Player is loading...
Refactor all the things!
Daniela Sfregola, organizer of the London Scala User Group meetup
Wednesday, June 12, 2019 · 3:31 p.m. · 44m 01s
Learning the syntax is just the first step towards mastering a new language. Idiomatic expressions and good practices must also be adopted to produce code that is both readable and performant. In this talk, we'll analyse snippets of code and highlight common Scala anti-patterns that make them difficult to understand. We'll also discuss how to refactor them to improve their readability. At the end of this session, you are going to be able to recognise code smells and refactor your code to make it easier to reason about and maintain, as well as avoid common pitfalls and possible bugs.
Embed
Transcriptions
Note: this content has been automatically generated.
00:00:01
okay welcome guys i area guideline contracts yeah as you got the the latch
00:00:09
i can't get because your candidate for the sock fish it's not gonna be the usual talk to you
00:00:17
guys are used it because guess what you gotta do
00:00:21
it all the work so why is this talk about
00:00:25
this joke about um how you can in the same well
00:00:31
why a perfectly your code and hopefully make your code better
00:00:35
and i'm like oh yes i'm the now and i i'm
00:00:42
a software engineer and typing brightness colour for a while um
00:00:47
but i haven't always of return star are actually started my career with java
00:00:53
and accidents caught hangs
00:00:59
well it was an amazing journey wasn't always that easy
00:01:03
um so this is a collection of all the things that are learn over the years and
00:01:10
collection of things that i've seen other people do i i'm i'm also writing a buck
00:01:16
um uh and my yeah these are give me a code that gives you forty percent off
00:01:21
on what the books not just might that that's more in stuff we wanna talk about curves are
00:01:28
yeah you're right some beautiful again piece of corrode and and you know um
00:01:37
you think is also um and and one of your
00:01:39
teammates starts reading it and they think that it's really
00:01:48
not right so there is this that means speech at
00:01:51
that tells you that the finding good code is really difficult
00:01:57
so depending on how many swear words you use um that could be a good better off
00:02:05
you know how good your code is but even with the basket you'll still have a couple of moments where
00:02:13
second tape does this code to write
00:02:18
said it's really important art to try and
00:02:24
right corrode that um it's the best that we can write any particular
00:02:29
when you're learning a new language is not just about the sink docks
00:02:34
he's also by the humans style that that's specific language has
00:02:39
which can be tricky some time to master in particular the beginning i
00:02:47
can i feel that then uh to us well but a while ago
00:02:51
i had the opportunity of looking at some off old clothes that um
00:02:58
i've written in one of my previous companies oh yes i write this part
00:03:08
that should be an idea get blamed because come up everybody both that everybody wants to know who's to blame and then
00:03:17
it and that was me weights with um embarrassing
00:03:23
ish but is it that making that myself six
00:03:27
years ago or six months ago or six weeks ago i was right in a different type of crowd
00:03:33
so my solve now he's up better co there there was
00:03:40
to freely second half and this is why i think talking about
00:03:46
could party giving people little tips on how to improve code having
00:03:52
the courage to stand up and say okay tell me if discuss good is
00:03:55
really important because it makes you a better person and attending events like place
00:04:01
i always had a friend in at the end of the data either wanna spend the entire night
00:04:07
going we fucked them into your code base because i've learned so many nice tricks i don't wanna
00:04:13
try them all to refer doing is the way of improving yourself as a
00:04:19
developer but obviously you know most of the time we don't work our by ourselves
00:04:27
so we always have to make our teammates happy in particular if one
00:04:34
of our team eight could arm for to know what i've so always good
00:04:41
as if the guy you whip and that's maintain your code will be violent rough who knows where you live
00:04:49
right we want to make our team life it's easy
00:04:53
as possible but i i'm not not only that um
00:04:57
when we write bad code we inject a lot of complexity that is not necessary
00:05:04
so there are two types of complexity wise because you know the problem that we are
00:05:10
actually trying to solve is complex to your one is just accidental complexity
00:05:16
it's just us making the code more convoluted than necessary
00:05:22
so off we are gonna play again today but isn't that great game of all
00:05:29
so that's where you guys come in there is called what's wrong with my code
00:05:36
we're gonna have time out um when we're gonna i'm gonna show you a snippet of code
00:05:43
and i'm gonna ask you what's wrong with my code and then we're gonna discuss
00:05:48
it together and see you know how we can make it better there than it
00:05:53
the different that was also difficulties some will be he's here
00:05:57
you know there's some will be a little bit more tricky and there might be
00:06:02
problems related to performance readability or simply code
00:06:07
that doesn't really look sky lucky enough okay
00:06:13
so are we ready for the first round yeah i gots um okay let's start
00:06:20
this high actually yeah i've done this not in production unit test it
00:06:25
to find out how the database right and you have a function connection
00:06:31
followed some configurations and then calls a function that
00:06:36
uh in eastern instantiate the connection to the database
00:06:42
so what about
00:06:45
if
00:06:50
if if if not i hear the word ass sorry i forgot to say n. so i experience don't worry about
00:07:00
it all the slides will be online and i know it's gonna be really to afford be um mike guy so
00:07:06
reach out and i would repeat all the answer to this is gonna be fine
00:07:11
so it must be fun for you as well so don't worry about it okay cool i i'm gonna give you at six
00:07:18
sarah had this that's it's and i was running it
00:07:23
anyone else you know the memory was going up enough enough happen it
00:07:28
was like how things so how much memory doesn't make any sense certain zealots
00:07:38
that every time i was calling the function connection and i was in stop wishing menu connection to the database
00:07:45
and the solution to fix this is to turn the function into of all that would avoid a lot
00:07:52
of memory allocation okay so that the da i'll take away is that every time you write the faction
00:08:02
and this function as an no argument you
00:08:06
should ask yourself if these function already is
00:08:09
this actually of all can i avoid a locating memory for this over and over again
00:08:18
the second
00:08:21
lots of but i still have the speed of workload that has
00:08:27
an a. p. i. cold my p. i. that has a function
00:08:32
that is called my into point and any calls to other functions
00:08:37
uh the function and another how functions are so let's row or it might be a
00:08:55
oh it it if the answer is it certain
00:08:59
oh it's a problem of ordering functions yes but
00:09:05
i'm looking for something more specific any other answers
00:09:13
like i said the problem is that these functions are not meant to be
00:09:19
public right these functions are out there functions that access my a. p. i.
00:09:27
right so i come from a java backgrounds and
00:09:33
in java at least when i used to write java d. d. for
00:09:36
taxes modify your was protected so you actually writes because by putting the function
00:09:43
within the sayings code will you are limiting the scope of the function
00:09:48
um but they come from the top background which i believe is the
00:09:53
same background of the majority of the star developers i separate use
00:09:57
not to worry about it right i don't let anything and that's it
00:10:01
but by so every single function they you over ah eat is gonna be topic
00:10:08
which has so i fax one it makes your e. p. i. a lot more complex years
00:10:14
because when you're user calls that functioning press the
00:10:17
ought is gonna have a much longer list of methods
00:10:22
very nice to pick from and this current is the incremental compilation
00:10:29
so i think that's a compiler wendy's tries he needs to recompile
00:10:34
the public matt uh the house to such where this probably mattered
00:10:39
what else this puppy method disuse body from actually spread the search is gonna
00:10:45
be a lot faster so i'm just saying why don't we have here is that
00:10:53
i know it's the hunting uh no it's easy to forget that but make sure
00:10:58
you use the most restrictive boxes modify your the makes you occurred
00:11:04
work so if something should be for that make it right it's it's something should be protected
00:11:10
make it protect because it's gonna help your compilation and your uses life okay
00:11:18
well it's it's rock and this function
00:11:24
that is called my function but that's and stuff
00:11:28
and i am a has a data that is again it's
00:11:35
what's wrong with my code
00:11:40
yes you guys got it it was the the return type it right yes that's an easy mistake to make
00:11:47
so actually these facts uh needs not actually returning data is just returning unit
00:11:53
and this is the classic copy and paste things like i it happens to everybody
00:12:00
what these functions should actually read had is you can
00:12:03
add because this will make shirt that data is not discarded
00:12:08
pays return s. value of the function my function is is that there are a lot of these tricky parts
00:12:16
that can easily be detected by using something that is called flat compilers
00:12:23
uh it's something that you can enable a within your field
00:12:26
that is b. t. file and i strongly recommend you to
00:12:31
enabled them because they cats a lot of these silly bites they
00:12:36
can happen and they're really really difficult to notice so r. i. e.
00:12:42
link you are made in noise article by robin reese wary lace all
00:12:47
the fact that we have available that we should all be using okay
00:12:54
the next wow it's we have these neat that's got a then we half of our a
00:13:01
that is the string that contains the string happy times
00:13:05
but they and then we do a bus scallop days
00:13:12
what's wrong with this code
00:13:15
i'll give you guys thirty seconds of what the stock is gonna be a winning ten seconds
00:13:23
okay get if you know the answer is string concatenation and the plastic yes absolutely so
00:13:33
this command is not wrong it does what is meant
00:13:37
to to live but string concatenation is a normal that
00:13:42
frost what you should do his dad is use the string interpret eight
00:13:48
or operator that is the little ice that you see before the string
00:13:52
in terms of performance this is gonna be a lot faster
00:13:57
so as a general rule in style not
00:14:00
use the plus the always use the um
00:14:06
the string operator okay cool
00:14:12
next rounds a set of our half
00:14:17
that's correct simple script that needs to be verified that exactly one on
00:14:23
argument as being passed so i receive somehow all the
00:14:29
least of all the arguments that i've received any do if argument has
00:14:34
the lamp of zero than i for an exception no parameters found a if
00:14:42
arts the lymph has small them one then afro an exception
00:14:49
too many arguments phones otherwise i do art style head
00:14:57
what's wrong with like a it no it's no no it the
00:15:05
too many people to okay let's go is why what's wrong with this guard
00:15:13
oh it's a um there's there are problems
00:15:18
with these code are you are you are close
00:15:21
uh and the biggest problem that i have we discovered is that had
00:15:27
dot had dog get an option are
00:15:31
really really really unsafe functions to use
00:15:35
in style reason why they're unsafe is that if you call dot had on
00:15:42
a collection that is empty or if you call don't get on the norm
00:15:47
shown that exempted they will for a nasty exceptions exceptions right so like this code
00:15:56
i probably every yeah yeah i'm after maybe hide the
00:16:00
night of drinking and i have a production her or and my boss is going up made
00:16:06
because everything is broken i need to go
00:16:09
into makes myself that the if statements before all
00:16:14
logically got antsy there i only have one elements
00:16:22
which is fine but in the requires in human intervention around
00:16:28
reasoning on the lack of right i have to thank that if it's not zero if it's
00:16:35
no more than one then it's one so that the
00:16:41
why over factoring this is using bottom matching like you
00:16:47
can patent much on oh right he and you can say
00:16:52
if i have the empty are righty de and high for the exception if
00:16:58
i have an array with exactly one had a mentor and returned the other man's
00:17:03
you know the other cases are returned the other exception this is a lot more readable
00:17:08
then trying to figure out all d. if animals
00:17:13
statements in particular because this isn't simplification i've seen functions there where
00:17:21
two underlines long there where one single if
00:17:25
else if else if all's if else else um segment place if
00:17:38
if you split might okay yep yeah time you i'll get to that in one second
00:17:46
um so the take away and is the is please dawns don's get
00:17:54
don't don't use that had dined used all get either use that option
00:18:01
or these use pattern matching of any of the i order
00:18:05
functions um there are available and now to the bus rounds
00:18:13
this one came directly from my to know their ski
00:18:17
um i wasn't sage you don't and giving this talk
00:18:23
and then suddenly marketing race that and saying i don't hit the something else running scared
00:18:29
and i know they serve learn something and so what is wrong
00:18:34
with discovered if i give you guys mean it up with him
00:18:45
okay okay
00:18:51
if
00:18:54
so the answer is are we should use the ensuring that it is not my answer
00:18:58
is a lot more simpler than that anything else will like to give it a try
00:19:09
yes the types i. e.
00:19:14
uh the typeset not cracked up but uh you might be right yeah somebody
00:19:20
before indy audience was shouting either and i absolutely agree with you but it's
00:19:25
not what i'm looking for i'll give you guys a little hint if it
00:19:35
a right to cave so apparently for
00:19:40
the compiler days is not as powerful moment
00:19:45
as a base is so first they said there wasn't aware of uh
00:19:54
it's always advisable to put the happy case first you know after matching
00:20:02
so it is a small formant serve thank you marty i wasn't aware of that and i don't something at fish
00:20:11
so my take away is that are you know can't
00:20:15
really is awesome right you can always always learn something
00:20:21
from showing your code to other people i'm not saying that you should
00:20:25
invite marked no that's good to utah and the bigger rooms that you're
00:20:30
cut these rock because there that a little bit traumatic i have to
00:20:34
admit that had a hard on something you why wasn't plates that's awesome
00:20:40
but i i get ready for the next not yeah yes awesome
00:20:46
we have a case to us a that takes a an argument that it's an x.
00:20:52
tight ends and then we'll have a class the if that as two parameters and x. and y. both
00:20:59
of type int and they extend and a of
00:21:03
x. but give you guys thirty seconds difficulties one left
00:21:13
it yes but the point is absolutely right it's dawned x. downs
00:21:21
case classes let me say this again i don't want
00:21:24
to extend the case classes because it's gonna be a mess
00:21:28
i'm gonna show you why because he likes than the case the us
00:21:34
then apparently if you compared to efficiency is you could break equality
00:21:40
so i say this is a disaster because
00:21:44
the quality is how we used a lot of
00:21:48
internet of scar which means that potentially you could
00:21:52
retrieve do wrong element from a mop or other
00:21:58
really really tricky parts to figure out if just don't do it doesn't right
00:22:05
are half time on forcefully to fully explain why this is
00:22:10
really a deeper at re read but it would feel like
00:22:15
and also we will have nickel us there is sitting right in front here
00:22:20
that tomorrow is gonna explain us a
00:22:23
little bit more why extending pace classes are
00:22:28
really pretty bad sorry my solution is to
00:22:33
play is always the clarity okay stresses the final
00:22:37
because it's gonna avoid a lot of headaches so let them marketing was in the audience we asked him
00:22:45
why it is that says i'm not final by
00:22:48
design and it turns out that there are some um
00:22:53
i it's crazy thing he started compiler code
00:22:57
that requires case classes to be not final
00:23:01
sell i present to you could wrap flow
00:23:05
all off she is not final k.'s classes uh_huh
00:23:12
okay martin and i see no my case argued about it
00:23:16
being the start complied a meaning are you pushing to scouts go
00:23:21
i'm not so please don't use case classes that are not find all
00:23:29
if you are not not as key or you are working
00:23:33
on the east dallas colour report and you know what you're doing
00:23:37
there is a really good chance that you probably need to put a final there
00:23:45
so these guys don't extend case classes because it's it's really tricky to it and figure out what's wrong
00:23:53
okay so to take away in case you didn't get a please may
00:23:58
all your case classes find all unless you really really know
00:24:03
what you're doing and if you really need to extend that
00:24:07
don't make them at case plastic case classes does something that the compiler
00:24:14
did you can rewrite the same methods their case class has just monday
00:24:22
okay the next round we have these beautiful piece of code
00:24:29
that he's a function that is cool or something it
00:24:33
takes two parameters a billion there is gold enable a
00:24:39
and now when there is called any will be and you returned whatever and unit
00:24:47
yeah so what's wrong with my kurt i would give you guys
00:24:52
a minutes otherwise we're gonna be here if done in ten minutes
00:25:03
t. v. audiences shouts name parameters absolutely so when you look at this signature
00:25:11
sorry when you do not don't look at the signature in just look at the function implication
00:25:18
just something true false what what that's true false means if i don't i i need to look
00:25:25
at this signature of the function to remember that the
00:25:29
first volume value is the city to enable i. e.
00:25:33
and the second bullion value is associated to enable be any particular factoring
00:25:40
and it's what them that's that it's gonna be a park
00:25:47
so my suggestion is to always use name parameter is full billions
00:25:55
always because they are so easy to mix up yes we have a question
00:26:10
with
00:26:14
so the question the if if i'm gonna be simplify a
00:26:17
if things will uh be bettering doughty a high hopes of um
00:26:25
these uh just under half is not to do these
00:26:28
just for millions but to do it every time you have
00:26:33
more than one parameter there he's at that has the same type of another one
00:26:40
and i seem functions that take time parameters and all of them are really and
00:26:47
which are really really tricky to use uh if
00:26:50
you have more then free parameters of the same type
00:26:56
please do you consider making something more tight said if you could
00:27:01
have an object or a case class that drops all these billions
00:27:05
um because this will make your code safer and easier to read as well
00:27:11
so the take away ace afraid it's amazing featuring star
00:27:17
use fully name parameters because it makes the code a lot more readable
00:27:26
okay ready for the next was yes sir where um of
00:27:31
this function that a scroll there's something there it turns and and and this function calls
00:27:39
three functions a. b. and c. a. plus b. plus c.
00:27:46
then you have a function that is private a the returns forty two
00:27:51
the function they spread that that does some read some stuff from
00:27:56
a database then we wait and and then we have a private function
00:28:01
see that returns plentiful i. d. v. guys i mean it's but what's wrong with discovered a
00:28:18
so so that you're from the audience the show the news for comprehensive exam periods there is
00:28:23
yep everywhere yes yes yes yes yes it's uh
00:28:30
the the idea of features in particular for beginners a it
00:28:34
sometimes a little bit difficult to grasp the general idea is that
00:28:38
don't lie about your types of eva function does some to synchronise computation
00:28:46
'kay to return type of function it's a clear indication
00:28:51
of what these function das so by blocking the future
00:28:57
into line there has been a lighted you'll
00:29:01
are basically asking of fried to step aside it's
00:29:07
and do nothing
00:29:11
instantly okay i'm back i'm done i'm gonna continue and continue my computation
00:29:19
so this is terrible because you are wasting a lot of computational power
00:29:26
so the ah about your types if the function be does some uh synchronise computation
00:29:32
it was a few sort of and then has somebody was shouting from the audience how to deal with features
00:29:41
useful action that's a much better way of extracting
00:29:47
the result batteries written by your future and combining it
00:29:53
and if you really really really really need to block future because at some
00:29:58
point you need to have the answer try to do it as late as possible
00:30:06
so the take away in this case it is not uh have a have a lock a future
00:30:15
if you are blocking future there is a really good chance that you're doing something wrong
00:30:22
okay next round said these these
00:30:28
um function factorial are that has
00:30:34
a recursive function very school nope that calculates for the orioles
00:30:41
by the way if you don't people nazi now you'll have a hundred percent coverage
00:30:45
on recursive questions during interviews so yeah that's that's good what's wrong with this got
00:30:55
oh okay zero lot work because we pass one
00:31:02
i'm not sure if that's the answer that i'm looking for it might
00:31:05
be the micro these are wrong and yet to happen before we make area
00:31:10
uh but it's not what i'm looking for but yet
00:31:13
uh let's assume that my my factorial is actually correct
00:31:19
i i take ah yes the audience is she is shouting to record an annotation
00:31:29
so when i go that means up the that recursion annotation
00:31:35
was struggling an optimisation in the compiler out that no no
00:31:41
the scroll bar there is a lot smarter so he's able to detect
00:31:46
that uh records it's function can be optimised even if there really is no
00:31:52
uh to record steve annotation so that's awesome our
00:31:56
act uh humans are not as good as the compiler
00:32:01
so every time one of my team mates right the recursive function
00:32:05
i want to make sure that is the recursive when you hop it's a
00:32:11
recursive function you are sure that it's not gonna blow up your stock so
00:32:16
you know it's safe to use in production uh the good thing is that if you are on these
00:32:23
annotation annual function is not to require see if the code will not compile
00:32:29
so even though this colour compiler doesn't
00:32:32
need these annotation to trigger the optimisation
00:32:37
because the petite optimisation i was still suggest to use it because it combines your teammates
00:32:43
and yourself that you are fighting a function daddy's a stack safe so are your writing up
00:32:53
recursive function please please make sure there is there a recursive
00:32:59
meaning that it's safe to use in production and use the
00:33:03
'kay wreck annotation so that you can prove that f.'s actually a case okay
00:33:13
next route so we have been so fireballs
00:33:18
and we have my map fat that has a type
00:33:24
o. option of uh my whole the string to string
00:33:28
we have my list that is an option of list the strain
00:33:34
we have my option option that has like the option options right
00:33:39
anyway half my body and that has a a is an option of body
00:33:46
so
00:33:48
what's wrong with mica
00:33:53
some in the oh that's yeah see shouting yeah you could use our ideas because it's hard to read
00:33:59
yes i really do it but that's not the point there oh yeah
00:34:02
arm to look in full but yeah that that could be a good improvement
00:34:10
type arrays are uh_huh yeah oh it's a lot simpler life then
00:34:17
oh yes option of map option also laced option
00:34:22
of option option of wood and are we don't it's
00:34:25
do we really need to distinguish between i receive poles and i received no value
00:34:33
do you really need to distinguish between have no laced
00:34:37
versus i have an empty list don't get me wrong there are gonna be cases where
00:34:42
y'all business requirement will require you to do that but we have this kind of types
00:34:50
'cause yourself is the is the simplest type that satisfies my requirements
00:34:56
right beside the option of mock is the same
00:35:03
c. a. is equivalent to empty mop and so on and so forth question from the audience
00:35:23
so the comment from yeah disease that option of billion
00:35:25
actually have a body case in particular when you read
00:35:30
from five and i absolutely agree with you but that is
00:35:33
probably one percent of the cases that i've seen are so
00:35:37
yes i agree with you there are specific cases that uh
00:35:42
require an option proper or to transform these in too much right
00:35:47
uh what i'm arguing is um make sure that your types are
00:35:54
simple enough don't be lazy and just although
00:35:59
complicate your types because then you will pay
00:36:02
a big price for it but you are absolutely right if you're reading
00:36:05
from a file you my actually want to keep an option of one okay
00:36:12
so in this case the take away ace don't be lazy try to think about your tights
00:36:20
and if possible try to simplify them if you're starting to have too many
00:36:27
rappers then there is probably something that is not quite right it's probably has the
00:36:34
he did the the gentleman you've been suggested maybe you need to try
00:36:40
right uh but when you start a nice thing to
00:36:43
many types uh it could be a design issue um
00:36:50
cool um so the dogs were all my mountains
00:36:56
and this is the summary of all the things that we have seen these obviously are no not uh
00:37:05
on global rules there are always exceptions um but these are my suggestion
00:37:12
uh for people that maybe just thought in the language of things to look out for so we
00:37:18
discussed about um function that takes no parameters and
00:37:25
the fact that sometimes they are located too much memory
00:37:29
in those cases maybe you should consider transforming your function into a file
00:37:35
we have discussed the about the fact that depending on your backgrounds
00:37:40
you might want to forget about our all access modifiers
00:37:45
and you should always try to make your function
00:37:49
as restrictive as possible it takes one second to make a function public
00:37:54
but try and control the visibility of your code because he will make
00:38:00
the combination fast the n. is gonna make
00:38:03
your um code easier to use for other people
00:38:10
enable compiler flags because there are some parts of the really see later really
00:38:16
easy not to see and they can really really save you so really strong you
00:38:22
i strongly suggest you to go and look at the problem laurie's article about it i am i don't use
00:38:31
don't use string concatenation if i think i said the ops the uh sorry um
00:38:37
be careful when you are combining strings always the big
00:38:42
string interpretation over string concatenation because it can really have
00:38:47
an impact on the performance of your problem uh if you're concatenating to strings is not a big deal but
00:38:54
if you scale this to all the big number of strings then obviously um
00:38:59
just can have an impact w.'s to hide don't use dog
00:39:05
gets because even the your coat my still work you are making
00:39:12
humans life more difficulties because people will have to
00:39:17
reason about the lines before you function call and complains then self that
00:39:22
in this context stuff had dock at is a safe operation to to
00:39:30
i'm always always always always make your case gases final any few
00:39:37
needs to extend something that you region if thought is the case costs
00:39:43
don't make it as a case does make it has a regular fuss when half parameters to have the same type
00:39:51
don't risk it it's really easy to just the fact that the function and create
00:39:56
really difficult attacked bucks to use fully name parameters
00:40:01
that will guarantee that the parameters are passed consistently
00:40:06
just by their names rather than their position of decoration
00:40:12
don't flocking futures it's really really inefficient
00:40:17
try to always have types that actually represent the computation that you are
00:40:22
uh performing and if you have a computation that is a synchronous
00:40:27
well done all the computations that use they will need to be a synchronous
00:40:33
and if you really really need to block block is the latest possible
00:40:40
use their records even the patient even if the compiler you smarter than
00:40:44
us any well anyway optimise your code we're not as much as the compiler
00:40:50
and it's a nice way of proving that your body's safe to use
00:40:56
and finally tried to simplify your talks off
00:41:01
just so that you are not gonna pay the cost of maintaining really complex that signature later on
00:41:08
so i hope that this was useful um effective remote for listening and for participating
00:41:15
i hope it was fun i am gonna post this light on my tutor account so if
00:41:20
you want to have a look you can go into that and uh thank you very much
00:41:32
exactly four questions so if you have some just freeze you may come to you okay sure see
00:41:40
yeah just version of the program is when you are even before yeah we used to use it you just use
00:41:51
research because shoe you're you should we be review shows you should interest usual
00:41:57
you should request absolutely i mean writing good code thousand
00:42:03
doesn't allow you not to write tests i hope they that was not the message
00:42:08
uh yes absolutely we still need to write unit test integration tests
00:42:12
whatever we need it does that uh avoiding certain functions avoids um
00:42:20
avoid the uncertainty when you are you the review decoder
00:42:24
re factoring occurred but you're absolutely right now using the
00:42:29
um the metadata had or to get doesn't mean that
00:42:32
we don't have to write a unit test to guarantee that
00:42:36
yeah things are actually working so you're absolutely spot on writing
00:42:39
did good code doesn't mean that you don't write tests of
00:42:45
so about a seizure production functions i got
00:42:48
beaten just a collection max correction mean a soul
00:42:55
no one knew him get or just be very very careful
00:43:00
oh yeah absolutely i mean a a stylus colour to thirteen very i
00:43:05
i came out like two days ago uh that as a matter of
00:43:09
fact are that there are still a lot of functions that that there
00:43:13
are not safe so as you were saying if you two don't max
00:43:17
uh on a on a list that is empty will bob absolutely sir dot
00:43:22
had don't get are not the only ones are that you need to be um
00:43:28
aware of but are definitely the most common ones
00:43:33
dot size an ethernet laced guy is alive is getting scary now yeah s. the
00:43:39
absolutely didn't part about there but that that's terrible yes absolutely that's question someone wants
Conference Program
A Tour of Scala 3
Martin Odersky, Professor EPFL, Co-founder Lightbend
June 11, 2019 · 5:15 p.m.
8337 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
