terrible programmers: my god, it's full of chars

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > terrible programmers: my god, it's full of chars

«‹›1446 »

Valeyard: Mar 30, 2012; Grimey Drawer

Ploft-shell crab posted:

are you defining the datatype? you can use record syntax to give your fields names.

naming them yes. the problem was that I have a map, with strings as keys, and datatype X as values. but when I print out the map, it prints out datatype X as a monstrosity.

for no, https://zuttobenkyou.wordpress.com/2011/02/27/simple-haskell-data-pretty-printing/ , this was quick enough to setup and is ok to use for tests so far

Valeyard fucked around with this message at 05:23 on Nov 20, 2014

# ? Nov 20, 2014 05:21

Adbot: ADBOT LOVES YOU

# ? May 10, 2024 01:01

Space Whale: Nov 6, 2014

Through CTCI, brain teasers, code rust, being good at selling myself, and being acceptable at code monkey things, I managed to get myself a mid level job. Woo hoo!

As such I'm now exposed to stuff like the actual ISS server and the messaging queue and other things that are beyond just "poke around project, mash F12, and do what BA says, and make sure stylecop is happy." I actually feel that "oh poo poo I'm a faker" feeling again, which means I'm learning.

What do you DO to really get all this stuff besides ask questions, stay late and buddy up to the IT guys?

# ? Nov 20, 2014 05:51

Shaggar: Apr 26, 2006

uncurable mlady posted:

Mac OS X is the worlds best operation system

lol

# ? Nov 20, 2014 05:53

MeruFM: Jul 27, 2010

MALE SHOEGAZE posted:

didnt that happen like 2 years ago

apparently still an issue

# ? Nov 20, 2014 07:25

Luigi Thirty: Apr 30, 2006; Emergency confection port.

Seaside Loafer posted:

That would be a good idea in theory but since nobody had already done it for the thing (wish I could remember what it was) the company couldnt really embark on a project to emulate it could they, that would have been more difficult than the actual project. If I get a brainwave and remember what model it was i'll post it. it was (is for another 5 years at least according to the contract) a 'mini-computer' haha, of some sort http://en.wikipedia.org/wiki/Minicomputer

one of my dad's friends wrote Ada for one of these things for the navy in the 80s and went crazy. makes you think.

Notorious b.s.d. posted:

I don't expect luigi thirty to pull himself up by his bootstraps or fix everything in his life at once or magically stop being sadbrains tomorrow

just a job search for your first job is a bad place to "phone it in." even a depressed person must make choices where to apply greater or lesser effort, and your first job search is a place where more is better.

i did try rewriting my resume again and sending out more applications in the last 2 days than in the preceding month so let's see how that goes. surely if i can teach myself enough 6502 assembly to write YOSPOS.BIN i can obtain enough c# skills (which i have used for about a year) and python skills (which i have used for several years) to impress some interviewer.

fart simpson posted:

just keep in mind that study that showed interviewers mostly make up their minds in the first 10 seconds of the interview, before you even speak.

I am white, therefore

hobbesmaster posted:

the tough love "power through" speeches will work if you don't have clinical depression and are just feeling bad. if you do have clinical depression it just comes off as "oh, I'm completely worthless it is completely impossible for me to do that" and things get worse

i feel that way and i know it is irrational but i can't stop feeling that way! that is the worst part. sometimes i have a breakthrough and end up hyper-focused working 16 hours a day on some dumb project that's as likely to be useful (a win8 app or django site i can show off) as not useful (a commodore 64 tf2 crate unboxing program or YOSPOS.BIN).

# ? Nov 20, 2014 10:27

qntm: Jun 17, 2009

VLADIMIR GLUTEN posted:

Hahaha what the hell is fonzsort? Does it involve jukebox punching?

fonzsort is when you go up to sort the array, but it's already sorted because you're just that awesome, eyyy

# ? Nov 20, 2014 10:42

gonadic io: Feb 16, 2011; >>=

Valeyard posted:

naming them yes. the problem was that I have a map, with strings as keys, and datatype X as values. but when I print out the map, it prints out datatype X as a monstrosity.

for no, https://zuttobenkyou.wordpress.com/2011/02/27/simple-haskell-data-pretty-printing/ , this was quick enough to setup and is ok to use for tests so far

it's possible to override show, but it's much better to just write your own printing function and call it something else, which is pretty much what ppShow does. sadly there's not a full debugger for haskell, but the most common tool to use is the function trace from the library Debug.Trace which bypasses the type system to print values from inside functions. it's a little finicky to use due to laziness, by far the easiest way to use is to put trace at the top level of a definition like 'function = trace string retVal' so the string is printed whenever the function is called

# ? Nov 20, 2014 12:26

Malcolm XML: Aug 8, 2009; I always knew it would end like ｔｈｉｓ．

Seaside Loafer posted:

That would be a good idea in theory but since nobody had already done it for the thing (wish I could remember what it was) the company couldnt really embark on a project to emulate it could they, that would have been more difficult than the actual project. If I get a brainwave and remember what model it was i'll post it. it was (is for another 5 years at least according to the contract) a 'mini-computer' haha, of some sort http://en.wikipedia.org/wiki/Minicomputer

IBM holds a ton patentns on stupid poo poo and will sue your rear end into oblivion if you attempt to virtualize it w/o paying them

# ? Nov 20, 2014 12:49

fart simpson: Jul 2, 2005; DEATH TO AMERICA

AlsoD posted:

it's possible to override show, but it's much better to just write your own printing function and call it something else, which is pretty much what ppShow does. sadly there's not a full debugger for haskell, but the most common tool to use is the function trace from the library Debug.Trace which bypasses the type system to print values from inside functions. it's a little finicky to use due to laziness, by far the easiest way to use is to put trace at the top level of a definition like 'function = trace string retVal' so the string is printed whenever the function is called

yeah just write your own function HugeDataType -> String and then another that wraps that Map -> String something like this:

code:

data Crap = Crap Int Float String [Float]
data MoreCrap = MoreCrap String Crap

showCrap :: Crap -> String
showCrap (Crap _ theThingYouWant _ _) = show theThingYouWant

showMoreCrap : MoreCrap -> String
showMoreCrap (MoreCrap s crap) = "MoreCrap " ++ s ++ " " ++ showCrap crap

idk that's the general way i'd do it

# ? Nov 20, 2014 12:59

gonadic io: Feb 16, 2011; >>=

also trace is one of the functions which tell the type system to gently caress off. other fun ones include unsafePerformIO :: IO a -> a which masks a method's side effects from the type system (trace is defined in terms of this or something similar) and unsafeCoerce :: a -> b. yes that's any a and any b. i think it's depreciated in favour of a more nuanced solution involving the compiler working out which things are safe to coerce into each other as it's main usecase was swapping between collections of newtypes (which have the same internal representation as each other, this was NOT checked by the compiler) without having to traverse the collection itself

# ? Nov 20, 2014 13:31

Moist von Lipwig: Oct 28, 2006; by FactsAreUseless; Tortured By Flan

qntm posted:

fonzsort is when you go up to sort the array, but it's already sorted because you're just that awesome, eyyy

eyyy lmao

# ? Nov 20, 2014 13:47

Notorious b.s.d.: Jan 25, 2003; by Reene

Malcolm XML posted:

IBM holds a ton patentns on stupid poo poo and will sue your rear end into oblivion if you attempt to virtualize it w/o paying them

ibm missed the boat on minicomputers. nobody bought s/36 or s/38. like, when the word "minicomputer" was a thing people said, ibm was not in the market.

ironically they then developed the s/38 architecture into the wildly popular as/400 in the 90s. people think of as/4040 as an archaic 1970s thing when in reality it had all its growth from like 1992 to 2000

# ? Nov 20, 2014 14:00

Captain Foo: May 11, 2004; we vibin'
we slidin'
we breathin'
we dyin'

Space Whale posted:

Through CTCI, brain teasers, code rust, being good at selling myself, and being acceptable at code monkey things, I managed to get myself a mid level job. Woo hoo!

As such I'm now exposed to stuff like the actual ISS server and the messaging queue and other things that are beyond just "poke around project, mash F12, and do what BA says, and make sure stylecop is happy." I actually feel that "oh poo poo I'm a faker" feeling again, which means I'm learning.

What do you DO to really get all this stuff besides ask questions, stay late and buddy up to the IT guys?

read up on it, i guess? No better way to learn a specific system than to get balls deep, tho

You're doing great :unsmith:

e- also remember that almost nobody is as good as X as you think they are, they are just a little bit better/have a bit more experience so don't be too intimidated

# ? Nov 20, 2014 14:50

Space Whale: Nov 6, 2014

Any good IIS sources besides msdn?

# ? Nov 20, 2014 15:42

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

Thinking pretty hard about committing some truly bad code.

Either that or spend 2 weeks working on a system i don't want to work on

# ? Nov 21, 2014 01:39

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

whenever I want to commit a bad code I change my git config to use my boss's name and email, works pretty well

# ? Nov 21, 2014 01:44

fart simpson: Jul 2, 2005; DEATH TO AMERICA

Soricidus posted:

whenever I want to commit a bad code I change my git config to use my boss's name and email, works pretty well

cool idea

# ? Nov 21, 2014 03:25

DimpledChad: May 14, 2002; Rigging elections since '87.

what thread are we in??? whenever i commit bad code i just check it into the repo. when i commit good code, i pop a bottle of bubbly and reward myself with a well-earned mani pedi

# ? Nov 21, 2014 03:44

bobbilljim: May 29, 2013; this christmas feels like the very first christmas to me

has anyone ever committed good code?

# ? Nov 21, 2014 04:00

Moist von Lipwig: Oct 28, 2006; by FactsAreUseless; Tortured By Flan

bobbilljim posted:

has anyone ever committed good code?

https://github.com/mapmeld/profanity65

# ? Nov 21, 2014 04:10

DimpledChad: May 14, 2002; Rigging elections since '87.

i commit code in the same way that one might commit murder

# ? Nov 21, 2014 04:10

Luigi Thirty: Apr 30, 2006; Emergency confection port.

bobbilljim posted:

has anyone ever committed good code?

Bitcoin

# ? Nov 21, 2014 04:11

Bloody: Mar 3, 2013

i took some objects and denormalized them into a database

getting them into and out of the database is such a mess

i have to hit like four different tables to get all the crap out

is this an impotence mismatch

should i be using mongoloid database

# ? Nov 21, 2014 05:47

Bloody: Mar 3, 2013

i literally thought lol gently caress it ill just serialize these objects to blobs and stuff them in the database but then i thought

# ? Nov 21, 2014 05:49

bobbilljim: May 29, 2013; this christmas feels like the very first christmas to me

Bloody posted:

i took some objects and denormalized them into a database

getting them into and out of the database is such a mess

i have to hit like four different tables to get all the crap out

is this an impotence mismatch

should i be using mongoloid database

this is a normal way to use a relational db. a lot of times people denormalize to some extent though. depends on the situation but there's not really an easy way with dbs

# ? Nov 21, 2014 05:54

Bloody: Mar 3, 2013

well its a pain in the rear end and it makes my code look like poo poo. if this was an orm it would be an ienumerable and i could just waltz into the sunset.

# ? Nov 21, 2014 06:11

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

Why do you care what the code that actually accesses the db looks like? I mean yeah it's a shitpile of querying multiple tables and munging the results into some objects, but it's not like that interferes with your real code so who cares?

OTOH if that stuff is leaking into your business logic somewhere you done hosed up.

# ? Nov 21, 2014 11:43

jesus WEP: Oct 17, 2004

just use stored procs and then ur messy sql queries will be on the db where they belong

# ? Nov 21, 2014 12:34

power botton: Nov 2, 2011

St Evan Echoes posted:

just use stored procs and then ur messy sql queries will be on the db where they belong

# ? Nov 21, 2014 13:02

bobbilljim: May 29, 2013; this christmas feels like the very first christmas to me

St Evan Echoes posted:

just use stored procs and then ur messy sql queries will be on the db where they belong

# ? Nov 21, 2014 13:16

Lord Of Texas: Dec 26, 2006

Jabor posted:

Why do you care what the code that actually accesses the db looks like? I mean yeah it's a shitpile of querying multiple tables and munging the results into some objects, but it's not like that interferes with your real code so who cares?

OTOH if that stuff is leaking into your business logic somewhere you done hosed up.

Database objects change, unmanageable SQL is technical debt just like any other code

St Evan Echoes posted:

just use stored procs and then ur messy sql queries will be on the db where they belong

Yuck, please don't do this (even if you are using database source control like Liquibase it's decades behind even the worst typical VC)

If you are in some utopia where you never have to maintain code ever than yeah go ahead

# ? Nov 21, 2014 14:06

Shaggar: Apr 26, 2006

stored procs are the way to go.

# ? Nov 21, 2014 15:16

prefect: Sep 11, 2001; No one, Woodhouse.
No one.; Dead Man’s Band

Shaggar posted:

stored procs are the way to go.

just don't call them "sprocs", because it irritates me :mad:

# ? Nov 21, 2014 15:17

jesus WEP: Oct 17, 2004

shaggar was right
i got that stored proc wisdom from him itt and its served me well ever since

# ? Nov 21, 2014 15:21

Janitor Prime: Jan 22, 2004; PC LOAD LETTER

What da fuck does that mean; Fun Shoe

Lord Of Texas posted:

Yuck, please don't do this (even if you are using database source control like Liquibase it's decades behind even the worst typical VC)

If you are in some utopia where you never have to maintain code ever than yeah go ahead

Maintaining a small codebase without liquibase is just fine if you don't make frequent changes to your schemas and for the rest liquibase is more than adequate.

Database code belongs in the database and should be properly version controlled just like any other code hth

# ? Nov 21, 2014 16:52

Shaggar: Apr 26, 2006

Microsoft should add automatic schema versioning to sql server

# ? Nov 21, 2014 17:04

Lord Of Texas: Dec 26, 2006

Janitor Prime posted:

Maintaining a small codebase without liquibase is just fine if you don't make frequent changes to your schemas and for the rest liquibase is more than adequate.

Database code belongs in the database and should be properly version controlled just like any other code hth

I guess part of it depends on your setup - if you as a developer have full control over your database AND it rarely changes AND your codebase is small, then just do whatever. Where I'm at, due to decisions out of my control my app is sharing a database which changes frequently so maintaining hundreds of stored procs that change frequently via Liquibase as opposed to in SVN would be a loving nightmare.

(also Liquibase does not handle stored procs particularly well anyway, I've heard Datical does a better job but we're still on the open source LB)

quote:

Database code belongs in the database and should be properly version controlled just like any other code hth

Unless you're in a land where you hand a database to end users, the line drawn between database code and application code is pretty much irrelevant, put it where it's the most maintainable (in traditional VC.) It's like shoving all of your validation into your presentation because "that's where it belongs!!!", then you end up with an untestable (by anything but manual tests or Selenium) unmaintainable mess.