v0.3.1
Esta DLL implementa funciones para parsear ficheros CSV, donde se utiliza el carácter ',' como elemento separador, y solo contienen valores numéricos. El carácter '#' marca que el resto de la linea se ignore, siendo útil para agregar comentarios y cabeceras en los ficheros.
Si un número comience con "0x" se interpretará en hexadecimal. Si comienca precedido de un 0, entonces se interpretará en base octal.
Todos los buffers internos son de 256 bytes, con lo puede que no pueda procesar correctamente lineas mas largas de 255 bytes;
Requires Watcom C 10.6 installed on a DOS enviroment (dosbox, dosemu, virtual machine or native install of MS-DOS/PC-DOS/DR-DOS, FreeDOs or OpenDOS).
To compile, simply launch make2.bat like if it was make coommand :
make2 all
A debug version could be generated. The debug version generates a CSV.LOG file with information about the data being readed by the DLL. On every invocation or execution, new content its append at the end of the log file.
make2 csv_dbg.dll
Actually only works with CicTec's fork on Linux (64 bit). Simple launch
./makeDll.sh and it will generate a csv.so that could be used on DIV DX.
INT setCSVSeparator(STRING separator)
Sets the seperator token. The string it's used by C's strok function.
INT readCSVToIntArray(STRING fileName, OFFSET offset, INT numberOfElements)
Reads a CSV file on the path stored on the string fileName, and stores the values in an integer array or struct pointed by offset. The max number of elements read will be determinated by numberOfElements. The return value by the function, it's the number of elements read or -1 if there was an error. If offset is 0, then will not store any read element. It only will return the number of elements that are on the CSV file.
INT readCSVToWordArray(STRING fileName, OFFSET offset, INT numberOfElements)
Reads a CSV file on the path stored on the string fileName, and stores the values in an word array pointed by offset. The max number of elements read will be determinated by numberOfElements. The return value by the function, it's the number of elements read or -1 if there was an error. If offset is 0, then will not store any read element. It only will return the number of elements that are on the CSV file.
INT readCSVToByteArray(STRING fileName, OFFSET offset, INT numberOfElements)
Reads a CSV file on the path stored on the string fileName, and stores the values in an byte array pointed by offset. The max number of elements read will be determinated by numberOfElements. The return value by the function, it's the number of elements read or -1 if there was an error. If offset is 0, then will not store any read element. It only will return the number of elements that are on the CSV file.
Uso típico:
/**
* Genera la ruta relativa a los ficheros del juego
*/
function pathResolve(file)
begin
return (/* "\foo\bar\" + */ file);
end
/**
* Lee un fichero CSV con datos de juego
*/
function loadData(dataFile, _offset, size)
private
string _path;
int _retVal = 0;
string _msg;
begin
_path = pathResolve(dataFile);
_retVal = readCSVToIntArray(_path, _offset, size);
if (_retVal <= 0)
_msg = "Error al abrir fichero de datos: " + _path;
write(0, 0, 0, 0, _msg);
loop
// abortamos ejecución
if (key(_q) || key(_esc))
let_me_alone();
break;
end
frame;
end
end
return(_retVal);
end
...
loadData("dat\mydata.dat", offset myArray, sizeof(myArray));
Determinar la cantidad de elementos en un fichero antes de cargar el contenido:
/**
* Función que carga el contenido de un fichero CSV en un array dinamico
(generado por malloc)
* Devuelve un puntero al array dinamico
*/
function loadAndAllocateData(dataFile)
private
string _path;
int _retVal = 0;
int _nElements = 0;
string _msg;
int pointer _data;
begin
_path = pathResolve(dataFile);
_nElements = readCSVToIntArray(_path, 0, max_int);
if (_nElements <= 0)
_msg = "Error al abrir fichero de datos: " + _path;
write(0, 0, 0, 0, _msg);
loop
// abortamos ejecución
if (key(_q) || key(_esc))
let_me_alone();
break;
end
frame;
end
end
_data = malloc(_nElements);
readCSVToIntArray(_path, _data, _nElements);
return(_data);
end