What are you working on? v67 - March 2017

[Ott]

I'm working on a CS degree.
But that's boring so I've been doing some visual programming on the side:
[video=youtube;e1yDit9aHuE]https://www.youtube.com/watch?v=e1yDit9aHuE[/video]

[Nookyava]

I work full time and am doing College full time so finding the time can be tough.

Doesn't help my work just blocked my connection to my repos so I can't pull or commit at work now. I'm help desk so if nothing is broken my free time was at least productive, now it's just spent watching YouTube.

[NoRegard]

[QUOTE=Sam Za Nemesis;52787003]How do you people cope with your daily schedule? I think that working on interesting out-of-work projects with other people is my best way to succeed in (and out) of this country but college and work seem to eat all my time and motivation to do anything and I feel like I'm deeply failing because of that[/QUOTE]

It's really tough.
I find it hard to balance between not working on any project at all or stressing out so much that I can't sleep and get stomach pain.
Currently I try to limit myself only to work on projects 1 weekday a week. I feel that as programmers we somehow need to work in our free time to ensure that we stay in touch and don't become irrelevant. As in learning new cool shit and having stuff to show off to new potential employees. You can't always show off the stuff you have done at your current job, as it may be under an NDA.

[Clive]

[QUOTE=Nookyava;52791128]I work full time and am doing College full time so finding the time can be tough.

Doesn't help my work just blocked my connection to my repos so I can't pull or commit at work now. I'm help desk so if nothing is broken my free time was at least productive, now it's just spent watching YouTube.[/QUOTE]

That's shitty, can't you just tell them you had some files on there to help you with your work. That's a pretty dick move by them.

[reevezy67]

Work on a local copy.

[editline]18th October 2017[/editline]

I used to have an uneventful help desk job, thinking back I wish I'd used the spare time like that.

[lvivtotoroyt]

A Super Mario Maker for the PC (Level editor's finished, but I don't have an artist to help so i'm probably fucked)

And recreating a couple of gamemodes from GMT.
Almost finished ball race
[video=youtube;mvY8qyUJgRU]https://www.youtube.com/watch?v=mvY8qyUJgRU[/video]

[Tamschi]

[QUOTE=lvivtotoroyt;52792426]A Super Mario Maker for the PC (Level editor's finished, but I don't have an artist to help so i'm probably fucked)

And recreating a couple of gamemodes from GMT.
Almost finished ball race
[video=youtube;mvY8qyUJgRU]https://www.youtube.com/watch?v=mvY8qyUJgRU[/video][/QUOTE]

I'd put the level end trigger either quite a bit below the finish hole or set it to full containment only, if that's possible. That way the players will get to actually fall into there.

[Zero-Point]

[QUOTE=Clive;52788368]I recommended that to my boss (or something similar) but he told me he wants something to be made by us so we can say that we made it for them, I'm not even a programmer for this company I'm a support guy so god knows why I am tasked with this.[/QUOTE]

Because to some folks, "knows about computers" == "knows EVERYTHING about computers".

[editline]18th October 2017[/editline]

[QUOTE=Nookyava;52791128]I work full time and am doing College full time so finding the time can be tough.

Doesn't help my work just blocked my connection to my repos so I can't pull or commit at work now. I'm help desk so if nothing is broken my free time was at least productive, now it's just spent watching YouTube.[/QUOTE]

I just find it funny that they blocked your ability to commit/pull, likely as a measure to ensure you're "not wasting company time", yet they still allow Youtube, which can be more of a time-pit than Wikipedia.
"Well that was an interesting video, now back to- Ooh, a video on how to pan for gold..."

[DrLuke]

[QUOTE=Nookyava;52791128]I work full time and am doing College full time so finding the time can be tough.

Doesn't help my work just blocked my connection to my repos so I can't pull or commit at work now. I'm help desk so if nothing is broken my free time was at least productive, now it's just spent watching YouTube.[/QUOTE]

Use a proxy!

[paindoc]

So I kinda mentioned the JSON work I'm doing already, and I think I mentioned that I'm working on something that is kind of our CEO's baby.

So I was a little stressed telling him that I'd like to replace the old text-based input system requiring tons of switches, if-elses, and edge-cases with the simplified JSON system that requires less code and makes writing setup files a lot easier. This was also exacerbated by the fact that we are both awkward and stumble around words and sentences a lot. When I pointed out that we can serialize structs and classes from his program to and from a file pretty freely, though, and that he could potentially "pause" the simulation by exporting the current summed program state to a JSON file, he seemed pretty interested. And I felt relieved that he was okay with me implementing this feature, thank god

Everything was well and good until he mentioned what NASA wants to do around this contract, and the kinds of absolutely mental math involved. Its the kind of stuff that has literally never been done before in the history of spaceflight... so yay me, I guess?

[sp]honest to god it sounds like something out of kerbal space program, except even kerbals might consider it too crazy[/sp]

[lvivtotoroyt]

[QUOTE=Tamschi;52792578]I'd put the level end trigger either quite a bit below the finish hole or set it to full containment only, if that's possible. That way the players will get to actually fall into there.[/QUOTE]

Thanks, I'll do that

[Nookyava]

[QUOTE=DrDevil;52792707]Use a proxy![/QUOTE]

Can't since they use a proxy themselves, and so far what I've seen is if I turn said proxy off it nukes my connection to websites.

I get why their security team is doing it - because why leave these ports open? But to even sniff out the HTTPS port and check if it's something like pulling/pushing to a repo versus a web page? That's going pretty far.

So yea I've been trying to find a work around but nothing so far.

[paindoc]

who in the how the fuck arrives at the conclusion of 1. allocating a new object to return from a function returning const T& 2. deletes [I]a fucking operand[/I]. How do you even arrive at the logic of #2? That's not how this works. That's not how [I][B]any[/B][/I] of this works.

[cpp]
const number_array &number_array::operator-( const number_array &right )
{
// Local Variables
int i;

number_array
*ptr_ResultArray = NULL;

// Create proper Array object
ptr_ResultArray = new number_array( dm_int_size ); // Create new Array for result

if ( &right != this ) // check for self-assignment
{
// Check to ensure the arrays are the same size
if( dm_int_size == right.dm_int_size )
{
for( i = 0 ; i < dm_int_size ; i++ )
{
// Add individual array elements
(*ptr_ResultArray)[i] = dm_pdb_ptr[i] - right.dm_pdb_ptr[i];
}

if ( right.mf_e_getRealTempState() == e_Temp ) // deallocate right if it was a temporary result
{
delete (&right);
}
}
else
{
fprintf( g_ErrorLogFile, "ERROR - number_array::operator-( const double &right ): Array is NOT initialized, it has NO size.\n");
fflush( g_ErrorLogFile );
}
}

// Set this num_array as temporary structure (intermediate)
ptr_ResultArray->mf_bl_setRealTempState( e_Temp );

return *ptr_ResultArray;
}
[/cpp]

:why:

[sp]there are FOUR instances of this class, one for doubles, ints, unsigned ints, and another one for double's with duplicated code where the ONLY difference is the default argument in the constructor[/sp]

[editline]18th October 2017[/editline]

I'll take what are templates and somehow deleting stack variables for 500 please

[editline]18th October 2017[/editline]

like how the fuck does this even work and why doesn't the compiler even try to warn me goddamn

[editline]18th October 2017[/editline]

who needs RAII when you have enums and if-statements in operators

[drrs]

All I have to say to the person that made that is :goodjob:

[WTF Nuke]

Been working on a tiny JSON parser. Pretty fun and easy spec, and it's fun to work in the constraints of constexpr.

[DrLuke]

[QUOTE=Nookyava;52793091]Can't since they use a proxy themselves, and so far what I've seen is if I turn said proxy off it nukes my connection to websites.

I get why their security team is doing it - because why leave these ports open? But to even sniff out the HTTPS port and check if it's something like pulling/pushing to a repo versus a web page? That's going pretty far.

So yea I've been trying to find a work around but nothing so far.[/QUOTE]

You could use an external http proxy

[editline]19th October 2017[/editline]

Is port 22 open? You could use an ssh forward to tunnel out

[Nookyava]

[QUOTE=DrDevil;52796113]You could use an external http proxy

[editline]19th October 2017[/editline]

Is port 22 open? You could use an ssh forward to tunnel out[/QUOTE]

Port 22 may be open but they may also be sniffing traffic among that port too. Give it a shot.

Any recommended http proxies? I can set Firefox dev up to utilize it by normal and they can't setup a group policy for Firefox. So if it works I may be good then

[DrLuke]

[QUOTE=Nookyava;52796288]Port 22 may be open but they may also be sniffing traffic among that port too. Give it a shot.

Any recommended http proxies? I can set Firefox dev up to utilize it by normal and they can't setup a group policy for Firefox. So if it works I may be good then[/QUOTE]

ssh is encrypted, so they can sniff until their noses fall off!

[Tamschi]

[QUOTE=DrDevil;52796913]ssh is encrypted, so they can sniff until their noses fall off![/QUOTE]

Does that apply to the initial handshake too? Unless both parties immediately start with e.g. a super plain DH key exchange, the tunnel itself should still be easily identifiable.

[DrLuke]

[QUOTE=Tamschi;52798252]Does that apply to the initial handshake too? Unless both parties immediately start with e.g. a super plain DH key exchange, the tunnel itself should still be easily identifiable.[/QUOTE]

Of course the presence of the tunnel is easy to detect. Just say that you need it to manage some remote servers, and the traffic must be encrypted to protect from man in the middle attacks.

[reevezy67]

Shadowsocks usually works well for this kind of thing. Looks similar to https traffic. I use it to access stuff when I'm in China.

[Nookyava]

[QUOTE=DrDevil;52800481]Of course the presence of the tunnel is easy to detect. Just say that you need it to manage some remote servers, and the traffic must be encrypted to protect from man in the middle attacks.[/QUOTE]

Issue is I don't think they're that dumb, and there's no reason I'd be able to give them that I need to access said remote servers.

So I'm pretty tied down.

[Topgamer7]

[QUOTE=Nookyava;52801622]Issue is I don't think they're that dumb, and there's no reason I'd be able to give them that I need to access said remote servers.

So I'm pretty tied down.[/QUOTE]

Buy a wifi card, build a cantenna, point it at the nearest starbucks. If anyone asks, tell them it's an art piece from your nephew.

[paindoc]

hmmm this seems odd, storing a pointer and a reference?

[t]https://i.imgur.com/dWJJF43.png[/t]

i... don't even. i can't even. how do you even [I]arrive[/I] at this fucking logic?

[t]https://i.imgur.com/nvrSY2G.png[/t]

.... :goodjob:

[editline]edited[/editline]

guy who made this has a phd and claims to be an algorithm developer - did some research and found a preview of his next algorithm: [URL="https://i.redd.it/lwin56fisdsz.png"]LINK[/URL]

What unadulterated brilliance!

[editline]20th October 2017[/editline]

how does this even fucking compile let alone work?

"you can't make a reference null", they said

"hold my beer"

[Karmah]

for what purpose would you even need to store both a pointer and a reference to something at the same time in the same place
let alone make the pointer null and then store a reference to said null pointer

Is the whole point lazyness so that he won't have to (->) and can just (.) them instead? :v:
Because if so :goodjob: he saves 2 characters every time, dropping _p-> to .

[WTF Nuke]

It's super useful when you want undefined behaviour.

[paindoc]

[QUOTE=Karmah;52803713]for what purpose would you even need to store both a pointer and a reference to something at the same time in the same place
let alone make the pointer null and then store a reference to said null pointer

Is the whole point lazyness so that he won't have to (->) and can just (.) them instead? :v:
Because if so :goodjob: he saves 2 characters every time, dropping _p-> to .[/QUOTE]

me and the two senior C++ devs at work (yay there's one more of us now) just spent half an hour pondering how the fuck they arrived at this conclusion, because really, how do you? Its the same person that tried to delete an operand and created a new stack object for a function returning a const reference - it has to be, because the brand of dumb is unique

i can't think of any language you'd do this in, or any reason you'd ever do this. its as much an enigma as it is a stupid fucking thing to do

[editline]20th October 2017[/editline]

I quite like refactoring code, pulling it apart into cleaner chunks, adding modern C++ bits to make things clearer and safer, and getting the feeling of renovating an existing system

but fuck this stuff is testing my goddamn patience

[Karmah]

Spent all day today learning how to actually configure my project for CMake. Holy hell that took a while, but only because I'm picky as fuck and needed things to look nice and neat.

I'm not sure to what degree this works with other compilers or IDE's, but I just needed a way to create and update projects with half a dozen dependencies that could be found in different directories on different systems.

And now that it [I]works on my machine™ [/I]&#8203;I can finally call it a day

[Dr Magnusson]

[QUOTE=paindoc;52803772]me and the two senior C++ devs at work (yay there's one more of us now) just spent half an hour pondering how the fuck they arrived at this conclusion, because really, how do you? Its the same person that tried to delete an operand and created a new stack object for a function returning a const reference - it has to be, because the brand of dumb is unique

i can't think of any language you'd do this in, or any reason you'd ever do this. its as much an enigma as it is a stupid fucking thing to do

[editline]20th October 2017[/editline]

I quite like refactoring code, pulling it apart into cleaner chunks, adding modern C++ bits to make things clearer and safer, and getting the feeling of renovating an existing system

but fuck this stuff is testing my goddamn patience[/QUOTE]

Only reason I can think of, is if the pointer might sometimes be null, but you *really* hate using the -> operator. Which, mind you, is a terrible fucking reason, but it sounds like reason isn't exactly this guy's specialty.

[Zero-Point]

I have no idea what anybody is really saying at this point, and even I think it sounds like a dumb idea.