Code for Chudnovsky Algorithm [or How to Translate an Equation into C++ Code]

[TechnoCat]

*shrugs*

Well, I've tacked .0 onto the ends of all of the integers in the line, and I'm now down to one error:

[path to project]/calculatePi.cpp:38: error: call of overloaded 'pow(double, long int&)' is ambiguous
/Developer/SDKs/MacOSX10.6.sdk/usr/include/architecture/i386/math.h:343: note: candidates are: double pow(double, double)
/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/cmath:373: note:                 long double std::pow(long double, int)
/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/cmath:369: note:                 float std::pow(float, int)
/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/cmath:365: note:                 double std::pow(double, int)
/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/cmath:361: note:                 long double std::pow(long double, long double)
/Developer/SDKs/MacOSX10.6.sdk/usr/include/c++/4.2.1/cmath:357: note:                 float std::pow(float, float)

, which I've isolated to pow (-1.0, k). I guess I could replace that with something that returns -1 if k is odd, and 1 if it's even, but I'm feeling a bit lazy right now :P

[simeonz]

A wild guess is that the second argument of this function is the result from computing factorial. If this is the case, cast the result from factorial to double.

[TechnoCat]

k, as in for (long k = 0 ; k < loops ; k ++ ); k, as in 1 / π = 12 [k = 0, Σ, ∞]

[I used k instead of i because k was in the original formula.]

[simeonz]

Well, cast k to double or float, like pow (-1.0, (double)k).

[TechnoCat]

Okay. Now I have two errors:[output]In file included from [path to project]/calculatePi.cpp:1:
[path to project]/header.h:10: error: 'string' has not been declared
[path to project]/calculatePi.cpp: In function 'int negOneToTheXthPower(long int)':
[path to project]/calculatePi.cpp:51: error: invalid conversion from 'const char*' to 'int'

[simeonz]

The string type is defined in header <string>. You need to include it in the beginning of header.h. The second error sounds obscure. I can't understand where in negOneToTheXthPower is char* needed.

[ne555]

( (k%2==0)? 1: -1 ) pow is overkill.
-1^k
if k is even -> 1
if k is odd -> -1

Okay, I fixed that error by replacing “1.5” with ‘3 / 2 ’

Integer division returns an integer. In this case 3/2=1

Where are you using a string?

Side note: the numbers in C have finite precision.

[TechnoCat]

@ne555

simeonz said that the caret is used for bit fiddling, not powers.

Okay, I've changed it back to ‘1.5’ .

Never mind; I forgot to type using namespace std ; is all. Now there's just the

Undefined symbols:
  "factorial(int)", referenced from:
      calculatePi(long) in calculatePi.o
      calculatePi(long) in calculatePi.o
      calculatePi(long) in calculatePi.o
ld: symbol(s) not found
collect2: ld returned 1 exit status

linking error.

I know; I'm planning to change all the doubles to this data type that supposedly has infinite precision (enough precision, anyway) . There's a bunch of problems with that that I'll need some help with though, so I'm just using doubles for now.

[TechnoCat]

Oh, and there's the control-reaches-end-of-non-void-function warning at the end of negOneToTheXthPower (long x)^*; how do I tell Xcode to ignore that? I think it's #pragma warning disable and then something, but I can't remember.…


* all of the return statements are in if statements, and if for some reason x is not odd or even there's an else statement that ensures that the script will never reach the final “}” , so that's not a problem

[TechnoCat]

Finally figured it out: factorial.cpp is in a totally different project. I had added a reference to it, not copied it into this project's folder. No wonder #include "factorial.cpp" wasn't working. Xcode couldn't figure out what I was talking about. Anyway, I included it using the full path and hit Build again, which got me protests about my calls to factorial (long x) . Apparently they were too ambiguous. I fixed them by chopping off the “.0”s in the parameters and hit Build, which got me a duplicate symbol error about factorial (long x) . I tried to fix that by removing the import for factorial.cpp , but that got me a symbol not found error. I think this is the last error. I'll try to figure it out, but again, if anyone can help, I'd really appreciate it.

[simeonz]

You should not include implementation files, only header files. Including implementation files (.cpp extension) with the include directive causes redefinitions. I guess this is what the compiler meant by "duplicate symbols".

[TechnoCat]

Okay, so what do I do? If I remove that include, I get a symbol not found error.

[simeonz]

I don't know exactly why your IDE is not compiling factorial.cpp and linking against the resulting code. You need to create and include "factorial.h", a header file that declares the factorial function without defining it. But I really don't know what you need to do to include factorial.cpp in your project. As a last resort, if you are in a hurry, copy-paste the contents of factorial.cpp in main.cpp. Investigate the problem later.

[TechnoCat]

I found the problem. I had declared factorial as int factorial (int num) in the header file and long factorial (long num) in the implementation file. *facepalms*