|
https://blog.joda.org/2018/09/do-not-fall-into-oracles-java-11-trap.html lolquote:Oracle JDK, the one all web searches take you to, is now commercial not $free.
|
# ? Sep 26, 2018 16:09 |
|
|
# ? Apr 25, 2024 07:21 |
|
Hopefully this will be the Perma Death of Java one less language to worry about Next up on the kill list: OpenJDK
|
# ? Sep 26, 2018 16:16 |
|
its against the terms to even run oracle's runtime in any sort of container or namespacing mechanism. it's possible that even using selinux with oracle's jvm is violating that. java is good though. oracle is not.
|
# ? Sep 26, 2018 16:28 |
Are there reverse zip bombs? Like where a reasonably-sized input has a gigantic zipped form. I know that by the pigeonhole principle there are inputs with zipped forms that are larger than the original input, but my guess would be that there is an upper limit on this potential growth.
|
|
# ? Sep 26, 2018 16:30 |
|
CRIP EATIN BREAD posted:java is good though. oracle is not.
|
# ? Sep 26, 2018 16:34 |
|
VikingofRock posted:Are there reverse zip bombs? Like where a reasonably-sized input has a gigantic zipped form. I know that by the pigeonhole principle there are inputs with zipped forms that are larger than the original input, but my guess would be that there is an upper limit on this potential growth. I think the answer is "no, not unless the particular compression algorithm is extremely bad"-- decent pseudorandom inputs will probably all compress about the same which is to say that lossless compression in the worst case has to fall back to a data stream that takes up "basically the input file, plus inline framing/metadata to specify that we basically couldn't compress it" bytes, which, assuming the compression algo isn't poo poo, takes up 1+delta (for some kinda delta closer to 0 than to 1) times the original file size e: in a degenerate case, lz77 has an "expansion ratio" of 1 / 1100% but who fuckin' cares about single byte files quote:for the default settings used by deflateInit(), compress(), and compress2(), the only expansion is an overhead of five bytes per 16 KB block (about 0.03%), plus a one-time overhead of six bytes for the entire stream. Even if the last or only block is smaller than 16 KB, the overhead is still five bytes. In the absolute worst case of a single-byte input stream, the overhead therefore amounts to 1100% (eleven bytes of overhead, one byte of actual data). For larger stream sizes, the overhead approaches the limiting value of 0.03%. prisoner of waffles fucked around with this message at 16:49 on Sep 26, 2018 |
# ? Sep 26, 2018 16:45 |
|
looks like it’s j9’s time to shine!!!
|
# ? Sep 26, 2018 16:58 |
|
It's actually apparently sort of supposed to be because they merged Oracle JDK into OpenJDK and Oracle JDK is now just a version of OpenJDK, so anyone who doesn't want a commercial version is just supposed to be using OpenJDK or something. However, the confusion around this is definitely going to kill Java, which is unfortunate. I guess it's just part of Java's continual effort to shoot itself in the foot so everyone uses crappy languages like go instead.
|
# ? Sep 26, 2018 16:59 |
|
carry on then posted:looks like it’s scala native’s time to shine!!!
|
# ? Sep 26, 2018 17:04 |
|
prisoner of waffles posted:I think the answer is "no, not unless the particular compression algorithm is extremely bad"-- decent pseudorandom inputs will probably all compress about the same how about writing an algorithm that creates a valid stream that the decompressor will understand but which purposely wastes space for instance when creating a huffman tree you could inverse the probabilities so that it assigns the longest codes to the most common symbols
|
# ? Sep 26, 2018 17:17 |
|
Zlodo posted:creating a huffman tree you could inverse the probabilities so that it assigns the longest codes to the most common symbols the cute thing about what you're proposing is that it would do okay on random noise and do badly on complex compressible data
|
# ? Sep 26, 2018 17:26 |
|
CRIP EATIN BREAD posted:its against the terms to even run oracle's runtime in any sort of container or namespacing mechanism. it's possible that even using selinux with oracle's jvm is violating that.
|
# ? Sep 26, 2018 17:37 |
|
mystes posted:However, the confusion around this is definitely going to kill Java, which is unfortunate. I guess it's just part of Java's continual effort to shoot itself in the foot so everyone uses crappy languages like go instead. it's going to kill java for people who post questions on stack overflow asking how to concatenate two strings. the java ecosystem itself will be fine.
|
# ? Sep 26, 2018 17:43 |
|
also the best compression algorithm: http://www.dangermouse.net/esoteric/lenpeg.html
|
# ? Sep 26, 2018 17:43 |
|
CRIP EATIN BREAD posted:also the best compression algorithm: http://www.dangermouse.net/esoteric/lenpeg.html lol
|
# ? Sep 26, 2018 18:19 |
|
http://www.dangermouse.net/esoteric/petrovich.html found the official yospos os.
|
# ? Sep 26, 2018 20:52 |
|
moonshine is...... posted:http://www.dangermouse.net/esoteric/petrovich.html reward
|
# ? Sep 26, 2018 21:05 |
|
CRIP EATIN BREAD posted:it's going to kill java for people who post questions on stack overflow asking how to concatenate two strings. the java ecosystem itself will be fine. yep oracle will have a few new lawsuit targets, some idiots will get made examples of, people will freak out and then all the adults using java who pay people money to understand software licenses will just keep on doing what they're doing today
|
# ? Sep 26, 2018 21:09 |
|
Nah, java will just become another cancer like the AGPL that no companies will touch.
|
# ? Sep 26, 2018 21:11 |
|
extern has been slathered all over functions in this pile of poo poo c program and I'm like 99% sure functions are implicitly extern in header files there are some instances of variables being EXTERN, with EXTERN being conditionally defined to either nullstring or extern, depending on if __foo__ is defined this feels gross, could someone recommend some good reading for all this? my gut tells me we're doing poo poo very wrong but I don't have any concrete reasons why E: I've done some googling but keep getting c++ stuff which while interesting doesn't really apply
|
# ? Sep 26, 2018 21:14 |
|
mystes posted:Nah, java will just become another cancer like the AGPL that no companies will touch. Except for all those companies that have no problem going all-in on lovely oracle db. They'll have middle managers writing java licensing checks faster than you can say "yospos bithc"
|
# ? Sep 26, 2018 21:31 |
|
Big corporations may license java for now, and milking them is probably the goal, but it's going to turn into a thing like React before the license change where everyone is trying to switch away from it as fast as possible regardless of whether that's actually a massive overreaction.
|
# ? Sep 26, 2018 22:13 |
|
mystes posted:Big corporations may license java for now, and milking them is probably the goal, but it's going to turn into a thing like React before the license change where everyone is trying to switch away from it as fast as possible regardless of whether that's actually a massive overreaction. ah yes, my company will jump at the chance to rewrite our 1.9 million lines of java into rust, not a problem. totally something to do on a whim e: also port 2 or 3 giant libraries we depend on into some other language, because there basically aren't meaningful alternatives in any other language e2: plus god knows how many lines of scala in another big project i forgot exists Arcsech fucked around with this message at 23:32 on Sep 26, 2018 |
# ? Sep 26, 2018 23:27 |
|
I wonder if Experts Exchange has upgraded from Java 8 yet. Probably not, since I'm the guy who pushed them to upgrade to Javas 5, 6, 7, and 8 and they laid me off two Januaries ago. But speaking of EE and rewriting stuff to get away from Java, EE is currently built off of a home-built CMS that's just absolutely terrible to work with and mixes concerns so thoroughly it's practially impossible to make any progress on fixing. The employee who "designed" this CMS, eight or nine years ago or so, declared it a functional prototype and proceeded to immediately suggest we rewrite the entire codebase in PHP. I can't even imagine how much of a mess that would have been.
|
# ? Sep 26, 2018 23:41 |
|
AggressivelyStupid posted:extern has been slathered all over functions in this pile of poo poo c program and I'm like 99% sure functions are implicitly extern in header files you’re almost correct, function prototypes are always implicitly extern in C when dealing with libraries, it can be important to ensure C++ name mangling isn’t applied to symbols derived from C declarations, that’s why instead of extern you’ll often see something like MYLIB_EXTERN used, where it’s defined like code:
this can also be a good place to put things like compiler specific directives to mark a symbol as exported (when building a library) or potentially NULL (when consuming a library at runtime) so it’s good practice in general
|
# ? Sep 26, 2018 23:46 |
|
theyre actually doing a lot of:code:
code:
i actually can't get the goddamn thing to build due to linker errors, straight out of source control, which is a cool and good feeling. there's approximately 200 .c and .h files in the folder, and only 90~ of them inside the codewarrior project. i dunno if embedded is normally this level of Hell or what but I kinda hope not
|
# ? Sep 27, 2018 01:46 |
|
mystes posted:Big corporations may license java for now, and milking them is probably the goal, but it's going to turn into a thing like React before the license change where everyone is trying to switch away from it as fast as possible regardless of whether that's actually a massive overreaction. everyone will switch to openjdk except for companies that are also ok with paying for things like oracle db if openjdk goes to poo poo for whatever reason then idk google will fork it or something
|
# ? Sep 27, 2018 01:53 |
|
AggressivelyStupid posted:i dunno if embedded is normally this level of Hell or what but I kinda hope not #ifdef guards to prevent multiple-inclusion (if that's actually what's happening) is better done with #pragma once screwing around with extern is another smell of "this was broken once, now i have this dead chicken to wave over every new declaration" sorry i have the memory of a goldfish, what's your goal here? just re-building something, trying to modify it, trying to do a new product with the old code base?
|
# ? Sep 27, 2018 02:11 |
|
JawnV6 posted:#ifdef guards to prevent multiple-inclusion (if that's actually what's happening) is better done with #pragma once i think this is exactly what is happening JawnV6 posted:sorry i have the memory of a goldfish, what's your goal here? just re-building something, trying to modify it, trying to do a new product with the old code base? fixing some bugs and just getting the drat thing to compile in the first place
|
# ? Sep 27, 2018 02:26 |
|
"#pragma once" used to be a bit problematic on GCC so people avoided it for a while iirc fake edit: from https://gcc.gnu.org/gcc-3.4/changes.html quote:File handling in the preprocessor has been rewritten. GCC no longer gets confused by symlinks and hardlinks, and now has a correct implementation of #import and #pragma once. These two directives have therefore been un-deprecated.
|
# ? Sep 27, 2018 02:31 |
|
woof, that's roughAggressivelyStupid posted:there's approximately 200 .c and .h files in the folder, and only 90~ of them inside the codewarrior project. another approach would be to find main() and add files one at a time to fix whatever build issue it complains about first. on adding each file, take the time to scrub the EXTERN dances away at a bare minimum ugh, i really don't envy your position. there's some happy-path through the files where everything's declared once, but the prior authors don't really care if you can find it again
|
# ? Sep 27, 2018 02:37 |
|
I was able to clear up the undefined linker errors by adding a few files to the project ended up with a multi defined error instead just before I left for the day!! Looking forward to untangling that tomorrow The annoying thing is that it's basically untouched since 2014 and these people don't have source control discipline so the compiled binaries that are in the PLM (god drat it) may well have been the result of changes that exist only on some dickhead who left last years computer I think this somehow makes me, the intern, the leading expert on this product lmfao AggressivelyStupid fucked around with this message at 02:59 on Sep 27, 2018 |
# ? Sep 27, 2018 02:53 |
|
sounds like your internship is teaching you a lot
|
# ? Sep 27, 2018 05:38 |
|
lol all the scala idiots in my org are gonna get so much louder now
|
# ? Sep 27, 2018 06:42 |
|
maybe everyone will switch to Swift after all, IBM’s doing lots of stuff with Swift on Linux
|
# ? Sep 27, 2018 08:40 |
|
eschaton posted:maybe everyone will switch to Swift 3 people have asked me to implement Hadoop in swift now. people must hate me
|
# ? Sep 27, 2018 08:41 |
|
call me when swift runs on the Motorola 68000
|
# ? Sep 27, 2018 09:14 |
|
Luigi Thirty posted:call me when swift runs on the Motorola 68000 be the change you want to see in the world: write the compiler
|
# ? Sep 27, 2018 09:25 |
|
Luigi Thirty posted:call me when swift runs on the Motorola 68000 well it could be an interesting project I guess https://github.com/M680x0/M680x0-llvm https://github.com/apple/swift
|
# ? Sep 27, 2018 10:03 |
|
|
# ? Apr 25, 2024 07:21 |
|
jit bull transpile posted:lol all the scala idiots in my org are gonna get so much louder now do they not understand that this affects scala in exactly the same way as it affects java
|
# ? Sep 27, 2018 10:34 |