
[19m 38s] yeah at the moment we require
[19m 41s] postgresql to run these tests and all
[19m 43s] that's doing under the hood
[19m 45s] is it checks if the current database
[19m 48s] is compatible and if it's not we mark
[19m 51s] the testers skipped
[19m 52s] now since last the last stream as well
[19m 58s] pest has been updated to again
[20m 2s] receive a callable to skip and inside
[20m 5s] this callable you have access to the
[20m 7s] test case again
[20m 9s] so you can actually
[20m 14s] if i go get the logic here right
[20m 19s] you can actually bring this in and
[20m 22s] and you can just drop that in there
[20m 26s] and that will skip if the database
[20m 29s] driver isn't postgres sql so you can get
[20m 31s] rid of this line
[20m 33s] but the only thing i don't like about
[20m 35s] this at this point
[20m 36s] is a we have to copy this
[20m 39s] into every test where we want to skip
[20m 43s] but also um you can't set the message
[20m 47s] here right
[20m 47s] or you can set the message but you'd
[20m 49s] also have to set the message for every
[20m 50s] text
[20m 52s] yeah copy it every single single time
[20m 55s] exactly
[20m 56s] which i'm not i don't want to do that
[20m 57s] either um but
[20m 59s] we can add something to the pest um
[21m 2s] the pest file to basically
[21m 6s] handle it for us in a really declarative
[21m 8s] call so i'm going to copy
[21m 10s] this here and just remove it i'll
[21m 14s] probably regret that later
[21m 17s] but i'll come down to functions here and
[21m 19s] just paste it in
[21m 21s] and we'll grab the db and obviously it's
[21m 25s] no longer in a class so we don't want
[21m 27s] a function and i'm not going to return
[21m 30s] void i'm going to return test
[21m 34s] and if the db driver is not equal to
[21m 37s] postgres sql
[21m 39s] rather than saying this i can just say
[21m 41s] test with
[21m 42s] no arguments passed so you'll you'll
[21m 46s] you'll be familiar with test from this
[21m 48s] right test it works
[21m 50s] or yeah if you don't pass out just the
[21m 52s] same
[21m 53s] like it exactly yeah but if you don't
[21m 56s] pass
[21m 57s] uh you're muted i i i hear oh
[22m 1s] now i'll say it's music
[22m 4s] there is music oh it's the music on
[22m 7s] on my headphones i
[22m 10s] uh i'm hoping there's not me i can't
[22m 12s] hear any music so
[22m 14s] if there is there's something very
[22m 15s] strange going on
[22m 19s] yeah i think my girlfriend connected my
[22m 21s] headphones and turned on the music
[22m 24s] wait
[22m 29s] uh let me just check catch up with the
[22m 31s] uh with the chat
[22m 32s] could you set this as a global
[22m 34s] functioning pest that returns a boolean
[22m 36s] couldn't use a data set now uh which
[22m 39s] test are we
[22m 40s] talking about here steve just so have we
[22m 42s] already passed that
[22m 43s] but i can always come back and have
[22m 45s] another look at it if
[22m 47s] if um you let me know which test
[22m 52s] make sure you get 32 gigs of ram
[22m 55s] yeah yeah i shall i just don't know
[22m 58s] whether to
[22m 59s] to try and wait it out until the the
[23m 1s] next generation
[23m 3s] hopefully just around the corner but
[23m 4s] yeah basically uh we can add this
[23m 6s] function to the pest file right
[23m 8s] requires postgresql it's the exact same
[23m 11s] logic but instead of accessing via this
[23m 13s] we access via test
[23m 14s] as a function with no arguments and that
[23m 17s] will always return the test case
[23m 19s] and you'll notice too that i return the
[23m 21s] test at the bottom which will allow me
[23m 22s] to chain
[23m 23s] in higher order expectations or higher
[23m 25s] order tests
[23m 26s] so now if we come back to our test case
[23m 30s] or rather to our user test instead of
[23m 33s] having to
[23m 34s] do the skipping here we can change it to
[23m 36s] the bottom so first thing is i'm just
[23m 37s] going to go and update the
[23m 40s] um i'm just going to go and update the
[23m 42s] test name so it
[23m 44s] finds a user
[23m 47s] by the github email
[23m 50s] hey tom and
[23m 54s] there we have our function call and
[23m 56s] because we've now got that global
[23m 57s] function
[23m 58s] we don't have to put it there anymore we
[24m] can remove that
[24m 2s] and chain it onto the end
[24m 7s] so now when pest kind of comes to this
[24m 11s] functionality we've not got it in the
[24m 13s] feature test anymore in the base test
[24m 15s] case so it won't look there
[24m 16s] it will look in the past.php file it
[24m 19s] will find this
[24m 21s] and it's going to still be able to
[24m 23s] access dbget driver name because by the
[24m 25s] time it runs we've already set up the
[24m 28s] test case
[24m 29s] and here we can mark the testers skipped
[24m 31s] and provide our own message
[24m 33s] which means that any time we want to
[24m 35s] skip a test now
[24m 37s] if i can choose the correct test case
[24m 39s] anytime we want to skip a test for the
[24m 41s] reason that it requires postgres
[24m 43s] we literally just chain that method onto
[24m 45s] the end of our test and we get
[24m 46s] the full description output everything
[24m 49s] you'd expect
[24m 50s] i just i think that's real clean um
[24m 54s] i should probably actually run the test
[24m 55s] to prove that rather than just
[24m 57s] saying go and do it and then not working
[25m 2s] yeah here we go so it skipped that test
[25m 5s] um and if we were to run that in the
[25m 7s] actual
[25m 8s] um runner you'd see the message as well
[25m 12s] so i think that kind of because that was
[25m 14s] one of the things you you mentioned
[25m 16s] last week i think tom wasn't it like how
[25m 18s] do we actually do that in a nice way
[25m 23s] and i after playing around a little bit
[25m 25s] i think that's the nicest way
[25m 30s] so only thing i don't like about this
[25m 34s] pest file is that you
[25m 36s] said you put everything in the global
[25m 39s] namespace
[25m 41s] so you really have to be careful with
[25m 44s] naming of these methods
[25m 46s] yes personally i think i would prefer
[25m 49s] something like
[25m 51s] this test helper function with a macro
[25m 54s] for example method
[25m 56s] so i could be sure that this method is
[25m 59s] available on the test but i don't
[26m 1s] i don't want it to be available
[26m 3s] everywhere so
[26m 4s] it's not a function i will call
[26m 8s] out of nowhere it's a function i would
[26m 10s] call on
[26m 11s] a test case yeah yeah i i i
[26m 14s] agree with you um for sure um
[26m 18s] is it there vs code this is phpstorm
[26m 22s] i just strip it right back so i feel
[26m 24s] quite honored that you think this is vs
[26m 26s] code
[26m 26s] it's like vs code but it takes
[26m 30s] 10 times more ram so you know
[26m 33s] [Music]
[26m 37s] i'd like to test immediately know that
[26m 39s] it's skipped yeah i get that i get that
[26m 41s] i mean there's nothing to stop you using
[26m 42s] the standard uh
[26m 43s] this mark as skip functionality but i
[26m 46s] tend to like to put
[26m 47s] everything um that's to do with
[26m 51s] this not to do with the test itself
[26m 54s] after the test um just
[26m 57s] that's how i like to do it i mean of
[26m 59s] course if we if we move this into a
[27m] higher order test which we'll try and do
[27m 2s] in a moment
[27m 4s] then you could obviously chain and put
[27m 7s] requires postgres sql as the first
[27m 10s] line in the higher order test so then
[27m 13s] you'd still get that functionality
[27m 15s] but to to to touch on what you were just
[27m 17s] saying as well
[27m 18s] um tom because i definitely agree that
[27m 21s] you can allow this file to fill up
[27m 24s] very quickly and it can be messy
[27m 27s] um i would say that
[27m 30s] most of the time you want to be trying
[27m 32s] to put these in your in custom test
[27m 34s] cases
[27m 34s] rather than putting them in global
[27m 36s] functions
[27m 38s] but in those cases where you want to put
[27m 40s] them in here and you need access to the
[27m 42s] test case
[27m 43s] it's the test helper that will get you
[27m 46s] there
[27m 48s] if i would put it in the abstract test
[27m 51s] case
[27m 51s] we already have and in that case i could
[27m 55s] use
[27m 56s] cis and also returns this or probably
[27m 59s] yeah it
[28m] behaves the same let's try it uh yeah
[28m 4s] i think i just wanted to kind of
[28m 9s] so one question by steve i think i can
[28m 11s] answer so
[28m 12s] he asked um why it needs post address
[28m 17s] at all um there's some special query
[28m 22s] um in the email scope of the user
[28m 27s] because there's an emails plural
[28m 31s] column on the user model and
[28m 34s] this operator i'm using there is a
[28m 37s] postgres
[28m 38s] specific so it doesn't work on escolite
[28m 40s] or i'm not sure if it
[28m 42s] works on mysql but definitely not on
[28m 44s] sqlite
[28m 45s] and that's why i only run cs if there is
[28m 49s] a postgres database
[28m 51s] to run against and why i've decided for
[28m 53s] postgres
[28m 54s] instead of i think mysql
[28m 57s] is because postgres has some pretty cool
[29m 1s] features
[29m 1s] i miss in mysql like
[29m 5s] space level indexes for example and
[29m 9s] it also has better support for json
[29m 10s] columns in my opinion so
[29m 13s] um top buyers
[29m 16s] patri and i um
[29m 21s] we haven't discussed it each other but
[29m 23s] we had the discussion for example with
[29m 25s] uh
[29m 26s] busan and some others on twitter
[29m 29s] why we prefer prosperous and
[29m 33s] in favor of mysql so it's definitely
[29m 37s] some
[29m 38s] a bit of personal opinion but for me
[29m 40s] there's also a real technical reason to
[29m 43s] do so
[29m 44s] um yeah
[29m 50s] and just so steve just asked israel to
[29m 53s] refresh the data part way through the
[29m 54s] higher order test
[29m 56s] so using the new data set pr could you
[29m 58s] switch data set part way through
[30m 1s] um not sure i follow steve i
[30m 4s] i'm going to say no based on what i'm
[30m 8s] trying to to discern from
[30m 11s] what you said there um
[30m 14s] but if i've got you wrong then then hit
[30m 17s] me
[30m 17s] from what i'm reading you're asking can
[30m 20s] you have a data set so can you use
[30m 22s] with and pass in a bunch of data into
[30m 24s] with right
[30m 26s] um and can you switch this out
[30m 30s] part way through and the answer is no it
[30m 33s] will always sequentially run through the
[30m 35s] items
[30m 36s] um if that's not what you're asking
[30m 40s] then then correct me please and so i was
[30m 43s] looking at this test here right
[30m 44s] and i realized the only reason we ever
[30m 48s] knew up a user to is to check that
[30m 51s] they're
[30m 51s] the same model yeah
[30m 55s] so we could actually refactor that a
[30m 57s] little bit
[30m 58s] by just working on a single
[31m 3s] user and then checking
[31m 6s] that that user but if you need callbacks
[31m 10s] it has to be closure you definitely do
[31m 13s] so
[31m 13s] so that would fail without this one so
[31m 16s] this is another cool thing
[31m 17s] right this is a custom method
[31m 20s] and doing this won't do anything because
[31m 22s] in our custom method in the test case
[31m 25s] we accept a user or nothing else so
[31m 29s] the question is how can we make it work
[31m 31s] so that we can also accept a callable
[31m 32s] and that callable has access to the test
[31m 34s] case
[31m 35s] so let's just go and do that now i'm
[31m 38s] just going to finish this
[31m 39s] off before we do that
[31m 42s] so to drop those in there which allows
[31m 46s] us to remove all of that
[31m 49s] and then we'll go into that base test
[31m 51s] case right
[31m 52s] uh or actually is it the test case or
[31m 54s] did we put that in the pest.php file
[31m 58s] i'm not sure i i think it's there yeah
[32m 3s] to be used to be model so
[32m 6s] no so it's a lower one to be model yes
[32m 9s] so here's the comparison right
[32m 10s] and at the moment yeah we're forcing
[32m 12s] that comparison to
[32m 14s] be basically a an identical model
[32m 19s] if it wants to pass but
[32m 22s] we could actually say well is the
[32m 24s] comparison an
[32m 25s] instance of a closure
[32m 30s] and if the comparison is an instance of
[32m 33s] a closure
[32m 33s] then we can reassign the comparison
[32m 37s] to closure
[32m 40s] wrong one close your bind
[32m 44s] and we pass the comparison in because
[32m 46s] it's obviously a closure
[32m 47s] and then as the second thing we can pass
[32m 49s] uh test
[32m 51s] again with no arguments and then
[32m 54s] you just have to invoke that right so
[32m 57s] this will call the closure that we pass
[33m 1s] in
[33m 2s] but it will provide the closure access
[33m 4s] to the test case
[33m 6s] but i think we could also use uh dollar
[33m 8s] sis instead of test function
[33m 11s] like we do in the other parts of some
[33m 13s] method so we return this we
[33m 16s] as a same this value
[33m 19s] ah yes okay obviously
[33m 23s] we are into expectations without this
[33m 25s] expectation here
[33m 28s] not to test good catch yeah so we
[33m 30s] definitely want to bind to the test
[33m 32s] there is there is actually a class
[33m 34s] called reflection
[33m 36s] um which is in pest support
[33m 40s] and that has bind callable which will
[33m 42s] bind straight to the test case always
[33m 44s] but reflection is internal so don't use
[33m 47s] that
[33m 51s] but yeah this is basically the same
[33m 54s] functionality as that but
[33m 55s] with a custom integration yeah
[33m 59s] so if we come back now to our user test
[34m 3s] let's see if we can run this
[34m 6s] hey now i'm always so happy when test
[34m 9s] pass on stream that's like
[34m 11s] that's the best thing
[34m 14s] okay so to be honest luke isn't running
[34m 18s] the test he's editing he has some
[34m 20s] background tests that
[34m 21s] just do as a true true yeah
[34m 25s] it's just the stone fire this is a
[34m 27s] pre-recorded video
[34m 32s] um yeah so that
[34m 35s] that's really cool because any function
[34m 38s] that you define
[34m 40s] can basically work in the same way as
[34m 42s] these higher order
[34m 43s] callables like expect and tap which
[34m 45s] means
[34m 46s] basically anything you can do in the
[34m 50s] um in standard pest tests you can also
[34m 52s] do in higher order tests with a little
[34m 54s] bit of
[34m 55s] work um and i think it's that's just so
[34m 59s] neat i mean
[35m 1s] that is so nice to actually read and
[35m 2s] look at
[35m 5s] okay so this one is
[35m 8s] a little more interesting isn't it again
[35m 11s] we're going to block them
[35m 13s] and then it looks like we're going to
[35m 15s] grab both of them
[35m 17s] and on each instance of that user
[35m 21s] we do those same checks again
[35m 25s] so again i was looking at this a little
[35m 28s] bit earlier
[35m 30s] and there are two two things you could
[35m 32s] think here first of all we could
[35m 33s] attempt to put all of this into a single
[35m 35s] higher order test
[35m 37s] which i imagine would have a lot of taps
[35m 41s] but let's take a test two tips
[35m 45s] well yeah because well we've got to have
[35m 46s] the safe let's well let's just try it
[35m 50s] so we've got to do this we need to tap
[35m 52s] to go and grab a
[35m 53s] a new user right yeah so it will be
[35m 56s] this prepare tab
[36m 3s] so this user equals this user
[36m 8s] and then you're gonna update the user
[36m 11s] which at that point has to be another
[36m 13s] tap right
[36m 22s] uh already got that there we go
[36m 25s] so now we can remove that you need
[36m 28s] another bracket uh yep
[36m 32s] i do expect is going to become a chained
[36m 38s] method
[36m 43s] and this is going to become a function
[36m 47s] as we're interested in using this again
[36m 50s] where we access this user and then we
[36m 52s] get a new instance of user which allows
[36m 55s] us then to check on each
[36m 56s] instance um and we could actually i
[37m] guess go one step further by
[37m 2s] always asserting that these are a github
[37m 4s] user
[37m 5s] or is it just is it just that it returns
[37m 6s] a user it's just user
[37m 8s] i've omitted the github prefixes because
[37m 12s] there is only a user from github there
[37m 15s] isn't any other user
[37m 20s] right
[37m 22s] there you go uh there you go is a bear
[37m 26s] the uh the tests are failing
[37m 29s] proof that this is real uh argument one
[37m 32s] must be type user pesticide
[37m 36s] um so if i remove that
[37m 41s] because obviously by the time the user
[37m 43s] gets passed in
[37m 48s] pesticide
[37m 54s] i can just imagine steve crying now
[37m 56s] because there are no named arguments
[37m 59s] sorry steve i'm pretty happy with it
[38m 4s] i'm the total opposite of steve
[38m 7s] i use them when saying i need it and not
[38m 10s] only because i have some free time
[38m 14s] you you guys should should get an
[38m 15s] agreement and battle it out
[38m 18s] oh no i i'm fine with c uh doing
[38m 22s] named arguments everywhere and me doing
[38m 24s] sir
[38m 25s] nowhere totally okay
[38m 28s] two sides of the coin um i mean
[38m 31s] right okay so again this might be
[38m 34s] pushing it too far it probably is and
[38m 35s] i'm probably gonna undo this
[38m 38s] but because we're in laravel we can make
[38m 41s] use of laravel's helpers right
[38m 43s] so technically here you could actually
[38m 45s] tap this using laravel's tap
[38m 48s] where you create a user called gummy
[38m 52s] bear
[38m 53s] stop it you can stop it
[38m 56s] and then we retrieve a user which we
[38m 59s] then update
[39m] right and then you only got to use one
[39m 2s] tap
[39m 4s] yeah only one tab there's another tab in
[39m 6s] it
[39m 10s] i mean this is quite clean in all
[39m 12s] honesty and feel free to disagree with
[39m 13s] me
[39m 14s] i actually think this is testing two
[39m 16s] things
[39m 17s] um it it tests that the user can be
[39m 20s] updated
[39m 22s] with a block reason which is kind of a
[39m 25s] mixture of this here
[39m 26s] and this this here but then it's also
[39m 29s] testing that if we
[39m 30s] pull that user again using the um
[39m 33s] github user fetch details
[39m 36s] methodology that it's it's going to be
[39m 39s] able to retrieve it even though we've
[39m 40s] gone
[39m 41s] blocked in is that would you agree with
[39m 42s] that
[39m 46s] i agree partially but if i would test
[39m 50s] that i can update fields for every
[39m 51s] single attribute i have
[39m 53s] we weren't gonna finish the live stream
[39m 56s] so
[39m 58s] oh yeah for me updating an attribute
[40m 1s] isn't a test so
[40m 2s] uh if there is an attribute that isn't
[40m 5s] fillable
[40m 6s] or is the column is deleted from the
[40m 9s] database there
[40m 10s] will be definitely any test exploding
[40m 14s] yeah and it i don't need
[40m 17s] one specific test telling me hey you
[40m 19s] deleted a columns that is still in use
[40m 23s] um i'm fine with just any randomly test
[40m 26s] exploding and telling me
[40m 28s] you dump don't change the
[40m 30s] database structure without chatting
[40m 32s] uh checking the columns are still needed
[40m 34s] so yeah
[40m 36s] so i guess then that being the case and
[40m 39s] again this is this might be much of a
[40m 40s] muchness
[40m 42s] um but technically when we're doing
[40m 45s] these tests on the user here
[40m 46s] they're a little bit redundant because
[40m 48s] that's going to be the case anyway
[40m 50s] because we trust this code up here so we
[40m 53s] could actually simplify this
[40m 54s] slightly by instead of just expecting on
[40m 57s] a new user
[40m 58s] and not requiring the each statement
[41m 4s] because the functionality is assumed
[41m 7s] that when you create a user and update
[41m 10s] it
[41m 11s] those columns are going to be equal to
[41m 13s] that on that particular object
[41m 15s] yeah good point so i mean i
[41m 21s] i am all for the each statement of
[41m 22s] course
[41m 24s] um you will not stop me writing in each
[41m 26s] statement but
[41m 27s] sometimes it just reads a little bit
[41m 29s] nicer to have
[41m 30s] like a single a single level of flow
[41m 33s] going on
[41m 34s] um all right we actually need to
[41m 38s] switch over to to actually write and
[41m 39s] change these tests here we need postgres
[41m 42s] actually i think one one thing we could
[41m 44s] do
[41m 45s] is um add one
[41m 49s] method val return value high order test
[41m 53s] because
[41m 54s] um on the user models there is a trade
[41m 58s] called blockable and this should define
[42m 1s] a method
[42m 4s] is blocked
[42m 7s] so
[42m 11s] yeah it's blocked exactly oh
[42m 18s] i'm proving that it's actually my
[42m 19s] computer and it wasn't code with me
[42m 21s] yesterday steve
[42m 26s] all right so we could actually say is
[42m 29s] blocked
[42m 31s] to be true right yeah
[42m 36s] so now i definitely agree that this test
[42m 38s] is testing multiple things
[42m 40s] but for me this is definitely fine
[42m 42s] instead of doing the same preparation
[42m 45s] multiple times only to have one test
[42m 48s] testing that the block reason
[42m 50s] is the same i filled in and one test
[42m 52s] checking
[42m 53s] unit test likes that this method returns
[42m 55s] true if the block reason is fit in
[42m 58s] yeah so and and also here
[43m 1s] like we've got this block column and the
[43m 3s] is block method is now obviously
[43m 5s] checking that that column is correct
[43m 7s] right
[43m 7s] so yeah the test is is more complete
[43m 11s] absolutely yeah awesome let's move on to
[43m 15s] this one then
[43m 15s] um so i have already gone ahead
[43m 20s] and um where's my xml gun
[43m 24s] there we go and prepared this
[43m 28s] earlier i can switch these two things
[43m 31s] over
[43m 33s] without disrupting the flow of the
[43m 35s] stream too much
[43m 37s] so if we run this test now it shouldn't
[43m 38s] skip and it should indeed
[43m 40s] pass
[43m 45s] and it does all right which means we can
[43m 47s] start
[43m 48s] hacking it to pieces yeah
[43m 51s] first things first we already have
[43m 53s] refactored this so let's go grab that
[43m 55s] from up here
[43m 59s] and we're just calling update on that
[44m 1s] from the looks of things
[44m 4s] so we'll call update on that
[44m 9s] and get rid of that and
[44m 12s] we go grab another user by email
[44m 17s] we okay this is where we might need to
[44m 20s] go and add a new expectation just to
[44m 22s] wrap this and no
[44m 24s] it's the same like we had before to be
[44m 26s] used yeah
[44m 27s] so we can say expect a url terminator
[44m 34s] again
[44m 40s] even if it sounded more like a like a
[44m 42s] tattoo machine
[44m 44s] this is fast spinning needle
[44m 54s] we lasted so long there i thought i
[44m 56s] thought that it had given up i thought
[44m 58s] that the
[44m 59s] the ghost in the machine had gone away
[45m 1s] apparently not
[45m 3s] uh awesome yeah so we have that and then
[45m 6s] we can higher order that right let's
[45m 8s] move this into here
[45m 14s] we expect that it's obviously a user and
[45m 16s] then we expect that the
[45m 18s] name is
[45m 21s] tell me there yeah
[45m 24s] do you know out of your head if to be
[45m 27s] useless as an equals or as the same
[45m 30s] yeah so to be is this there's two
[45m 32s] there's two equal which is
[45m 34s] which is type um loose
[45m 38s] and then there's to be which is type
[45m 40s] script
[45m 41s] okay so yeah if you if you want it to
[45m 45s] check that it's the subject it's to be
[45m 48s] basically um
[45m 52s] awesome and obviously for strings
[45m 53s] because they're um
[45m 58s] a word that isn't coming to my head now
[46m] then
[46m 1s] they will patch to be as well as long as
[46m 3s] they're the exact same content
[46m 5s] yeah uh that's cool yeah i think we can
[46m 8s] inline this now and this will make
[46m 9s] the bear happy i think because we should
[46m 11s] be able to put this at the top of the
[46m 13s] test
[46m 15s] and um we should be able to now say tap
[46m 24s] do this little piece of logic here
[46m 31s] and then exit
[46m 36s] again i think i would at least line
[46m 39s] breaks this
[46m 40s] update array so having email and email
[46m 43s] verified ad on
[46m 44s] on separate lines
[46m 59s] and prefect that's the next one let's
[47m 2s] just see
[47m 8s] that
[47m 10s] oh there we go all right
[47m 13s] all right so that was a little bit of
[47m 16s] attached to me as well actually i didn't
[47m 17s] know whether that was gonna work at the
[47m 18s] top it does work at the top
[47m 20s] we're happy all right moving on to the
[47m 23s] next one uh it finds a
[47m 24s] user by the github email very similar
[47m 27s] thing here right we can
[47m 28s] first of all
[47m 32s] get rid of that we can move this call
[47m 34s] here to the top
[47m 36s] it's the exact same test but just with
[47m 39s] another email value so
[47m 41s] possibly let's let's let's really
[47m 44s] compare them it could be that we could
[47m 46s] use a data set here
[47m 48s] yeah that's a good idea so the
[47m 51s] difference update is the same
[47m 52s] from github is the same as a user
[47m 55s] and we assets the same name so yeah
[47m 59s] it's literally just that in which case
[48m 2s] i think we can can't we we can say
[48m 8s] with open up a data set
[48m 12s] and in each of those we need to pass in
[48m 15s] an array of values and i will put these
[48m 18s] on separate lines
[48m 20s] to appease everybody
[48m 24s] don't need to shoot me yet
[48m 29s] and we grab that put that in there
[48m 32s] and now in the expect from today
[48m 36s] um so hot off to press this we can pass
[48m 40s] in
[48m 40s] the email which is being passed from the
[48m 43s] data set right
[48m 45s] so we pass the email in there
[48m 49s] which lets us remove these tests
[48m 55s] and moment of truth i guess right
[49m 1s] two tests that's a positive start
[49m 4s] these tests are obvious to take a little
[49m 6s] longer because at this point we're
[49m 7s] booting up
[49m 9s] a a postgres database both passed
[49m 12s] how happy so yeah that's the that's the
[49m 15s] new functionality
[49m 17s] could you check how the name of the test
[49m 20s] was we have deleted
[49m 22s] uh yes i will bring so on onto left
[49m 26s] yeah so what always the name
[49m 30s] [Music]
[49m 33s] no that is by their github email okay
[49m 38s] so could we use name data set
[49m 41s] no no not a name data set but an
[49m 44s] associative area in the data set
[49m 48s] so yeah so so you're saying that we'd
[49m 50s] say this is the github
[49m 53s] exactly and this is normal yep
[49m 56s] yeah i'm pretty sure that works um it
[49m 59s] does
[50m 1s] so this is something i've already used
[50m 3s] and knows that
[50m 9s] [Music]
[50m 10s] oh then is this something else
[50m 14s] i think it's uh if you delete it so test
[50m 17s] again
[50m 19s] i that's
[50m 24s] fine using it that's a different test
[50m 25s] isn't it oh then
[50m 27s] one way to find out let me just remove
[50m 28s] these
[50m 37s] it could be set to terminators back
[50m 45s] yeah it was you
[50m 58s] hello i think hey i think i'm gonna have
[51m 1s] to
[51m 1s] maybe try um streamyard next week or
[51m 4s] something
[51m 5s] yeah see if it it it's got to be
[51m 7s] something like that
[51m 9s] um anyway so i
[51m 12s] i know that pest will give us more
[51m 14s] descriptive outputs by default anyway
[51m 16s] like these will be included in the
[51m 17s] description
[51m 19s] so it will say like test testing find
[51m 22s] users by email
[51m 23s] with and then it'll it'll show this
[51m 25s] email in the in the test description
[51m 28s] uh in fact you might be able to see it
[51m 29s] here look yeah here we go
[51m 32s] so you'll see that which is more than
[51m 33s] php unit but for some reason passing it
[51m 35s] as an associative array doesn't work
[51m 37s] perhaps one to look at
[51m 38s] so so it's the same output like php unit
[51m 44s] php unit also displaces
[51m 48s] yeah but the real problem here is
[51m 51s] for this normal email it's totally fine
[51m 53s] but like you see
[51m 54s] versus a github email for example or if
[51m 56s] you have
[51m 57s] multiple arguments so you could have a
[51m 59s] data set with like
[52m 1s] two or three arguments you want to pass
[52m 3s] in as this
[52m 5s] format is really horrible so
[52m 8s] yeah yeah that's actually skipped it
[52m 11s] hasn't as well
[52m 12s] it's it's skipped over this one
[52m 16s] that's interesting that's interesting
[52m 17s] that's uh that's one to file
[52m 21s] next time
[52m 25s] back report yep pr
[52m 28s] going up i was going to share the uh
[52m 32s] i was going to do the pr for the data
[52m 34s] provider values
[52m 36s] on stream today and then by the time i
[52m 39s] i'd actually finished thinking about
[52m 40s] that i'd already got the p already and i
[52m 42s] was like oh no
[52m 43s] i'm just going to leave it maybe
[52m 47s] we'll have to look at that one it might
[52m 48s] be one that we could do on stream um
[52m 51s] so it finds user by
[52m 54s] name only github email yeah
[52m 59s] okay so what's going on here you can
[53m 2s] again scroll up and duplicate the
[53m 6s] the first data set entry
[53m 11s] no no only uh first six one
[53m 15s] eight seven exactly
[53m 18s] you just want that yeah no no only
[53m 21s] just adding a new entry to city center
[53m 24s] oh
[53m 24s] this one yeah
[53m 29s] i wanted to say it duplicate it and
[53m 32s] remove the id in front but yeah
[53m 36s] and could we move the it's only a small
[53m 38s] thing but could we move this dev at
[53m 40s] coming beer dot the e to the top
[53m 43s] so it checks the normal emu at first
[53m 45s] yeah that's a good plan yeah
[53m 47s] and this email this test is doing
[53m 48s] nothing other than that yeah
[53m 51s] i think again absolutely it's the same
[53m 53s] awesome
[53m 55s] and then this is the same as well right
[53m 58s] yep but only
[53m 58s] id okay let's get rid of that
[54m 5s] so for all of you interested in these
[54m 8s] github emails
[54m 8s] um the first one is the officially
[54m 11s] supported one
[54m 12s] so this is id plus name the last one is
[54m 16s] also officially supported because this
[54m 18s] plus name is
[54m 19s] like on gmail you can use these email
[54m 22s] suffixes that aren't
[54m 24s] handled by the real mail inbox
[54m 28s] delivery whatever
[54m 31s] dispatch service but it's only for you
[54m 34s] so it's that you can assign labels for
[54m 36s] example
[54m 36s] but this mid one is the name only this
[54m 39s] is not officially supported but
[54m 41s] i think it was some time back
[54m 44s] and this is why some users still use
[54m 47s] this format
[54m 48s] and as long as i'm able to find
[54m 52s] someone i thought that it's okay to
[54m 54s] support it as well
[54m 57s] that's very inclusive of you
[55m 2s] i try my best okay this is actually the
[55m 5s] last
[55m 6s] this is a beast right but it's the last
[55m 7s] test case
[55m 9s] of the of the of the class
[55m 13s] this is one uh susanna and i created
[55m 16s] because uh she found a bug let's name it
[55m 20s] and we solved it not on stream but in a
[55m 24s] in a mentoring live coding session and
[55m 27s] she said
[55m 28s] yeah let's add a test for it to prevent
[55m 30s] the same error
[55m 32s] in the future and so this price is test
[55m 35s] cases set
[55m 37s] okay yeah so what's the actual
[55m 42s] error like what was actually happening
[55m 44s] what's the test testing for
[55m 47s] problem was uh github usernames can be
[55m 50s] upper and lowercase and github keeps it
[55m 52s] and like i think nearly all of you know
[55m 56s] by default sorting in php and i think
[55m 59s] all other languages
[56m] is a case sensitive so if you have lexus
[56m 3s] algolia and that's atomic both starting
[56m 5s] with an a
[56m 6s] but one with a lower case the next one
[56m 8s] was an upper case so they have
[56m 10s] different let's say sword priorities
[56m 14s] and this was a problem because um
[56m 18s] in this test case we have this algolia
[56m 20s] with a lower a barry
[56m 22s] thunderhead with a lower b and
[56m 24s] astrotomic with an upper a
[56m 26s] and if you sort alphabetically
[56m 30s] the normal user would expect okay a
[56m 32s] goliath first astronomic second
[56m 33s] and very third but what was happening
[56m 36s] was
[56m 36s] algolia first barry second at s atomic
[56m 39s] cert
[56m 40s] and this is what we have fixed and this
[56m 42s] is also what this test
[56m 44s] tests said this order is always
[56m 48s] uh to the lowercase enforced order and
[56m 50s] what the user would expect to see
[56m 53s] yeah that makes sense that makes
[56m 55s] complete sense
[56m 57s] so i think first of all let's go about
[57m] um like 10 tip methods
[57m 4s] well i think i think at some point
[57m 7s] you've got to give up on
[57m 8s] on higher order but let's at least first
[57m 10s] of all move over to expectations
[57m 12s] reflect the way we can and then we'll
[57m 14s] see we'll see if it makes sense for a
[57m 17s] higher order or whether it's more
[57m 18s] sensible to leave it inside the function
[57m 21s] um so it returns
[57m 25s] vendors alphabetically
[57m 29s] yeah i would add something like sorted
[57m 32s] or ordered
[57m 34s] uh it returns ordered vendors
[57m 37s] alphabetically regardless of case
[57m 40s] yeah so we've already got this
[57m 43s] it's just going to be this user
[57m 48s] um and we already have that expect
[57m 53s] user the user
[57m 58s] and we already have this which is going
[58m 1s] to be
[58m 4s] this user we don't need
[58m 7s] that user's tag
[58m 10s] so what we don't have yet is a um an
[58m 13s] organization helper
[58m 14s] which i think might might be helpful
[58m 17s] yeah definitely we already needed three
[58m 20s] times
[58m 20s] only for this test and very likely
[58m 24s] some more times in the future if we
[58m 26s] write tests for organization model
[58m 30s] so we can say organization which should
[58m 33s] have
[58m 34s] a name and i use organization with the
[58m 37s] set
[58m 38s] if i'm right yeah so not with an s but
[58m 41s] with a
[58m 42s] with a set uh my english is
[58m 45s] coming out one second
[58m 51s] i'm pretty sure it's one of these links
[58m 53s] like the uk english and
[58m 54s] u.s english or whatever
[58m 58s] i don't want to start that argument on
[59m] stream that would it'd be a bad screen
[59m 2s] wouldn't it
[59m 3s] um what's your most common one with
[59m 7s] should i use astrotomic as the default
[59m 11s] yeah i think so yeah so
[59m 14s] that will basically be doing that by
[59m 16s] default and
[59m 18s] we will go
[59m 24s] return organization
[59m 27s] this will return an organization
[59m 31s] and here we can just pass the name in
[59m 38s] i missed off a little hey
[59m 41s] that's nice gives it
[59m 45s] a nice helper we can reuse in many
[59m 47s] places
[59m 48s] so let's go select those
[59m 54s] organization
[59m 59s] and then i think you have a double sis
[60m 2s] on
[60m 2s] the second one
[60m 6s] organization oh yeah
[60m 11s] okay so let's jump over that for now
[60m 16s] we sync the repositories
[60m 20s] book the vendor names so assert equals
[60m 22s] can become expect
[60m 26s] render names all
[60m 30s] to equal and then we'll take that
[60m 35s] there is um a another one that you might
[60m 39s] want to use or you can use is to match
[60m 41s] array
[60m 42s] but the only thing with that is it
[60m 44s] matches array partials as well so if you
[60m 46s] want to be specific and say it has to be
[60m 49s] the whole array
[60m 50s] um you want to use two equal instead of
[60m 52s] to match array
[60m 54s] but in in certain cases like if you're
[60m 55s] just testing a part of a
[60m 57s] api response for example you can use to
[61m] match array and you'll just get the
[61m 1s] subset which is really useful when
[61m 3s] you're not interested in the id or
[61m 4s] whatever further up
[61m 6s] yeah cool um and i think we can actually
[61m 11s] do a sweet sweet sequence here right
[61m 15s] effect vendor names sequence
[61m 23s] um let's say vendor
[61m 29s] to be how golia
[61m 35s] out a few times
[61m 49s] so only thing i'm not sure if we used to
[61m 51s] be instead
[61m 52s] of to equal yeah
[61m 56s] does to be or as a same
[61m 59s] compare every positions i think so
[62m 9s] yes this will be an endless repertoire i
[62m 12s] think it's faster to
[62m 13s] just switch positions of some arguments
[62m 16s] and
[62m 17s] so it's a in the upper one because we
[62m 20s] we test on the vendor names and both
[62m 22s] times so
[62m 23s] just replace to equal with to be
[62m 28s] yeah and switch algolia and s atomic for
[62m 31s] example
[62m 32s] in this array here yeah
[62m 38s] and let's run the test and if it fails
[62m 42s] on this expectation we can
[62m 46s] we can remove the sequence because
[62m 49s] in that case we already checked
[62m 55s] yeah oh
[62m 58s] transaction aborted
[63m 2s] exception
[63m 6s] it's interesting
[63m 10s] let's have a look
[63m 14s] the transactions board command's ignored
[63m 17s] until end of transaction block
[63m 25s] ah any clue
[63m 34s] not from this point
[63m 47s] this is nice and silly but let me just
[63m 49s] come back here
[63m 53s] so only idea but i think this one also
[63m 57s] requires
[63m 58s] what does it
[64m 1s] ah well that's an interesting one isn't
[64m 3s] it it might be something to do with the
[64m 5s] refresh database trait perhaps
[64m 8s] possibly yeah maybe the interaction
[64m 11s] is uh yeah okay
[64m 15s] so let's switch them back and we can
[64m 17s] remove the sequence
[64m 24s] simplifies that nicely also means that
[64m 26s] we can
[64m 27s] inline that
[64m 33s] [Music]
[64m 36s] what's the idea with these two lines
[64m 38s] here tom
[64m 44s] i think the expectation here can be
[64m 46s] removed
[64m 51s] now only is expectation you still need
[64m 53s] to be user
[64m 55s] yeah oh yeah because we think their
[64m 57s] contributions later down the line
[65m] but we could move down
[65m 3s] the user to where the contributions are
[65m 6s] synced
[65m 7s] i like that cleaner keeps it all grouped
[65m 10s] together
[65m 11s] and then we still need the user don't we
[65m 23s] okay
[65m 26s] let's see if that works
[65m 35s] it does man awesome so
[65m 38s] by looking at this let's be quite honest
[65m 42s] i don't think
[65m 43s] it's going to get much simpler is it
[65m 47s] we have to do these steps if we
[65m 50s] move to higher order we're just going to
[65m 53s] be tapping over and over again
[65m 55s] but we don't win anything yeah it's not
[65m 58s] going to make it any cleaner
[65m 59s] so i i think in this case we we have to
[66m 3s] put the white handkerchief out and and
[66m 6s] kind of just accept that
[66m 7s] that it's not going to be higher order
[66m 9s] test but still very clean
[66m 11s] still very clean which is nice yeah
[66m 14s] um all right cool
[66m 17s] that's this test case then completely
[66m 20s] moved over to
[66m 22s] higher order tests slash slash using all
[66m 24s] the latest updates
[66m 26s] um and even the new
[66m 30s] the new data higher order data providers
[66m 31s] which is cool
[66m 34s] um awesome all right
[66m 37s] have we got any other tests we've got
[66m 39s] this example test but that's not really
[66m 41s] doing anything is it
[66m 44s] i don't it's just what it takes i don't
[66m 48s] think we've got any other tests
[66m 49s] that have been created that we haven't
[66m 51s] migrated to pest have we
[66m 54s] like said there aren't many tests
[66m 57s] so most of these user tests were added
[67m] to
[67m 1s] provide an example how i want to test
[67m 4s] some of these github related and
[67m 8s] keyword related things yeah yeah
[67m 13s] but no one in the community provided a
[67m 15s] new test
[67m 18s] hey you just have to wait you just have
[67m 19s] to imagine the right people
[67m 22s] and push this up so let's commit and
[67m 26s] yeah we could start with the first
[67m 29s] organization test for example
[67m 42s] i might have to go back and change the
[67m 45s] uh
[67m 47s] revert the phpunit.dist file
[67m 50s] well they're quite useful to have their
[67m 53s] people can just go in and quickly
[67m 55s] uncomment those lines
[67m 59s] i mean what do you usually do how do you
[68m 2s] usually handle that time like if you
[68m 4s] were testing
[68m 6s] and you wanted to test using postgres
[68m 8s] would you go into the the
[68m 10s] um php unit file and manually alter
[68m 13s] those lines or
[68m 14s] how do you usually tackle that
[68m 19s] i'm doing what php unit would expect me
[68m 22s] to do
[68m 22s] and copy the php unit xml dist to a php
[68m 26s] unix some lfi
[68m 28s] and alter this one because this file is
[68m 31s] on the git ignore
[68m 32s] so i don't have to worry about
[68m 35s] not committing the file and php unit has
[68m 38s] a priority of picking
[68m 40s] first the phpunit.xml file if it is
[68m 43s] present and if not it will pick up the
[68m 45s] php.xml.list
[68m 47s] which is part of the repository so
[68m 51s] if i want to adjust my local php unit
[68m 54s] environment without touching
[68m 56s] the repository php unit
[68m 59s] i just copy
[69m 2s] some php on the dxml dist and adjusted
[69m 7s] yeah which makes complete sense
[69m 10s] um all right
[69m 14s] so i think for appender i even have a
[69m 17s] php unit.xml
[69m 20s] dot dot postgres or something like that
[69m 24s] so i can just
[69m 27s] rename this file
[69m 31s] yours would literally be although they
[69m 33s] get ignored it's not picked up
[69m 35s] interesting
[69m 42s] below the yarn arrow yeah
[69m 46s] cool in which case i can come in keep
[69m 49s] those there
[69m 51s] head back into here remove those lines
[70m] exactly
[70m 3s] and everyone stays happy
[70m 8s] all right everyone okay everyone's still
[70m 11s] with us
[70m 13s] we're not moving too much
[70m 18s] a lot of time i'm not sure if they all
[70m 21s] fall asleep or
[70m 24s] have left us alone i mean we're not
[70m 27s] moving
[70m 28s] extreme pace but i i kind of like to uh
[70m 31s] take my time when i'm doing side
[70m 33s] projects because it's where i learn
[70m 35s] things
[70m 36s] yeah boys this deep with us that they
[70m 40s] aren't
[70m 41s] able to write comments so they could
[70m 44s] they could well be
[70m 45s] so deep in it and i think you
[70m 49s] again wrote organization with some s
[70m 51s] instead of saying
[70m 52s] um you you would point that out wouldn't
[70m 56s] you
[70m 56s] i'm just trying to slide one through
[70m 58s] okay just one
[71m 1s] [Laughter]
[71m 4s] all right um so organizations
[71m 7s] what what can you tell us about
[71m 8s] organizations one question i have first
[71m 11s] is
[71m 12s] you have this uh pest plugin installed
[71m 14s] and
[71m 16s] in the tests feature models directory
[71m 20s] these two files have this pest icon why
[71m 22s] does this
[71m 24s] update user details test doesn't have
[71m 26s] the test icon
[71m 27s] right okay the only thing i can think
[71m 30s] and i need to ask oliver about this
[71m 31s] because he's the guru
[71m 33s] is that at some point in higher order
[71m 36s] syntax
[71m 38s] whatever the plug-in is doing to detect
[71m 40s] a pest test
[71m 42s] like stops understanding what's going on
[71m 48s] so i the the only thing i can think of
[71m 51s] um is that because if i simplify this if
[71m 55s] i get rid of these
[71m 56s] here look look at that okay but when we
[71m 59s] bring in higher order tests
[72m 1s] in higher order expectations which to be
[72m 3s] fair is like
[72m 7s] quite crazy anyway but when we do that
[72m 10s] it does disappear so i think i just need
[72m 12s] to to chat with oliver and
[72m 14s] uh no doubt he can update it to also
[72m 18s] allow for properties yes i wrote the
[72m 21s] tests
[72m 22s] i think the easiest would be just check
[72m 24s] in if the file contains
[72m 26s] uh a call to the global namespace
[72m 29s] to the global function it or test yeah
[72m 33s] yeah because this is what identifies a
[72m 36s] pest test
[72m 37s] yeah yeah and absolutely not caring if
[72m 40s] it has a
[72m 41s] closure argument or some higher order
[72m 44s] chaining to
[72m 45s] the end of life i i feel like super
[72m 48s] sorry for oliver though because
[72m 50s] he does like such an awesome job
[72m 51s] maintaining that plugin and then
[72m 53s] every week one of us goes in and creates
[72m 56s] some new monstrosity that he's got
[72m 59s] an understanding poor guy you think
[73m 2s] he's also on his couch now and saying
[73m 5s] stop
[73m 6s] providing new ps i want just to
[73m 10s] to finish this project yeah i think he
[73m 13s] is
[73m 14s] i'm really sorry okay so
[73m 17s] organizations are yeah nearly the same
[73m 21s] like a user
[73m 22s] from um business logic level
[73m 25s] the only difference is that immigration
[73m 28s] said an organization has a relationship
[73m 30s] to users as well
[73m 32s] and so users can be members of
[73m 36s] organizations
[73m 37s] and organizations can't be contributors
[73m 40s] to repositories
[73m 42s] so in fact an organization has members
[73m 45s] it can own repositories and
[73m 48s] sets it okay so
[73m 52s] i mean is there any any point in kind of
[73m 55s] copying some of the testing
[73m 57s] from here into our organization test and
[74m] then seeing what correlates and what
[74m 1s] doesn't
[74m 3s] um scroll to the top and i can tell you
[74m 7s] so i think the first one can be copied
[74m 10s] and the second one also
[74m 15s] the third one as well because
[74m 17s] organizations can be blocked
[74m 20s] and the other ones
[74m 24s] are unique to users okay that's quite
[74m 26s] good there that only puts us with
[74m 28s] two tests that can't be copied
[74m 32s] so let's go grab those
[74m 36s] drop this in here and
[74m 39s] we haven't created a 2b organization
[74m 41s] expectation yet have we
[74m 44s] we don't have so you can go do that
[74m 56s] perfect i had to
[74m 59s] i had like a brain freeze there just
[75m 1s] trying to work out what i had to do
[75m 6s] and yeah doesn't exist and would also
[75m 11s] have to be written with a set okay okay
[75m 16s] i'm sorry no worries what is a cert user
[75m 19s] actually doing
[75m 22s] a lot all right
[75m 25s] this is fine though i mean that's the
[75m 26s] the most basic is this right
[75m 28s] and then we want to let the id is an
[75m 31s] integer
[75m 32s] and i think you changed this only to
[75m 36s] showcase
[75m 36s] that we can use expectations insist
[75m 40s] and you wanted to proceed but you've
[75m 42s] also seen that
[75m 43s] as a lot going on and stopped it yeah
[75m 49s] you know me too well i don't like this
[75m 52s] so personally
[75m 52s] i would also be fine if we refactors
[75m 55s] expect actually
[75m 56s] to back to the php unit
[75m 59s] uh like
[76m 2s] someone's no not this but like this
[76m 5s] asset is inter
[76m 6s] isn't yeah that's a instance of
[76m 18s] all right so
[76m 21s] we sort of want to take this user
[76m 24s] assertion
[76m 25s] almost duplicate it out but then change
[76m 29s] quite a few things
[76m 31s] a lot of things
[76m 37s] organization assertions yes
[76m 42s] organization
[76m 45s] actual and
[76m 48s] yeah this is where we want to also say
[76m 50s] and perhaps we want to go back and
[76m 51s] update the other one as well but that
[76m 53s] you could pass in a
[76m 54s] closure right no you can't pass this to
[76m 58s] to spawn you can pass it to
[77m 2s] to the pest expect extension
[77m 6s] yeah but you can change the param app
[77m 9s] models
[77m 10s] organization nice nice absolutely
[77m 13s] correct
[77m 13s] you will also have to change the trade
[77m 15s] name
[77m 18s] you don't have to no just
[77m 24s] and yeah
[77m 27s] you could have also just press enter
[77m 33s] and so id has to be an integer
[77m 37s] name is a string you can remove the
[77m 40s] email
[77m 40s] as certain organizations don't have one
[77m 43s] github email will be the same avatar url
[77m 47s] is twitter is emails can be removed
[77m 52s] github access token can be removed
[77m 58s] description keeps in location should
[78m 1s] we keep in and i'm thinking
[78m 4s] said full name is display name on
[78m 9s] its organization is the have we got
[78m 12s] every
[78m 12s] organization's table can we check that
[78m 16s] you can also go to the organization
[78m 18s] model because i have all attributes on
[78m 21s] in the doc block are you an organized
[78m 23s] person
[78m 24s] yeah i'm using the idea but so full name
[78m 27s] oh so as a full name okay
[78m 33s] oh yeah so display name was what i'm
[78m 36s] using
[78m 36s] if i'm not sure what i need because this
[78m 39s] planning falls back to the normal name
[78m 41s] if there isn't a full name defined as
[78m 44s] you can see
[78m 45s] the full name can be nullable as well
[78m 49s] right okay yeah and can we so are we
[78m 52s] saying that we can
[78m 54s] we could also do the same with display
[78m 55s] name
[78m 58s] it's display no display name is not an
[78m 59s] audible right so it has to have a
[79m] display name
[79m 2s] yeah so we could almost uh does it have
[79m 5s] a name name
[79m 7s] lots of names so this display name
[79m 11s] is the actual oh no we do have a name
[79m 13s] yeah
[79m 16s] display name is an overloaded attribute
[79m 21s] so it's not in the database it's just in
[79m 23s] peach people
[79m 24s] and you want to do are we wanting to
[79m 26s] assert that that is a string
[79m 32s] i haven't tested overloaded attributes
[79m 35s] on the user
[79m 36s] but we could edit in both cases sure
[79m 42s] so yeah it's got it's got to return it
[79m 45s] hasn't it
[79m 47s] so might as well give this a test uh
[79m 50s] without having to do individual tests
[79m 53s] yeah
[79m 54s] and then it is an organization can be
[79m 56s] blocked she said didn't
[79m 58s] it's also blockable yeah yeah so
[80m 1s] we don't need this email can be removed
[80m 8s] argument has to be changed to
[80m 9s] organization
[80m 14s] and everything keeps the same
[80m 19s] avatar url also changes the argument to
[80m 21s] organization
[80m 23s] but everything keeps the same so twitter
[80m 27s] one
[80m 28s] argument has to change the organization
[80m 30s] but everything
[80m 32s] also keeps the same
[80m 36s] and emails can be moved
[80m 40s] not so bad brings up
[80m 47s] awesome okay so now back into our past
[80m 51s] we can say
[80m 55s] organization assertions
[80m 59s] so organization this value
[81m 3s] yep
[81m 6s] well that was simple
[81m 11s] right organizations test so we actually
[81m 13s] want this to be
[81m 17s] organization
[81m 20s] um this is going to change to
[81m 23s] organization right
[81m 25s] and we probably want to pass the default
[81m 28s] of astrotomic
[81m 31s] yeah that's about that and so so test
[81m 33s] name is also creates organization from
[81m 35s] github
[81m 38s] thank you
[81m 43s] i guess that comes from the fixture
[81m 44s] right
[81m 47s] yeah name is right
[81m 50s] also name is correct i'm just gonna
[81m 54s] move that off to the side but keep it
[81m 55s] open in case we want to reference it
[81m 57s] and email can be removed
[82m] yeah and this register
[82m 4s] isn't doesn't exist on the organization
[82m 7s] it can be also removed is there anything
[82m 10s] organization specific at like the low
[82m 12s] level or
[82m 13s] or like like well i guess the opposite
[82m 15s] way actually like the high
[82m 16s] organization model level that we want to
[82m 18s] test oh i forgot one attribute specific
[82m 21s] to organizations we can add to the
[82m 23s] organization assertions
[82m 25s] yeah and it's a php unit asset is
[82m 30s] boolean
[82m 33s] let's put it here
[82m 39s] and is verified should it be
[82m 51s] and marion i'm using the organization
[82m 53s] with the set because
[82m 54s] it's how github writes it and because
[82m 58s] i'm using github as a base of
[82m 59s] everything i've also just copied how
[83m 3s] organization is written and i think it's
[83m 7s] like yeah like i said it's one of these
[83m 9s] words that
[83m 11s] uh both things are correct i think
[83m 14s] so with an s and this is set
[83m 17s] i think it could be something like one
[83m 19s] in the us at once the uk
[83m 22s] um but i don't have any idea so
[83m 27s] best to stick to um convention if in
[83m 29s] doubt yeah
[83m 32s] and johnny my light theme i lasted like
[83m 35s] six hours man and then
[83m 37s] i couldn't do it i tried honestly i
[83m 40s] tried hey
[83m 41s] have you tried for tone scenes uh i've
[83m 44s] downloaded it i've yet to give it a
[83m 45s] proper
[83m 46s] i gave it like 10 minutes but i need to
[83m 48s] give it a proper coding session
[83m 50s] before i can give judgment
[83m 53s] give it a try now on live stream because
[83m 55s] light seems are easier for
[83m 57s] for the viewers to to watch oh my
[84m 2s] goodness that doesn't mean that that's
[84m 3s] going to make it even harder for me to
[84m 4s] code though
[84m 7s] i'll tell you what let's let's save it
[84m 9s] for next time because it's
[84m 10s] uh we're one hour 24 in and i don't want
[84m 12s] to try and keep these to like an hour
[84m 14s] and a half
[84m 15s] yeah so let's
[84m 18s] tell you what i'm going to comment all
[84m 20s] of these out other than this top one
[84m 24s] and then we'll go in this is going to be
[84m 26s] the problem now look
[84m 28s] tell you what that's here's a cool
[84m 31s] thing if you're using pest tests you can
[84m 34s] tag only on which means that when you
[84m 36s] run the test suite the only one that's
[84m 37s] going to be run
[84m 39s] is the one with only cool okay so
[84m 42s] failed asserting that null is of type
[84m 44s] bull and this is obviously from our
[84m 45s] little organization assertions
[84m 47s] so let's go in and
[84m 52s] just look at the properties
[84m 55s] get attributes is is verified a dynamic
[84m 59s] attribute
[85m 2s] oh yeah i think yeah it should be a
[85m 5s] nullable as certain
[85m 10s] now you have to replace php unit with
[85m 12s] nullable assertions
[85m 14s] it should be auto completed because we
[85m 16s] are using yeah
[85m 18s] and asset is nullable boolean yeah
[85m 21s] i think with that i'll probably put this
[85m 23s] down here
[85m 24s] yeah exactly
[85m 29s] and you can move it to to the location
[85m 31s] and description block
[85m 33s] so no need for this empty line
[85m 39s] cool and we've got a passing test again
[85m 41s] right
[85m 43s] perfect same thing here but this is
[85m 47s] both of these are going to become
[85m 48s] organization
[85m 51s] i did that one without even thinking
[85m 52s] about it look at that
[85m 56s] and it finds an organization from github
[86m 15s] happy days
[86m 19s] and this test here
[86m 23s] start with this essentially in fact i'm
[86m 25s] trying to be clever
[86m 26s] use a user user
[86m 30s] user you should have dances for the
[86m 33s] whole file
[86m 36s] hey man we're learning as we go okay
[86m 39s] yeah
[86m 41s] this is a past stream not a php storm
[86m 43s] stream
[86m 47s] yeah but being able to do this just
[86m 49s] shows
[86m 50s] how similar user and organization model
[86m 53s] are
[86m 54s] yeah so which is nice
[86m 57s] and again let's just tap only on there
[86m 59s] while we run this test
[87m 6s] happy days so we can remove that
[87m 9s] and now we have three organization model
[87m 12s] tests right
[87m 13s] yeah that discovering the same
[87m 17s] behavior we ensure the user has
[87m 20s] is there any organization specific
[87m 22s] behavior that we should
[87m 24s] that we should think about or tackle
[87m 28s] yeah i think same like um
[87m 31s] for the user so we have it partially
[87m 35s] covered with the last test to send an
[87m 37s] eye added to the user
[87m 38s] but we aren't testing any of the
[87m 41s] relationships
[87m 42s] so we don't test that organization can
[87m 44s] be had members
[87m 46s] and we don't test such an organization
[87m 48s] can own repositories
[87m 51s] yeah okay same user so and when you
[87m 54s] should
[87m 55s] eat things but if you want to stick with
[87m 57s] one and a half hour i think
[87m 59s] it's also fine to yeah to end
[88m 2s] commit here yeah that's just so just
[88m 4s] thinking about next time then
[88m 7s] when you when you think about when
[88m 9s] you're saying we're testing the
[88m 10s] relationships are you talking about the
[88m 13s] a level of test where we're literally
[88m 15s] using model factories to build those
[88m 17s] things or you're talking about going
[88m 18s] through the system
[88m 19s] coming out the other side and checking
[88m 21s] that it has
[88m 23s] i don't know it has repositories for
[88m 25s] example
[88m 27s] yeah i think like the
[88m 31s] the last test we have in this user test
[88m 34s] file
[88m 35s] so preparing some uh preparing a user
[88m 38s] filling in some repositories i don't
[88m 41s] have any factories
[88m 42s] because i load all the data from these
[88m 44s] picture files
[88m 46s] and yeah for
[88m 49s] testing relationships i have a model
[88m 51s] assertions
[88m 52s] class in my php unit composite assertion
[88m 56s] package
[88m 57s] which also has a asset related method
[89m 2s] i think uh
[89m 6s] let me not lie but i think
[89m 10s] adrienne casey had the idea to edit
[89m 22s] so
[89m 25s] aiden sorry aiden casey
[89m 28s] and
[89m 32s] so if next time then we start taking a
[89m 35s] look at perhaps those relationships
[89m 40s] yeah and how we might we might go about
[89m 42s] testing those
[89m 44s] on the organization that sets us up
[89m 46s] nicely
[89m 47s] um yeah we've got quite a few little a
[89m 50s] few little tests coming to the fore now
[89m 51s] though
[89m 52s] starting to build up isn't it yeah
[89m 54s] definitely
[89m 55s] so other idea would be to test the jobs
[89m 58s] in more detail
[89m 59s] yeah so we have this update user details
[90m 1s] test
[90m 2s] we can add the same one for update
[90m 5s] organization details
[90m 6s] and we have a lot of more
[90m 9s] jobs so like loading the repositories
[90m 13s] and loading contributors and
[90m 17s] these job tests will also test uh some
[90m 20s] relationships already
[90m 21s] so yeah yeah i mean just just as like a
[90m 25s] final point
[90m 26s] and this is i guess this is just a bit
[90m 27s] of a discussion to
[90m 29s] finish things off do you tend to
[90m 32s] like as you were building would you
[90m 34s] always go for the
[90m 36s] the level of um i want to test something
[90m 39s] that my application does
[90m 40s] or would you test kind of the bare bones
[90m 43s] of how things are put together
[90m 44s] and then build up to and this is how
[90m 46s] they work as a unit
[90m 49s] in general i'm a big fan of feature
[90m 51s] tests because they are
[90m 53s] a lot closer to how you
[90m 56s] think your application behaves and are
[90m 59s] also similar to
[91m 1s] if i click this button i want this
[91m 3s] happen
[91m 5s] and in general we don't think in unit so
[91m 11s] i don't care which services parts there
[91m 13s] or how many classes or units are
[91m 15s] implemented i just want to know if it's
[91m 18s] a
[91m 18s] controller or if i dispatch this job
[91m 21s] i want this final result and
[91m 26s] this is what feature tests cover very
[91m 28s] nicely for me
[91m 30s] yeah and with the new hdp
[91m 33s] fake which i absolutely love so it's one
[91m 37s] of
[91m 38s] really it's one of the best things
[91m 40s] laravel has added
[91m 42s] since yes so not reducing the value of
[91m 45s] all the other things but
[91m 47s] uh mocking and faking castle itself was
[91m 50s] a pain in the u.s
[91m 51s] and now it's hdp fake it's so easy it's
[91m 54s] it's yeah it's like you would expect the
[91m 57s] formula
[91m 58s] from level so it's so easy it's easy to
[92m 1s] understand it's easy to
[92m 2s] do and
[92m 5s] yeah we were able to get it yeah
[92m 8s] definitely
[92m 9s] i think as well obviously the
[92m 11s] application is already kind of
[92m 12s] it's built it's finished it's there it
[92m 15s] exists
[92m 16s] so you you're not wondering what an
[92m 19s] object is going to look like
[92m 20s] because you have it there in front of
[92m 21s] you it's not like you're building with
[92m 23s] tdd where you're thinking what's this
[92m 25s] going to
[92m 26s] how is this going to look when i'm
[92m 27s] finished i never do
[92m 29s] tdd in its plane definition so i always
[92m 33s] write the code first and says test
[92m 35s] second
[92m 37s] tend to be similar in my experience
[92m 40s] there's nearly
[92m 41s] no time uh when you're
[92m 45s] when your ticket or task to do is cis
[92m 48s] specific
[92m 49s] uh written out what say the author of
[92m 53s] the issue expects you to do
[92m 55s] and which attributes you have to add and
[92m 58s] how the route will be named and
[93m 1s] all these things so i think we all know
[93m 4s] it's if you do something
[93m 5s] you see another problem and have to
[93m 8s] solve it
[93m 9s] and as a sole developer
[93m 13s] writing the implementation first and
[93m 15s] this test second matches much
[93m 17s] more my
[93m 20s] my way of doing it because otherwise i
[93m 23s] would have to adjust to the test
[93m 25s] several times and separate those tests
[93m 27s] also it's the same like it
[93m 28s] like i iterate over the implementation
[93m 31s] and
[93m 33s] yeah yeah yeah i think i think
[93m 36s] i think you are muted again
[93m 39s] right
[93m 43s] but not terminators this time only only
[93m 46s] silence
[93m 48s] um yeah i think that um it kind of goes
[93m 51s] to show there's no like
[93m 52s] right way still silence is it is that
[93m 54s] only me i don't
[93m 56s] oh no can you hear me now he's coming up
[93m 58s] here saying that i
[94m 1s] i am talking it there still someone
[94m 4s] watching and could write in the comments
[94m 5s] if you can hear
[94m 7s] luke
[94m 11s] hi i'm not speaking so i can't hear you
[94m 14s] i have no idea what to say
[94m 16s] oh then oh then it's coming up for
[94m 18s] saying that
[94m 19s] see what someone says
[94m 23s] hi hey luke yeah okay synchro csu so it
[94m 26s] seems like it's
[94m 26s] it's me
[94m 31s] i mean i think let me just
[94m 34s] private chat you because
[94m 40s] right um so
[94m 44s] hopefully tom can hear us just before
[94m 46s] the end but i think we've kind of come
[94m 48s] to a logical conclusion for today's
[94m 51s] uh stream i hope you've enjoyed some of
[94m 54s] the cool stuff
[94m 55s] that's been added to past recently it is
[94m 58s] becoming honestly like an absolute joy
[95m] to work with
[95m 1s] um and going back to
[95m 5s] standard tests feels quite gross to be
[95m 8s] honest
[95m 8s] so if you've not yet picked test up in a
[95m 10s] real project to go and do it
[95m 12s] and uh yeah if this stuff that you you
[95m 16s] have a questions you have then
[95m 17s] get in touch with myself
[95m 21s] or you know you know if he's not busy
[95m 25s] on twitter and we'll try and pick it up
[95m 27s] um
[95m 28s] or any of the past team there it's a
[95m 30s] real cool community to be fair
[95m 32s] and obviously getting larger all the
[95m 34s] time seems to have
[95m 36s] spiked in interest a lot lately which is
[95m 37s] really nice so
[95m 40s] thanks for watching really appreciate
[95m 41s] you all being here again and we'll be
[95m 43s] streaming again next week
[95m 44s] so we'll see you there can you hear us
[95m 47s] again tom you
[95m 48s] i can hear you you're back i just want
[95m 50s] to say thanks
[95m 51s] uh very much for joining us again really
[95m 53s] appreciate that
[95m 57s] and it's been it's been cool to hang
[96m] and uh yeah i will see you all next week
[96m 3s] then
[96m 4s] yeah thank you for inviting me again so
[96m 8s] have a nice week and thank you for
[96m 10s] contributing the tests
[96m 15s] yo it's my pleasure it's nice to work on
[96m 17s] someone else's code
[96m 21s] all right take care guys and have a good
[96m 23s] evening or whatever time zone you're in
[96m 26s] enjoy it bye
[96m 29s] bye