|
Shaggar posted:descriptive names are good. theres no reason to be using acronyms or abbreviations. I agree, HyperTextMarkupLanguageDocumentObjectTreeElement is a good name.
|
# ? Feb 15, 2017 08:49 |
|
|
# ? May 23, 2024 17:00 |
|
Sapozhnik posted:that was me for some reason enterprise java/c# programmers loving hate plain old static functions
|
# ? Feb 15, 2017 11:11 |
|
reflection based dependency injection is cool as heck when it works debugging that poo poo is the worst nightmare i have ever had to deal with though
|
# ? Feb 15, 2017 16:15 |
|
Xarn posted:I agree, HyperTextMarkupLanguageDocumentObjectTreeElement is a good name. If somebody else gives you an abbreviation, go ahead and use it. If you are creating your own abbreviations - ever - stop it.
|
# ? Feb 15, 2017 18:07 |
|
At past jobs I refuted away any attempts to use reflection in our code as too much of a performance hit. So far, it's saved me from having to work too hard to understand it. Current job? lol no one gives a poo poo about DI when your platform is the code equivalent of the London Fire.
|
# ? Feb 15, 2017 18:08 |
|
HoboMan posted:reflection based dependency injection is cool as heck when it works guice goes into extremely specific detail about why it cannot instantiate a thing, op
|
# ? Feb 15, 2017 18:25 |
|
Finster Dexter posted:At past jobs I refuted away any attempts to use reflection in our code as too much of a performance hit. So far, it's saved me from having to work too hard to understand it. We're in the final stages of implementing a dynamic odata api that builds the entities on demand based on each individual customer's bespoke data structures so lol the only parts of the codebase that aren't reflection are in IL lol kill me.
|
# ? Feb 15, 2017 18:50 |
|
Sapozhnik posted:guice goes into extremely specific detail about why it cannot instantiate a thing, op dagger is also cool and good for reducing runtime overhead
|
# ? Feb 15, 2017 18:56 |
|
Chalks posted:We're in the final stages of implementing a dynamic odata api that builds the entities on demand based on each individual customer's bespoke data structures so lol the only parts of the codebase that aren't reflection are in IL lol kill me. haha omg ok that sounds pretty bad
|
# ? Feb 15, 2017 18:58 |
|
Sapozhnik posted:guice goes into extremely specific detail about why it cannot instantiate a thing, op otoh ot's called "guice" and you know the least funny borat-quoter in your office will non-stop say "the guice is lose" whenever they see it in a stack trace
|
# ? Feb 15, 2017 19:12 |
|
Sapozhnik posted:guice goes into extremely specific detail about why it cannot instantiate a thing, op that stuff is pretty straightforward i am talking about poo poo like: "hey, can you tell me why this one specific value is wrong? or where is is coming from and what modifies it because we have a customer that insists it's wrong and we either need to prove them wrong or find the faulty business logic. (by the way, this is all undocumented)" this was very hard to answer without exactly recreating what the user was doing because everything happens at runtime
|
# ? Feb 15, 2017 19:25 |
|
Cocoa Crispies posted:least funny borat-quoter
|
# ? Feb 15, 2017 20:19 |
|
oh god all my references in my project got hosed somehow like it can't even find the namespace "System" help?
|
# ? Feb 15, 2017 22:52 |
|
HoboMan posted:oh god all my references in my project got hosed somehow we switched his c# compiler with java. let's see if he notices.
|
# ? Feb 15, 2017 23:17 |
|
carry on then posted:we switched his c# compiler with java. let's see if he notices. it was actually using the wrong compiler lol
|
# ? Feb 15, 2017 23:20 |
|
HoboMan posted:it was actually using the wrong compiler lol How did it end up using the wrong compiler?? Is this a practical joke I can play on someone?
|
# ? Feb 15, 2017 23:23 |
|
somehow the csproj file had an assert for checking if some c++ compiler existed in the packages folder
|
# ? Feb 15, 2017 23:37 |
|
https://youtu.be/56RsdDNjGI4
|
# ? Feb 16, 2017 00:09 |
|
comedyblissoption posted:reflection based dependency injection frameworks are poop from a butt are you talking about static functions as factories, or static functions as LOB logic? because the reason the latter are out of style is because you're locked to one implementation which means you can't supply a new one or unit test it without breaking the VM impure static functions are bad
|
# ? Feb 16, 2017 00:14 |
|
reposting this from the pl thread because it fits better here: don't use node if you do use node, don't use mongo if you do you node and mongo, definitely do not ever use mongoose mongoose is a javascript orm for mongo it is exactly as dumb as you would expect if you know anything at all about mongo and javascript (ask me about about chasing performance bugs for a week and a half only to find mongoose is spending approximately 1000x as long constructing objects as it is querying mongo and parsing the json combined) anywho, we use it a) to enforce a schema on our schemaless database, except it only enforces it from the node-based processes that talk to mongo, not the java-based ones, b) to make crud slightly simpler, c) to badly reinvent LISTEN/NOTIFY using a dumbass plugin a coworker wrote, and d) because we're masochists i guess we inherited this project from another group that is currently in the process of perpetrating a new node/mongo atrocity i very very badly want to rewrite this piece of garbage in anything but javascript but unforunately, deadlines
|
# ? Feb 16, 2017 00:17 |
|
Luigi Thirty posted:the C macro system is incredibly garbage did anyone say port g++ to saturn?
|
# ? Feb 16, 2017 00:22 |
|
So I figured out how to write a reverse polish notation parser in haskell (before finding out that that's also in LYAH so I was kinda proud of that) Then I wanted to see if I could do it that way in C#, and I can: C# code:
|
# ? Feb 16, 2017 00:22 |
|
or you know, just make the cross compiler bro
|
# ? Feb 16, 2017 00:23 |
|
Mongo is a bad name because it makes me want to call users of Mongo problematic names
|
# ? Feb 16, 2017 00:27 |
|
imagine caring about you are data so little you would use a non relational db
|
# ? Feb 16, 2017 00:29 |
|
AggressivelyStupid posted:Mongo is a bad name because it makes me want to call users of Mongo problematic names mongo users deserve them
|
# ? Feb 16, 2017 00:31 |
|
Cocoa Crispies posted:mongo users deserve them
|
# ? Feb 16, 2017 00:37 |
|
AggressivelyStupid posted:Mongo is a bad name because it makes me want to call users of Mongo problematic names get in the gas chamber
|
# ? Feb 16, 2017 00:45 |
|
oh no blimp issue posted:did anyone say port g++ to saturn? the full GCC compiler suite supports SuperH because people still use it for some reason the library I'm using is C89 so in theory I should be able to use g++ with it I guess? the problem is the system only has 1MB of work RAM and g++ produces huge object files Luigi Thirty fucked around with this message at 01:13 on Feb 16, 2017 |
# ? Feb 16, 2017 01:09 |
|
St Evan Echoes posted:imagine caring about you are data so little you would use a non relational db what if I told you that an acid-compliant nosql db was out there.... waiting for you
|
# ? Feb 16, 2017 01:16 |
|
oh there we go, the application compiles with sh2-g++ 3.4 and -fpermissive turned on because sega loved malloc void*
|
# ? Feb 16, 2017 01:38 |
|
I dunno, I don't think node is that bad. Like it's definitely bad, but it's not really worse than any other p-lang. npm is a reasonably good package manager too.
|
# ? Feb 16, 2017 03:16 |
|
nrook posted:I dunno, I don't think node is that bad. Like it's definitely bad, but it's not really worse than any other p-lang. npm is a reasonably good package manager too. left pad
|
# ? Feb 16, 2017 03:46 |
|
nrook posted:I dunno, I don't think node is that bad. Like it's definitely bad, but it's not really worse than any other p-lang. npm is a reasonably good package manager too.
|
# ? Feb 16, 2017 03:50 |
|
i wish all the languages that de facto require a package manager would do rpm/apt integration instead of write their own poo poo mountain. the whole cpan/pip/npm cottage industry is a waste of time
|
# ? Feb 16, 2017 04:02 |
|
nrook posted:npm is a reasonably good package manager too. thats a good one haha
|
# ? Feb 16, 2017 04:40 |
|
hifi posted:i wish all the languages that de facto require a package manager would do rpm/apt integration instead of write their own poo poo mountain. the whole cpan/pip/npm cottage industry is a waste of time how do I rpm/apt on my macbook or my surface and what does it mean to "de facto require a package manager"? I can copy/paste left-pad np m
|
# ? Feb 16, 2017 04:45 |
|
speaking of this poo poo, today is the first day of us going full web dev. i installed node and npm and used npm to get react and typescript and webpack i already hate this toolchain wish me luck!
|
# ? Feb 16, 2017 04:49 |
|
nrook posted:I dunno, I don't think node is that bad. Like it's definitely bad, but it's not really worse than any other p-lang. npm is a reasonably good package manager too. js itself isn't significantly worse than say, ruby (I'd still rather use python if I had to use a plang tho) the libraries on npm though, oh my god. there's basically zero way to tell what will be a horrific mess under the hood and what will be kind of okay until it explodes on you also similar to ruby, something about js makes people desperately want to abuse the dynamism and make bugnuts incomprehensible bullshit
|
# ? Feb 16, 2017 05:32 |
|
|
# ? May 23, 2024 17:00 |
|
hmm, it compiles fine under GCC 3.4 but doesn't work under GCC 6.x because of some problem with libgcc for sh2 missing arithmetic emulation functions. rip
|
# ? Feb 16, 2017 05:43 |