awful programming: feature flags and suffering

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > awful programming: feature flags and suffering

«‹›1115 »

AggressivelyStupid: Jan 9, 2012

simble posted:

you just need enough reputation and you can do stuff

https://stackoverflow.com/help/privileges

if you dont have enough fake points it gets put into a queue where its approved/denied by morons with enough points

# ? Mar 21, 2019 02:03

Adbot: ADBOT LOVES YOU

# ? Apr 25, 2024 11:40

FlapYoJacks: Feb 12, 2009

Shaggar posted:

setting specific versions for libraries is pretty normal and allows you to build the same thing again reliably. allowing any version can introduce unexpected changes in functionality

There is nothing in the code that requires super specific versions, and even if a new version of a dependency library broke something, I would then agree to do a <= on the requirement. Or fix the issue and update the version.

# ? Mar 21, 2019 02:07

Jabor: Jul 16, 2010; #1 Loser at SpaceChem

okay, but how is carrying on using the version that worked yesterday suddenly going to break things?

# ? Mar 21, 2019 02:17

brand engager: Mar 23, 2011

Lutha Mahtin posted:

is this the first time you've encountered an inversion-of-control MVC framework? i had a similar reaction when i first tried to learn android but it makes more sense the more you read about how apps work in android. either way it's how the android API is structured so you will pretty much have to learn to live with it

the dependency injection stuff, it depends on a number of factors whether it is good or not (i don't have a lot of experience with this). there are a lot of valid reasons for an android app to use one but again it depends on the specifics

I've seen MVC stuff before but this setup feels excessively abstracted. We've got a settings activity that dynamically loads in some SettingsFragments and SettingsSomethingFragments inside there too. The injection stuff and magic view binding annotations on top of all that make it hard to see where the stuff that gets put into the view is even specified. If it was regular code I could use the IDEs find usages search or similar, but magic @ annotations seem to mess that up since they generate code at build-time. Anyways the stuff I was looking for ended up being in an xml config file that wasnt mentioned in any of the view or presenter classes

# ? Mar 21, 2019 02:30

floatman: Mar 17, 2009

I think when it comes to abstraction, people tend to focus on how the abstraction is done rather than what the abstraction achieves.
I.e. good abstraction: you get in your car and can accelerate, turn left, right, break etc without actually knowing how your car works.
Most software abstraction: you get into the car and there's two buttons only, "OFFICE" and "AIRPORT". you're now tasked to build a machine that will alternatively hit both buttons until you reach "SHOPPING CENTRE". You cannot build a third button because if you do, you might as well build a new car. A co-worker is on a tricycle screaming about microservices.

# ? Mar 21, 2019 02:38

brand engager: Mar 23, 2011

The list of menus and submenu options are all unchanging so this system of loading in fragments and then replacing them with new ones when you travel into the menu hierarchy didnt need to exist

# ? Mar 21, 2019 02:46

Dijkstracula: Mar 18, 2003; You can't spell 'vector field' without me, Professor!

YOSPOS: A co-worker is on a tricycle screaming about microservices.

# ? Mar 21, 2019 04:31

Sapozhnik: Jan 2, 2005; Nap Ghost

python and its community are trash

if they took interface stability seriously they wouldn't have created an incompatible new revision of the language and then immediately set about killing the old one that people were foolishly depending on

# ? Mar 21, 2019 04:47

DONT THREAD ON ME: Oct 1, 2002; by Nyc_Tattoo; Floss Finder

Dijkstracula posted:

YOSPOS: A co-worker is on a tricycle screaming about microservices.

# ? Mar 21, 2019 04:52

Symbolic Butt: Mar 22, 2009; (_!_); Buglord

Dijkstracula posted:

YOSPOS: A co-worker is on a tricycle screaming about microservices.

# ? Mar 21, 2019 07:14

DrPossum: May 15, 2004; i am not a surgeon

Chalks posted:

Commit messages that link to the work item/ticket they were assigned is always useful no matter how terrible the developer. When ever you find insane code, you always want to know what they were trying to achieve when they did it.

So many times I've been looking at some huge obvious problem caused by some commit and thinking that the code is complete nonsense, but when I go to check the work item associated with it I suddenly understand what they were struggling with - and the fix I was just about to make would have re-broken the scenario they'd fixed.

Gonna emptyquote this

# ? Mar 21, 2019 08:34

Zlodo: Nov 25, 2006

Shaggar posted:

setting specific versions for libraries is pretty normal and allows you to build the same thing again reliably. allowing any version can introduce unexpected changes in functionality

shaggar was right

# ? Mar 21, 2019 08:57

Soricidus: Oct 21, 2010; freedom-hating statist shill

Shaggar posted:

wales, Scotland, etc.. are more like counties than states.

please tell us more about how our own kingdom works, o wise foreigner :allears:

# ? Mar 21, 2019 09:01

gonadic io: Feb 16, 2011; >>=

can't wait to get my Scottish passport within the year, giving me the full trifecta

Back to my twitter stuff: since there's no processing I'm not actually convinced doing the auth serverside and then the calls clientside will really save me much bandwidth. Like how big is a request with no body?

E: I guess one advantage to doing it clientside would be that the user auth can stay loaded in the js, no need for it to be on the server

gonadic io fucked around with this message at 09:37 on Mar 21, 2019

# ? Mar 21, 2019 09:24

Aramoro: Jun 1, 2012

Soricidus posted:

please tell us more about how our own kingdom works, o wise foreigner

Lets give it a year and see how that holds out.

# ? Mar 21, 2019 12:44

mystes: May 31, 2006

Sapozhnik posted:

python and its community are trash

if they took interface stability seriously they wouldn't have created an incompatible new revision of the language and then immediately set about killing the old one that people were foolishly depending on

Where "immediately" means "over 10 years".

# ? Mar 21, 2019 13:21

champagne posting: Apr 5, 2006; YOU ARE A BRAIN
IN A BUNKER

if only the death of python 2 had been immediate

# ? Mar 21, 2019 13:24

AggressivelyStupid: Jan 9, 2012

Soricidus posted:

please tell us more about how our own kingdom works, o wise foreigner

not very well

# ? Mar 21, 2019 13:49

cinci zoo sniper: Mar 15, 2013

AggressivelyStupid posted:

not very well

depends if its main output product is or isn�t laughter

# ? Mar 21, 2019 13:50

cinci zoo sniper: Mar 15, 2013

also angloboos need to remember that england etc are countries in interpretation of local laws only, where uk is centre of the galaxy. for everyone else uk is one country, at least until march 29

# ? Mar 21, 2019 13:54

Carthag Tuek: Oct 15, 2005; Tider skal komme,
tider skal henrulle,
sl�gt skal f�lge sl�gters gang

one would think i wrote these forums in flask in a half hour

# ? Mar 21, 2019 14:27

Finster Dexter: Oct 20, 2014; Beyond is Finster's mad vision of Earth transformed.

cinci zoo sniper posted:

also angloboos need to remember that england etc are countries in interpretation of local laws only, where uk is centre of the galaxy. for everyone else uk is one country, at least until march 29

wait what's happening march 29? is this something to do with brexit?

# ? Mar 21, 2019 17:55

cinci zoo sniper: Mar 15, 2013

Finster Dexter posted:

wait what's happening march 29? is this something to do with brexit?

yeah, the current deadline. if nothing changes they are gtfo�d from eu with no agreements on that day

# ? Mar 21, 2019 18:09

Finster Dexter: Oct 20, 2014; Beyond is Finster's mad vision of Earth transformed.

cinci zoo sniper posted:

yeah, the current deadline. if nothing changes they are gtfo�d from eu with no agreements on that day

yikes

good luck, uk friends :ohdear:

# ? Mar 21, 2019 18:35

Gazpacho: Jun 18, 2004; by Fluffdaddy; Slippery Tilde

Holy cow I sure do like programming with strings. Whose dumb idea was it to invent other types?

# ? Mar 21, 2019 18:36

mystes: May 31, 2006

Gazpacho posted:

Holy cow I sure do like programming with strings. Whose dumb idea was it to invent other types?

Welcome to the terrible programming thread! Languages where everything is a string sure are great!

# ? Mar 21, 2019 18:39

elite_garbage_man: Apr 3, 2010; I THINK THAT "PRIMA DONNA" IS "PRE-MADONNA". I MAY BE ILLITERATE.

The real fun begins when you have to come up with your own string implementation.

# ? Mar 21, 2019 18:50

Doom Mathematic: Sep 2, 2008

Gazpacho posted:

Holy cow I sure do like programming with strings. Whose dumb idea was it to invent other types?

This is called "stringly typed" programming.

# ? Mar 21, 2019 18:53

Symbolic Butt: Mar 22, 2009; (_!_); Buglord

this may sound dumb but I like ruby's distinction of strings and symbols

# ? Mar 21, 2019 18:58

gonadic io: Feb 16, 2011; >>=

Symbolic Butt posted:

this may sound dumb but I like ruby's distinction of strings and symbols

Scala technically has this too

code:

val fooVal = "foo"
val BarVal = "bar"

def butt(input: String): Unit = {
  input match {
     case BarVal => println(s"found bar")
     case fooVal => println(s"found foo")
     case _ => println(s"found nothing")
  }
}

butt("foo") // prints found foo
butt("bar") // prints found bar
butt("baz") // prints found foo

# ? Mar 21, 2019 19:24

Carthag Tuek: Oct 15, 2005; Tider skal komme,
tider skal henrulle,
sl�gt skal f�lge sl�gters gang

wtf

# ? Mar 21, 2019 19:31

gonadic io: Feb 16, 2011; >>=

These days you do at least get the warning: suspicious variable shadowing by wildcard match 'fooVal' or something

# ? Mar 21, 2019 19:41

Carthag Tuek: Oct 15, 2005; Tider skal komme,
tider skal henrulle,
sl�gt skal f�lge sl�gters gang

but why does butt("baz") match foo??

# ? Mar 21, 2019 20:02

ComradeCosmobot: Dec 4, 2004; USPOL July

Krankenstyle posted:

but why does butt("baz") match foo??

I assume BarVal is interpreted as a constant value and therefore the case statement matches only when exactly equal, while fooVal is interpreted as a variable name, and bound to �baz� as a result of Scala�s pattern matching case statement.

Hard-coding variable/constant naming conventions as a behavioral difference seems dumb and bad to me in a way syntactically significant white space could only dream of but hey, forget it Jake, it�s Scalatown.

# ? Mar 21, 2019 20:14

gonadic io: Feb 16, 2011; >>=

Krankenstyle posted:

but why does butt("baz") match foo??

This isn't the horror, although it is a bit surprising if you haven't done much pattern matching. It's the same as if that line was "val fooVal = input", you're making a new value which shadows the old one.

The horror is that this behavior changes depending on if the name starts with a capital or not since that's how a symbol is specified. As comradeCosmobot said, symbols (and string literals) match exactly.

# ? Mar 21, 2019 20:18

cinci zoo sniper: Mar 15, 2013

ComradeCosmobot posted:

I assume BarVal is interpreted as a constant value and therefore the case statement matches only when exactly equal, while fooVal is interpreted as a variable name, and bound to �baz� as a result of Scala�s pattern matching case statement.

Hard-coding variable/constant naming conventions as a behavioral difference seems dumb and bad to me in a way syntactically significant white space could only dream of but hey, forget it Jake, it�s Scalatown.

now once more but slower

# ? Mar 21, 2019 20:21

gonadic io: Feb 16, 2011; >>=

gonadic io posted:

These days you do at least get the warning: suspicious variable shadowing by wildcard match 'fooVal' or something

In fact you might even get another warning that the third case can never be reached

# ? Mar 21, 2019 20:24

gonadic io: Feb 16, 2011; >>=

cinci zoo sniper posted:

now once more but slower

The fooVal at the top and the fooVal in the match are entirely unrelated. This is fairly typical behavior in flangs, and if you replace the match with a val or variable declaration, I would say most langs would behave the same

gonadic io fucked around with this message at 20:30 on Mar 21, 2019

# ? Mar 21, 2019 20:28

ComradeCosmobot: Dec 4, 2004; USPOL July

the hosed up part, as gonadic io pointed out, is that the only reason it doesn�t work for BarVal is that the capitalized first letter (and only the capitalized first letter) makes it a symbol/constant and therefore it won�t be reassigned by the pattern match

most other languages do something sane like having an sigil or explicit quoting mechanism like the apostrophe in �BarVal in Scheme (and Lisp?) or the colon in :BarVal in Ruby.

Scala, being Scala, happens to make this call based on whether or not the first letter is capitalized.

# ? Mar 21, 2019 20:40

Adbot: ADBOT LOVES YOU

# ? Apr 25, 2024 11:40

cinci zoo sniper: Mar 15, 2013

gonadic io posted:

The fooVal at the top and the fooVal in the match are entirely unrelated. This is fairly typical behavior in flangs, and if you replace the match with a val or variable declaration, I would say most langs would behave the same

oh right I get what you mean

# ? Mar 21, 2019 20:41

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > awful programming: feature flags and suffering

«‹›1115 »