|
Bloody posted:oh also it was nice to discover that the clients gracefully handle the server dropping off the network for 2+ days by just sittin there and waiting til it comes back
|
# ? Oct 10, 2014 04:52 |
|
|
# ? May 25, 2024 00:46 |
|
ugh, there's no API for selling items on the store via their OpenID stuff. to automate selling of items in my app the user has to log into steam community using the webview embedded in the inventory page. i'm not publishing something that looks like a phishing app.
|
# ? Oct 10, 2014 09:18 |
|
Luigi Thirty posted:ugh, there's no API for selling items on the store via their OpenID stuff. to automate selling of items in my app the user has to log into steam community using the webview embedded in the inventory page. i'm not publishing something that looks like a phishing app.
|
# ? Oct 10, 2014 12:12 |
|
|
# ? Oct 10, 2014 15:02 |
|
Im flesh and blood
|
# ? Oct 10, 2014 15:27 |
|
Shaggar posted:chef and puppet or w/e are script based build systems. you write scripts to make the build happen. your build is more code which means you need to test your build code. puppet is exactly as declarative as maven and works very similarly inside. you write a definition of the end state of the system, dependencies are calculated as a directed acyclic graph, and then it builds a list of changes to make. sound familiar? and yes, you still should write tests for your puppet and maven work. there is still mutable state and there are opportunities for everything to look OK and not actually be OK (you probably don't need a test for a java build job: did it poo poo out a war successfully? TEST PASSED. but if you are gonna configure your httpd and deploy to servers with maven, you need to test that poo poo)
|
# ? Oct 10, 2014 15:39 |
|
chef is much closer to straight up scripting. it definitely 100% has the "we put code in your code so you can code while you code" problem. the upside is that it is deterministic: you know the exact order resources will converge and things will execute to continue the analogy, if puppet is maven-like, chef is much more like gradle: a library/dsl to help you write a certain class of scripts quickly and with minimum custom error handling
|
# ? Oct 10, 2014 15:41 |
|
200 get
|
# ? Oct 10, 2014 15:44 |
|
Notorious b.s.d. posted:chef is much closer to straight up scripting. it definitely 100% has the "we put code in your code so you can code while you code" problem. the upside is that it is deterministic: you know the exact order resources will converge and things will execute yeah gradle is dumb and primarily for java-but-don't-call-it-java users who can never allow themselves to use good tools. puppet does seem more maven like tho
|
# ? Oct 10, 2014 15:49 |
|
my new dog posted:200 get out of yospos
|
# ? Oct 10, 2014 16:28 |
|
oh good, Tomcat also has a version of javax.annotation.Generated and it has a "comment" field instead of the "comments" field that Java's version has. great.
|
# ? Oct 10, 2014 17:35 |
|
Captain Foo posted:out of yospos 402 payment required
|
# ? Oct 10, 2014 18:48 |
|
Notorious b.s.d. posted:it definitely 100% has the "we put code in your code so you can code while you code" problem. you mean how parts of the product are literally named after memes?
|
# ? Oct 10, 2014 18:52 |
|
my software packaging approach is write it all myself libraries are for n00bs
|
# ? Oct 10, 2014 19:04 |
|
gcc main.c bam done
|
# ? Oct 10, 2014 19:04 |
|
lol if u have more than 1 file per project just lol
|
# ? Oct 10, 2014 19:07 |
|
i find it best to not waste time pretending like im going to reuse any of the liquid hot molten garbage i write so why bother writing in a composable reusable way more like a compostable way
|
# ? Oct 10, 2014 19:42 |
|
let the computer worms eat and poop it all so they become free bits everyone can use again
|
# ? Oct 10, 2014 21:08 |
|
MeruFM posted:let the computer worms eat and poop it all so they become free bits everyone can use again rms thought this that's why he ate the foot fungus
|
# ? Oct 10, 2014 21:09 |
|
maven is piss easy and owns owns owns
|
# ? Oct 10, 2014 23:02 |
|
anybody have a cheat sheet for the rote-memorization-of-CS-undergrad-material portion* of a development interview? i have botched a couple interviews because I never took CS courses that made me rewrite established algorithms/data-structures, so i have taken them for granted my whole life note: i'm not trying to restart interviewchat. i understand that it is a good way to select people who paid attention during their CS degree, but I feel like I am getting false-negatived for positions I would be great at because I havent memorized a thing that will do me absolutely no good to have memorized when i'm on the clock *example: the last bad interview i had asked me to write my own implementation of a* to calculate the shortest path from a to b on a grid with complicated distance costs
|
# ? Oct 11, 2014 17:48 |
|
If I were presented with that question during an interview, I would have to write that I never learned it, but figure I could look up some pseudocode on Wikipedia, then hope that I'd get a point for honesty. (We learned Dijkstra's algorithm in one of our intro classes, but I bet I don't remember how to do that, either.)
|
# ? Oct 11, 2014 18:21 |
|
Corla Plankun posted:anybody have a cheat sheet for the rote-memorization-of-CS-undergrad-material portion* of a development interview? cracking the coding interview is the normal rec i think
|
# ? Oct 11, 2014 19:35 |
|
Corla Plankun posted:*example: the last bad interview i had asked me to write my own implementation of a* to calculate the shortest path from a to b on a grid with complicated distance costs rote memorization is dumb. i would tell the interviewer i used a* for an 8 tile puzzle and try and have a discussion about a good admissable heuristic. try because i assume he'd not do very well at it.
|
# ? Oct 11, 2014 20:11 |
|
i am thinking of starting on a RICH INTERNET APPLICATION using vaadin. is this a terrible idea? if so, why? it is for an existing personal project i started w/ a boring scripting language but holy gently caress it exploded so fast as i kept doing more and more lovely js work
|
# ? Oct 11, 2014 20:37 |
|
Notorious b.s.d. posted:i am thinking of starting on a RICH INTERNET APPLICATION using vaadin. is this a terrible idea? if so, why? I looked into vaadin a while back and it looked ok if you want to write a web app in java and don't mind hideously slow iterations because lol gwt build times ended up not using it because it didn't have some feature i needed unless I wrote the gwt binding myself and I figured straight js was less painful at that point
|
# ? Oct 11, 2014 21:02 |
|
i'm trying to work out how to use VS (2012 Ultimate) if it matters. it's just a single file, and i turned precompiled headers off. the file has "#include "OsiClpSolverInterface.hpp"" which initially caused a whole bunch of red underlines in the file. then i, in the project properties, added the library and include directories to the "Configuration Properties > Linker > General > Additional Library Directories" and the "Configuration Properties > C/C++ > Additional Include Directories" respectively. the red underlines went away. however, when trying to build (debug or release mode) i get lots of errors like: quote:Error 1 error LNK2019: unresolved external symbol "public: virtual __thiscall OsiSolverInterface::~OsiSolverInterface(void)" (??1OsiSolverInterface@@UAE@XZ) referenced in function "public: void __thiscall OsiClpSolverInterface::`vbase destructor'(void)" (??_DOsiClpSolverInterface@@QAEXXZ) C:\code\phd\osi\ConsoleApplication1\ConsoleApplication1\MinimumOSI.obj MinimumOsi and so I added the /VERBOSE option to the build so that it'd list all the libraries that it's searching for. however, none of the libraries I've added (as mentioned above) show up in this at all if the red underlines went away, does that mean that VS recognises the library files, so why won't the linker even though I told it to? e: fwiw, this is the file I'm trying to compile: http://www.coin-or.org/Cbc/ch02s02.html gonadic io fucked around with this message at 15:56 on Oct 13, 2014 |
# ? Oct 13, 2014 15:49 |
|
Coursera's Design and Analysis of Algorithms class starts today. I'm not sure how good it is but maybe if you are a bad algorithmist it will make you a slightly less bad algorithmist?
|
# ? Oct 13, 2014 15:55 |
|
i don't like coursera bc you have to keep on their schedule and when i'm just doing something in my free time i always fall behind. udacity ftw
|
# ? Oct 13, 2014 15:58 |
|
Werthog 95 posted:i don't like coursera bc you have to keep on their schedule and when i'm just doing something in my free time i always fall behind. udacity ftw this but i like to bang out a course in a weekend
|
# ? Oct 13, 2014 16:36 |
|
AlsoD posted:i'm trying to work out how to use VS (2012 Ultimate) if it matters. so it looks like it wsa able to find OsiClpSolverInterface.hpp so it can handle the semantics checking. its not finding the definition of the OsiSolverInterface's destructor, which would be in a compiled lib.
|
# ? Oct 13, 2014 16:54 |
|
FamDav posted:so it looks like it wsa able to find OsiClpSolverInterface.hpp so it can handle the semantics checking. its not finding the definition of the OsiSolverInterface's destructor, which would be in a compiled lib. i see. there's a bunch of .la, .lib, and .pc files non of which are named "OsiSolverInterface". i suppose now i should get some kind of batch file to call DUMPBIN /symbols on them and filter it for the ones that VS can't find?
|
# ? Oct 13, 2014 17:11 |
|
AlsoD posted:i see. there's a bunch of .la, .lib, and .pc files non of which are named "OsiSolverInterface". i suppose now i should get some kind of batch file to call DUMPBIN /symbols on them and filter it for the ones that VS can't find? this is the point where i would say take it to CoC and ask in the c++ thread. there are a few windows users who will be able to tell you how to fix the problem, not just someone like me who can explain it. EDIT: oh cool you already did. yeah crossposting is helpful because you never know who reads here and not there, but that thread should be your goto if youre on windows.
|
# ? Oct 13, 2014 17:17 |
|
FamDav posted:this is the point where i would say take it to CoC and ask in the c++ thread. there are a few windows users who will be able to tell you how to fix the problem, not just someone like me who can explain it. am doing this, thanks
|
# ? Oct 13, 2014 17:18 |
|
i always mean to help out more in this thread but sadly haskell is the only language i know well enough to help, and even then it's very rarely stuff like "why am I getting this type error?". maybe I should hang around on the haskell IRC more, but there's always at least 10 people there ready to jump up for this kind of stuff.
|
# ? Oct 13, 2014 17:19 |
|
I just leaned that VS lets you visually design your data model and it will create the database and all the model source files for you, which is really cool and good.
|
# ? Oct 13, 2014 17:37 |
|
AlsoD posted:i always mean to help out more in this thread but sadly haskell is the only language i know well enough to help, and even then it's very rarely stuff like "why am I getting this type error?". maybe I should hang around on the haskell IRC more, but there's always at least 10 people there ready to jump up for this kind of stuff. the frustrating part about c++ and c is that things can get needlessly tricky for no good reason. if you want to help me out with a haskell question, i was thinking about writing a haskell implementation of our service interface for work. what i'd like is for the service to be a monad transformer so the actual service calls have to run in IO, but you could then mock services if you wanted to test code. So there would be something like a "NameService" monad which also implemented getName, and then you would have the NameServiceIO and NameServiceMock implementations of it. does this all sound right so far? am i going off the rails?
|
# ? Oct 13, 2014 17:38 |
|
triple sulk posted:I just leaned that VS lets you visually design your data model and it will create the database and all the model source files for you, which is really cool and good. nononononononononono
|
# ? Oct 13, 2014 17:49 |
|
don't do it sulk the results will be worse than the DNA that makes up your dog
|
# ? Oct 13, 2014 17:50 |
|
|
# ? May 25, 2024 00:46 |
|
stored procedures are pretty cool and I'm glad I read all those shaggarposts
|
# ? Oct 13, 2014 17:51 |