Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[dreadmullet]

If anything, "~=" should mean "is almost equal to", as a way of comparing floats and such. I know Lua uses "~=" as a replacement for "!=", which is strange, since it's supposed to be heavily influenced from C++. Eh, I don't understand what's wrong with "!=".

[Mego]

A coworker, our current senior-most-programmer, has decided the new system he's going to be working on will use "~" to represent the not symbol (and "~=" to represent not equals) instead of the more tradition ruby "!" and "!=".

His reason is that "I don't like how bangs look, the tilde looks much better."

This explains why he previously created his own "Not" method and used that almost exclusively up until this point, I suppose.

In other news, I'm now officially hunting for a new job.

If he had cited unary operators as a reason, then that would be somewhat valid. Preferring how tildes look over bangs is not valid. He is bad and should feel bad.

[Nadaka]

but ~= is the approximately equal operator!

[Mego]

but ~= is the approximately equal operator!

In mathematics, yes. But computer science has a tendency to shit on mathematical notation. Also, if you can find me a good use for the ~= operator in a programming language that can't be done just as succinctly and lucidly, or more so, then I will give you an internet cookie.

For example, the ~= operator would require 3 arguments: the two floats being compared, and the greatest allowed difference for the floats to still be considered approximately equal. That can just as easily be done like so:

Code: (Python) [Select]

import math
def approx(a,b,epsilon):
    return math.abs(a-b) <= epsilon


[GlyphGryph]

but ~= is the approximately equal operator!

And you get the real meat of why this makes me so immensely negative emotioned. The code it's replacing even USED that mean approximately equal - not it will mean the exact opposite!

I'm going to cry if I keep talking about this. I need to go back to trying to figure out event handling for my game and calm down.

[MadocComadrin]

A coworker, our current senior-most-programmer, has decided the new system he's going to be working on will use "~" to represent the not symbol (and "~=" to represent not equals) instead of the more tradition ruby "!" and "!=".

His reason is that "I don't like how bangs look, the tilde looks much better."

Ruby? How about nearly any widely used language that doesn't use the explicit word "not?"
Please tell him the last time you see him that you don't like the way his face looks and that a paper bag would look much better. :/

[kaijyuu]

What the fuck use is an "approximately equal" operator? Within what bounds? It just seems so... vague and arbitrary.

[eerr]

What the fuck use is an "approximately equal" operator? Within what bounds? It just seems so... vague and arbitrary.

Float values aren't perfectly accurate.

They are NOT decimals, they are fractions, and get rounded to the nearest fraction.

From the larger double operations, no less.

The perfect accuracy is lacking.

[Mego]

What the fuck use is an "approximately equal" operator? Within what bounds? It just seems so... vague and arbitrary.

Float values aren't perfectly accurate.

They are NOT decimals, they are fractions, and get rounded to the nearest fraction.

From the larger double operations, no less.

The perfect accuracy is lacking.

Maybe, instead of what I thought, it could be to determine if two floating-point numbers could be results of the same operation?

Damnit, we're putting too much effort into something stupid.

[MadocComadrin]

What the fuck use is an "approximately equal" operator? Within what bounds? It just seems so... vague and arbitrary.

Significant Figures vs Accepted Values/Values with more Significant Figures.

[Mego]

What the fuck use is an "approximately equal" operator? Within what bounds? It just seems so... vague and arbitrary.

Significant Figures vs Accepted Values/Values with more Significant Figures.

SigFigs? IN MY COMPUTER SCIENCE?

GTFO.

[MadocComadrin]

You don't understand: we need to ally with the Physicists and the Chemists before they ally with the Biologists so we can make androids to fight their genetic super-mutants!

[alway]

Pfft, we're programmers. We use magic-boxes to build universes. We beat all three by distracting them until they forget all they have learned. "I could make a super-mutant army.... but on the other hand.... just one more turn. OuO"

I have a hard enough time not distracting myself by opening up that erosion simulation, letting it run for hours, and just sitting there, watching it. It's terrible for productivity >_<

[Skyrunner]

HELP Dx
The problem: There is a vector of a structure that includes a string, "plantname". It's referenced in a different class to draw it to the screen. However, somewhere between writing it and reading it, it gets wiped.

Here's the snippets of relevant code:

Spoiler (click to show/hide)

Code: (SoilMap class.h) [Select]

enum planttype {DOUBLEQUOTE = 34, SINGLEOPENQUOTE = 39, SINGLECLOSEQUOTE = 96};

class SoilMap
{
public:
  struct plants
    {
      string plantname;
      void setName(string str) {plantname = str;}
     };

  void GenerateGrass(); //generate the grass map;.
  plants plants_at(int x, int y) {if (y*h + x < plantmap.size()) return plantmap.at(y*w + x); else {plants temptile; return temptile;}}
 
protected:
  vector<plants> plantmap;
};
 

Code: (SoilMap class.cpp) [Select]

#include "SoilMap class.h"

void SoilMap::GenerateGrass()
{
plants *temp;
temp = new plants();

/* code that is not relevant */

for (int ycounter = 0; ycounter < h; ycounter++)
  for (int xcounter = 0; xcounter < w; xcounter++)
    {
      /* not relevant */
      string tempstr;
      if (temp->character == DOUBLEQUOTE)
        tempstr.assign("Fescue grass");
      else if (temp->character == SINGLECLOSEQUOTE)
        tempstr.assign("Ryegrass");
      else if (temp->character == SINGLEOPENQUOTE)
        tempstr.assign("Ryegrass");
      else
        tempstr.assign("Unidentified");

temp->setName(tempstr);
    }

   plantmap.push_back(*temp);    //add to plantmap vector
}

Code: (Display.h) [Select]

class sideDisplay: public GUIHandler
{
public:
sideDisplay();
void draw();
void update(SoilMap &map, Cursor &cursor); //while cursor is visible

protected:
vector<pair<string,pair<TCODColor, TCODColor>>> infovector;
TCODColor darkbrown;

};


Code: (Display.cpp) [Select]


void sideDisplay::update(SoilMap &map, Cursor &cursor)
{
SoilMap::tile *tilepointer;
SoilMap::plants *plantpointer;

if (cursor.visible == true)
if ((cursor.getX() < 53) & (cursor.getY() <= 46) & (cursor.getY() >= 10))
{
plantpointer = & map.consoletoplant(cursor.getX(), cursor.getY());
if (plantpointer->plantname.size() != 0)
infovector.push_back(pair<string, pair<TCODColor, TCODColor>>(plantpointer->plantname, pair<TCODColor, TCODColor>(TCODColor::lerp(plantpointer->background, TCODColor::black,0.5f), plantpointer->foreground)));

tilepointer = & map.consoletotile(cursor.getX(), cursor.getY());
infovector.push_back(pair<string, pair<TCODColor, TCODColor>>(tilepointer->getTileName(), pair<TCODColor, TCODColor>(TCODColor::lerp(tilepointer->foreground, TCODColor::black,0.5f), tilepointer->background)));

}
}

void sideDisplay::draw()
{
console->clear();

int counter = 0;
for (auto it = infovector.begin(); it < infovector.end(); it++)
{
console->setDefaultBackground(it->second.first);
console->setDefaultForeground(it->second.second);
console->print(0, counter, "%s", it->first.c_str());
counter++;
}

infovector.clear();
}

Basically what the update() function does is get the coordinates from a cursor, translate them into the Soilmap's coordinates, and fetches the correct plants stored in the vector. It then pushes back the plant name, two TCODColor variables.
Then, the draw() function takes the colors and sets them, then takes the c_str() of the string and blits that to the screen.

Debugging checks that (1) Using a constant string in the update() function makes it work correctly (2) In the update() function, the pointer to the plants object tells me that the plantname variable is empty, (3) An almost identical structure tile works perfectly fine, though it uses a slightly different way of looking up the string _{(it uses a lookup function, though, not a lookup table)}.

If this doesn't suffice, I could provide a dropbox link to the entire project. :<
It is using SDL and libtcod.


Question: What could be possible reasons the string in the plants is wiped between writing it and reading it?

This has stumped me for a good hour.

[DrKillPatient]

I've got a programming-logic quandary:

In a tile-based game, caves are generated randomly. Some are disconnected from others, and the game flood-fills these areas, setting each isolated group of tiles as a separate subregion. These subregions are stored as a vector of 2D coordinate points, which are run through one-by-one to apply various changes to a subregion. Each 2D point stored in the vector refers to a point on the map that is within the subregion. In addition, each tile on the map stores an integer referring to the index of the subregion in which it resides (-1 if it does not belong to a subregion).

If I want to ensure that each subregion is connected to, say, all other subregions within 20 tiles (IFF those subregions can be reached with a straight line), what's the best way to compute that? All I need is to find one point on the edge of one subregion, then find a reasonably close point on the nearest subregion. With that data, I can begin to draw tunnels.