|
Athas posted:Is this source code really something a competitor would want to steal? The actual value of the code isn't the problem, it's that he's now actually unemployable as a developer if anyone ever identifies him based on his SA account.
|
# ? Nov 24, 2014 22:47 |
|
|
# ? Jun 8, 2024 01:54 |
|
EAT THE EGGS RICOLA posted:The actual value of the code isn't the problem, it's that he's now actually unemployable as a developer if anyone ever identifies him based on his SA account. I'm just waiting for someone to exploit it, and some rigpig in Alberta being really confused as to why one of the rigs is now called "The Yosfracker."
|
# ? Nov 24, 2014 23:41 |
|
EAT THE EGGS RICOLA posted:The actual value of the code isn't the problem, it's that he's now actually unemployable as a developer if anyone ever identifies him based on his SA account. or his dropbox account, which he used
|
# ? Nov 25, 2014 01:04 |
|
How does the anti-switch camp implement FSM's? Do you just resign yourself to a Mealy taste?
|
# ? Nov 25, 2014 01:21 |
|
TopShelfer posted:I have struck gold... Kids don't do what that man did.
|
# ? Nov 25, 2014 01:58 |
|
Can someone loving edit his post and put him out of his misery?
|
# ? Nov 25, 2014 02:00 |
|
On a related note: my consultant rates for installing and administering DLP software across an enterprise are very reasonable. Maybe throw my resume at your boss? Ninjaedit: or host it on dropbox and send them the link.
|
# ? Nov 25, 2014 06:31 |
|
I am not a book posted:On a related note: my consultant rates for installing and administering DLP software across an enterprise are very reasonable. Maybe throw my resume at your boss? Too late, he already checked your resume into source control along with some cat videos.
|
# ? Nov 25, 2014 06:34 |
|
JawnV6 posted:How does the anti-switch camp implement FSM's? Do you just resign yourself to a Mealy taste? With tail recursion
|
# ? Nov 25, 2014 07:06 |
|
JawnV6 posted:How does the anti-switch camp implement FSM's? Do you just resign yourself to a Mealy taste? Tail recursion, coroutines or lookup tables of function pointers. Or in languages that are not terrible: pattern matching on sum types.
|
# ? Nov 25, 2014 08:25 |
|
Athas posted:Or in languages that are not terrible: pattern matching on sum types. I don't even know what this is.
|
# ? Nov 25, 2014 11:29 |
|
Skuto posted:I don't even know what this is. It's like switch statements, but better.
|
# ? Nov 25, 2014 11:52 |
|
Well i have seen lots of code in my life, i even have copies of some of the bad ones but i never give out that code, - some of it is currently over 10 years old. but the horror is knowing that there is code out there like a java class that is currently running at 15000 lines with 250 lines of imports and a cyclic complexity in the hundreds.
|
# ? Nov 25, 2014 12:19 |
|
GrumpyDoctor posted:With tail recursion Why not both
|
# ? Nov 25, 2014 13:21 |
|
Athas posted:Is this source code really something a competitor would want to steal? It's not about competitive advantage, more about things that were not written to be published. Compare to what he might have posted to "upcoming partnerships horrors ITT" or "PYF expense reports". When companies don't know that something's going to be published, there tend to be lots of little things in there that you don't want leaking.
|
# ? Nov 25, 2014 13:31 |
|
Voted Worst Mom posted:Why not both Is there language with both goto-able switch cases and tail call elimination? I am unaware of any.
|
# ? Nov 25, 2014 17:34 |
|
GrumpyDoctor posted:Is there language with both goto-able switch cases and tail call elimination? I am unaware of any. C++
|
# ? Nov 25, 2014 17:38 |
|
GrumpyDoctor posted:Is there language with both goto-able switch cases and tail call elimination? I am unaware of any. Verilog.
|
# ? Nov 25, 2014 17:46 |
|
PleasingFungus posted:or his dropbox account, which he used I think his posts in TCC might beat out the leaked source code as a reason not to employ him.
|
# ? Nov 25, 2014 18:08 |
|
I guess I'm nervous about trusting not-required-by-standard C++ behavior that just happens to be implemented by all major compilers (which I didn't actually know, so that's cool).
|
# ? Nov 25, 2014 18:11 |
|
Soricidus posted:It's like switch statements, but better. In what sense? I've never heard of these before, so I looked up some code examples. Can you explain the advantages for us?
|
# ? Nov 25, 2014 18:38 |
|
GrumpyDoctor posted:I guess I'm nervous about trusting not-required-by-standard C++ behavior that just happens to be implemented by all major compilers (which I didn't actually know, so that's cool). I imagine tail call elimination is only implemented with compiler optimisation turned on, and I'd be nervous about it always working in any case where it not working would suddenly blow up my stack. You are not guaranteed any particular optimisation that's not in the language standard after all.
|
# ? Nov 25, 2014 21:28 |
|
feedmegin posted:I imagine tail call elimination is only implemented with compiler optimisation turned on, and I'd be nervous about it always working in any case where it not working would suddenly blow up my stack. You are not guaranteed any particular optimisation that's not in the language standard after all. Right. But some languages do guarantee tail-call elimination; it's not an "optimization," it's the semantics of the language.
|
# ? Nov 25, 2014 21:35 |
|
Today I found this, wonderfully complex considering the DLL that's being loaded is created by a project in the same folder as this source (and in fact is already added to some solutions depending on what product you're building)code:
|
# ? Nov 25, 2014 21:38 |
|
QuarkJets posted:In what sense? I've never heard of these before, so I looked up some code examples. Can you explain the advantages for us? Strong typing and exhaustiveness guarantees. You can't easily get those if enums are just aliases for ints. Java-style enums get you part of the way, but still have annoying limitations. Then proper sum types extend the enum concept by allowing you to associate variable data with some members, which is similar to, say, using a combination of an enum and a union in C, but again with actual type-checking. And pattern matching then lets you destructure those and use bits of their data in the matching process. It gives you a very nice natural representation for things like trees with associated data.
|
# ? Nov 25, 2014 21:44 |
|
GrumpyDoctor posted:With tail recursion please keep the furry stuff in the brony threads
|
# ? Nov 25, 2014 21:55 |
|
feedmegin posted:I imagine tail call elimination is only implemented with compiler optimisation turned on, and I'd be nervous about it always working in any case where it not working would suddenly blow up my stack. You are not guaranteed any particular optimisation that's not in the language standard after all. It's actually specifically difficult to do TCO in C++ because of the destructor ordering rules. You need a very carefully written function in order to trigger it.
|
# ? Nov 26, 2014 03:04 |
|
So, uh, I just realized that we have a homespun web server in production at work that will crash and shut down... Instead of sending a 404 response to the browser. I just got brought into the project today. Our whole thing is "minimal viable product" which for whatever the gently caress reason translates to, "Write the worst poo poo you can, as long as it's done quickly." It's Node.js, and we can't use Express or any of the other tried and true options because, "There's too much of a learning curve." So naturally we should hack together some poo poo that uses fs.readFileSync() everywhere and completely fails instead of sending 404 responses.
|
# ? Nov 26, 2014 03:22 |
|
Holy lol, why even use node if you're planning on blocking for simple reads. Bleeding edge mannnnnn...
|
# ? Nov 26, 2014 04:11 |
|
It's real loving bullshit. There's so much wrong with it. We have another server somewhere else that works unless you put a '/' at the end of the URL.
|
# ? Nov 26, 2014 04:23 |
|
GrumpyDoctor posted:Is there language with both goto-able switch cases and tail call elimination? I am unaware of any. This is really easy in Forth. In stack based languages, TCO is a trivial peephole optimization- replace any pattern of "CALL X, RETURN" with "JUMP X". But why use a case statement when you can write a DSL and express your FSMs as tables? code:
|
# ? Nov 26, 2014 04:27 |
|
Electronic Health Record systems are the worst The documentation is scarce, incomplete, and outright wrong in places. The EHR itself runs on a browser version that is slated to EOL soon, and has an insane list of runtime requirements (eg needs to run as the real Administrator superuser, not just an admin-privileged account). And it looks there's no EHR db-level audit records from the API and it would be almost trivially easy to dump all the PHI in the DB. Yesterday, I found this gem in one of our classes (from memory): code:
Paul MaudDib fucked around with this message at 09:00 on Nov 26, 2014 |
# ? Nov 26, 2014 08:52 |
|
JawnV6 posted:How does the anti-switch camp implement FSM's? Do you just resign yourself to a Mealy taste? Come on man, give them a break. What Moore do you want out of them?
|
# ? Nov 26, 2014 09:40 |
Paul MaudDib posted:Yesterday, I found this gem in one of our classes (from memory): Lol, I kind of like this one. It's almost cute. "DateSub... DateDiff... I know it's something like that. Meh, I'll just write a loop." Does the actual function just ignore the day and month the way your version does? Anyone whose day of birth is before the current day gets another year tacked on? That's extra precious.
|
|
# ? Nov 26, 2014 10:13 |
|
Centripetal Horse posted:Does the actual function just ignore the day and month the way your version does? Anyone whose day of birth is before the current day gets another year tacked on? That's extra precious. Java Date is millisecond accurate, so ignoring day/month is not the problem. It's just running the loop one time too much. Edit: I guess for most ways of initializing the birth date it won't matter. Hiowf fucked around with this message at 10:46 on Nov 26, 2014 |
# ? Nov 26, 2014 10:39 |
|
fcvtbuf(3) [which is non-standard but ships as part of newlib] and fcvt_r(3) both kept experiencing memory corruption issues in a recently-launched product when it ran for a while. Couldn't tell what the root cause was, and the source for both in the version of newlib that my toolchain uses was macro'd to hell to the point it confuses both me and the debugger so after a few weeks of fighting I gave up and wrote a limited re-implementation of fconvert(3), which appears to work properly. I am the coding horror.
|
# ? Nov 26, 2014 15:38 |
|
Last night I found a file on one of my old servers I was going to decommission called "gently caress-YOU-BOOST.txt". I can't paste it here because of message length, but have a taste.code:
|
# ? Nov 26, 2014 18:03 |
|
650 lines? call me when the error output is hundreds of megabytes. or gigabytes .
|
# ? Nov 26, 2014 18:23 |
|
FamDav posted:650 lines? call me when the error output is hundreds of megabytes. or gigabytes . No. I will call you when C++ isn't a trash language for idiot babies. (So never)
|
# ? Nov 26, 2014 18:35 |
|
|
# ? Jun 8, 2024 01:54 |
|
GrumpyDoctor posted:Right. But some languages do guarantee tail-call elimination; it's not an "optimization," it's the semantics of the language. Yes, which is why I phrased it the way I did. If you're doing Scheme, it's guaranteed by the language standard so you can rely on it. C++, not, so no.
|
# ? Nov 26, 2014 21:03 |