Print

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

[breadman]

Alright so I now have usable maps, but have to re-write my pathfinding code.

Again, I'm using libtcod with Python. I want each tile on the map to have a specific movement cost  (which ranges from 1 to 4) based on its terrain. Then whenever I hover my mouse over a unit, it should display all the tiles that unit can move to (based on the unit's hard-coded range).

Heh.  When I used libtcod for a PyWeek project, I was accused of not having done anything myself.  Yes, I started with what it gave me, but had to write my own map generator, field of view, and path finder to deal with a feature that the reviewer hadn't even noticed.  I also revised the tileset for another feature.

Anyway, for this particular case, it seems like a flood-fill would be more efficient than pathfinding.  Mark each tile next to the unit with the remaining range, then mark each unmarked tile next to those tiles, continuing until you run out of range.

Static typing saves a lot of headaches down the track. The main "pain in the ass" is that you get more compiler errors. But this doesn't mean you had more errors than in a dynamic typed program, it just means that the compiler picked up potential flaws in your logic before you actually ran the program.

Yes, I've found myself making about the same number of stupid mistakes in both static and dynamic languages, but I've tended to find the important ones faster in Python than in C++.  Partly because automated test cases replace the compiler as my first line of defense, and partly because Python's error messages are actually useful.

Think about how often you wrap "ParseInt" or "IsANumber?" around variables in PHP/Javascript programs. It's something you always need to guard against.

Almost never, if I'm the one calling the function.  Unless we're talking user or network input, which is so much more convenient to deal with in PHP, Javascript, and Python than in C/C++.

Python is just slow compared to c++. It has zero to do with your optimization abilities. The same exact algorithm will almost always be faster in c++.

That's only true while cpython remains the default implementation.  And doesn't usually matter, because I/O wait time tends to dominate almost everything.

I didn't say that C++ is slower than Python. I said that it's harder to optimize things in Python than it is in C++. Mainly because it hides a lot of memory thingies from you.

Still possible to get algorithmic optimization, and this applies to any language. :v

Fair point about the memory thingies; it's so much simpler to write code that does stupid things to memory when you don't have to worry about your own allocations.  Then again, Python sometimes makes it more convenient to change the hotspot implementation without having to change everything that touches the memory structures involved.  Likewise, certain algorithmic optimizations are far easier in a dynamic language.  (I've written iterators in C++, and was appalled at the required boilerplate, both in implementation and in use.)

[Antsan]

@spehss_
Do you have

Code: [Select]

#include <string>
at the beginning of your library or are you using C-style strings (eg. "char * string")?

The to_string function is in the string library, so you need to do the include. It will probably return a "string" and not a "char*".

[Antsan]

I searched on Google for different things and I wasn't lucky in finding at least a LLVM C++ API reference or documentation.
Does anybody know whether something like that even exists?
If not, how would I go about learning the API? I know there is this trick with compiling something to llvm and then compiling it via a C++ backend, but that is relatively confusing to me, as what anything does is quite cryptic and I am also not sure how to translate what I learn there into something I can use for anything else.

[cerapa]

@spehss_
Do you have

Code: [Select]

#include <string>
at the beginning of your library or are you using C-style strings (eg. "char * string")?

The to_string function is in the string library, so you need to do the include. It will probably return a "string" and not a "char*".

He posted his code. It was all written correctly.

[miauw62]

http://stackoverflow.com/questions/12975341/to-string-is-not-a-member-of-std-says-so-g

You have to specially patch MinGW, apparently.

Might be easier to use the C functions to convert them to strings and then convert the char*s to strings?

[Spehss _]

So I'm skipping the graphics thru curses part of the program and coming back to that once I have the core stuff put together.

Seems like my program isn't changing the variable values like I want it to. Say you have 1 d6, and it rolls a 1, so it should be removed and you have 0 d6s, but instead the program continues to display that the amount of d6s is at 1. And if there's a net loss of coins, or negative coins should be added to the total coins, it doesn't deduct the coins. I'm thinking I used the & refer to bit address thing incorrectly.

I don't know.

Spoiler: code (click to show/hide)

Code: [Select]

#include <iostream>
#include <string>
#include <sstream>
#include <stdlib.h>             //srand, rand
#include <stdio.h>              //printf, null
#include <time.h>               //time for srand and rand seed
#include <vector>               //vectors
#include "pdc34dllw/curses.h"   //curses

using namespace std;

string numToString(int num)
{
    stringstream stringStream;
    stringStream<<num;
    return stringStream.str();
}

int flipCoin(int& coinNum)
{
    int result=0;
    for(int i=0; i<coinNum; i++)
    {
        int flip=((rand()%2)+1);
        if(flip==2)         //heads=2
        {
            result+=1;      //if heads, get 1 coin
        }
        else if(flip==1)    //tails=1
        {
            result-=1;      //if tails, lose 1 coin
        }
    }
    if(result>0)
        cout<<"You gained "<<result<<" coins from flipping more heads than tails."<<endl;
    else if(result<0)
        cout<<"You lost "<<result<<" coins from flipping more tails than heads."<<endl;
    return coinNum+result;
}

int rollD6(int& diceNum, int& coinNum)
{
    int result=0;
    int gains=0;
    for(int i=0; i<diceNum; i++)
    {
        int roll=((rand()%6)+1);
        if(roll==1)
        {
            result--;
        }
        else
        {
           gains+=roll;
        }
    }
    if(result<0)
        cout<<"You lost "<<result<<" d6 from rolling 1's."<<endl;
    if(gains>0)
        cout<<"You gained "<<gains<<" coins from rolling six sided dice."<<endl;
    coinNum+=gains;
    return diceNum+result;
}

void displayInventory(int coinNum, int d6Num, int d20Num)
{
    if(coinNum==0)
    {
        cout<<"You have no coins."<<endl;
    }
    else
    {
        if(coinNum==1)
        {
            cout<<"You have only "<<coinNum<<" coin."<<endl;
        }
        else
        {
            cout<<"You have "<<coinNum<<" coins."<<endl;
        }
    }
    if(d6Num!=0)
    {
        if(d6Num==1)
        {
            cout<<"You have only "<<d6Num<<" six-sided die."<<endl;
        }
        else
        {
            cout<<"You have "<<d6Num<<" six-sided dice."<<endl;
        }
    }
    if(d20Num!=0)
    {
        if(d20Num==1)
        {
            cout<<"You have only "<<d20Num<<" twenty sided die."<<endl;
        }
        else
        {
            cout<<"You have "<<d20Num<<" twenty-sided dice."<<endl;
        }
    }
}

void takeTurn(int& coinNum, int& d6Num, int& d20Num)
{
    if(coinNum!=0)
    {
        flipCoin(coinNum);
    }
    if(d6Num!=0)
    {
        rollD6(d6Num, coinNum);
    }
    if(d20Num!=0)
    {
        //add rolld20 function
    }
}

int main()
{
    srand(time(NULL));
    char player_in='y'; //gets player input
    int coinTotal=1;
    int d6Total=1;
    int d20Total=0;

    do
    {
        cout<<"\n\n"<<endl;
       displayInventory(coinTotal, d6Total, d20Total);

       cout<<"What will you do? (f)lip, (s)tore, (b)ank, (q)uit"<<"\n>>";
       cin>>player_in;

       if(player_in=='f')
       {
           takeTurn(coinTotal, d6Total, d20Total);
       }
    }while(player_in!='q'&&player_in!='Q');

//    initscr();
//    refresh();
//    getch();
//    endwin();

    return 0;
}


[cerapa]

I think you forgot what you were doing.

flipCoin() calculates the result of flipping the coins, but doesn't actually change the number of coins you have. You have to replace
flipCoin(coinNum);
in taketurn() with
coinNum = flipCoin(coinNum);
or make flipcoin() set the amount of coins.

In rollD6() you do the same thing, except you change the number of coins correctly, but return the number of D6's without changing their amount.


It's a very easy mistake to make when you are just starting out, but try to pay attention to where you are supposed to be changing your variables. I'm guessing originally you wanted to have it return the value, but then wanted to make it use references instead?

[Reelya]

In this case, you basically need to track through the program, find the line where it is being changed, then make sure the changes are applied to where they're needed. But there doesn't seem to be any line where coinNum is being updated. you're returning "coinNum+result", but that return value isn't stored anywhere, it's just thrown away. And then sometimes you're using pass by reference, and updating things that way. You should exclusively use return values or pass-by-reference, since mixing and matching has made things hard to comprehend.

Functions that change their parameters should be used very sparingly. They can cause issues especially if you provide your code to someone else, since changing parameters that are passed to a function breaks the principle of encapsulation. In this case, it has made your program hard for you to work out what it's actually doing, since you have separate coinNum variables at several levels of scope, all chaining together.

Having a single global variable coinNum would actually be better than that. But globals are frowned upon too, so the best solution would be to encapsulate everything in a class, make coinNum a member variable of the class, and have the flipCoin etc be member functions. Then they don't need to have the coinNum passed around as a function parameter at all. So you know that any time the coinNum is changed, it's actually the same variable, not separate variables which happen to have the same name. You'd end up with a much simpler program structure and less scope issues.

[Dutrius]

Ok, so In C#, if I do this:

Code: [Select]

int[] somearray;

somearray = new int[5];
//put values in somearray

somearray = new int[6];
//do something

Does the second array declaration completely overwrite what's in the previous declaration, or does it just add a new element to the array?

[Reelya]

This will help you determine in general whether it can be the case:

When you see:

x = y

in a program, the program first calculates the value "y", then blindly sends the resultant value to "x". So what is on the left hand side of an equation can never affect what occurs on the right hand side of an equation. Basically, "x" is not an input to process "y". (There could be some fringe cases with overloading the = operator in c++, but they'd be unlikely).

In your case, "new int[6]" allocates a block of memory that can hold 6 ints, and has a return value which is a pointer to the start of the new block. The "new" operator is basically a function which knows what values have been fed into it: the int type, and the number 6, so it has no idea what you intend to do with the output or even whether "somearray" exists.

So yup, new int[6] makes a whole new array, then the "=" operator assigns the start of that array to the label "somearray". If there was anything already in "somearray" it would be deleted and "somearray" now points to the new array's location.

C# provides a resize command as a function inside the array type:

http://www.dotnetperls.com/array-resize

What this does is (1) creates a new bigger array using "new", then (2) copies the old values into the new array, then (3) points the array's name at the new table.

[Dutrius]

Thanks. I thought this was the case, but I just wanted to be sure. I didn't know about the resize command though. That will come in useful.

[FallacyofUrist]

This thing is dense. Wow. I do a little JavaScript, but anything else isn't within my abilities. I might stop by here on occasion when I'm not doing forum games.

[alway]

I distinctly remember having a similar issue with to_string at some point, but that might have just been because I didn't have c++11 turned on.

Meanwhile you can convert numbers to string through a stream. Less convenient than to_string, but you gotta do what you gotta do.

Code: [Select]

#include <sstream>

stringstream streamofstrings;
streamofstrings << yourint;
resultingstring = streamofstrings.str();
String doesn't look like a word anymore.

Abominations like this is why one of the first things a programmer should do in C++ is implement their own string class for all future projects. Null terminated C-style strings will give you nightmares in debugging, std::string is an abomination that looks like you're trying to summon cthulhu 50% of the time you want to do something simple with it.

[cerapa]

I distinctly remember having a similar issue with to_string at some point, but that might have just been because I didn't have c++11 turned on.

Meanwhile you can convert numbers to string through a stream. Less convenient than to_string, but you gotta do what you gotta do.

Code: [Select]

#include <sstream>

stringstream streamofstrings;
streamofstrings << yourint;
resultingstring = streamofstrings.str();
String doesn't look like a word anymore.

Abominations like this is why one of the first things a programmer should do in C++ is implement their own string class for all future projects. Null terminated C-style strings will give you nightmares in debugging, std::string is an abomination that looks like you're trying to summon cthulhu 50% of the time you want to do something simple with it.

I don't see how rolling your own string class fixes the lack of to_string?

The example there is just an ass-backwards way to convert variables into strings. It's an abomination, I admit, but it isn't a feature of std::string.

[i2amroy]

Didn't C++11 fix up the vast majority of that though? It's not quite as nice as python, but they at least have a nice handful of object->string conversion functions now.