ref: 74df3901c04f70c1ba89e71b78dee9dc2882e116
dir: /src/to-kana.c/
/*
*******************************************************************************
* Author: Ethan Long
* Licence: Public Domain
* Email: ethandavidlong@gmail.com, u7281759@anu.edu.au
* Description: to-kana is a program for converting streams of romaji to
* hiragana or katakana.
*/
#include <u.h>
#include <libc.h>
#include <stdio.h>
void eval(int, int);
void kanafill(Rune, char*, int, Rune*);
void kanalook(char*, Rune*, int);
void printhelp(void);
int strappend(char*, char);
void
main(int argc, char *argv[])
{
int fpin, fpout;
ARGBEGIN{
case 's':
fpin = fileno(stdin);
fpout = fileno(stdout);
eval(fpin, fpout);
break;
case 'h':
printhelp();
exits(0);
default:
fprint(2, "usage: %s [-s][-h]\n", argv0);
printhelp();
exits("usage");
}ARGEND;
exits(nil);
}
void
printhelp(void)
{
print("options:\n");
print(" [-s] - use stdin/out streams\n");
print(" [-h] - show this help\n");
return;
}
/*
* Runs through the input at the file pointer fpin until it reaches EOF
* outputs kana to file pointer fpout.
*/
void
eval(int fpin, int fpout)
{
int reading = 1;
int katakana = 0;
char charin;
char buf[10];
Rune kana[10];
while(reading){
if(read(fpin, &charin, 1)){
if(charin == '!')
katakana = (0 == katakana);
else if(strappend(buf, charin)){
kanalook(buf, kana, katakana);
fprint(fpout, "%S", kana);
for(int i=0; i<10; i++){
kana[i] = 0;
buf[i] = 0;
}
}
}
else{
reading = 0;
}
}
fprint(fpout, "%s", buf);
return;
}
/*
* Appends the provided character to the end of the provided string.
* If this results in a complete kana being formable, return 1.
*/
int
strappend(char* string, char in)
{
int end;
int i;
for(i=0; string[i] != 0; i++);
end = (in == 'a' || in == 'i' || in == 'u' || in == 'e' || in == 'o');
if (in == 'n'){
if (string[i-1] == 'n')
end = 1;
}
string[i] = in;
return end;
}
/*
* kanalook takes in a pointer to a string and a pointer to a string of Rune
* and converts the string in romaji to a string of runes of kana.
*/
void
kanalook(char* buf, Rune* str, int katakana)
{
int index;
char* end;
katakana = katakana * 96;
index = 0;
if(buf[0] == buf[1]){
str[index] = L'っ';
index ++;
}
end = buf + 1 + index;
switch(buf[0]){
/* あ family */
case 'a':
str[0] = (Rune)((int)(L'あ') + katakana);
break;
case 'i':
str[0] = (Rune)((int)(L'い') + katakana);
break;
case 'u':
str[0] = (Rune)((int)(L'う') + katakana);
break;
case 'e':
str[0] = (Rune)((int)(L'え') + katakana);
break;
case 'o':
str[0] = (Rune)((int)(L'お') + katakana);
break;
/* か family */
case 'k':
kanafill((Rune)((int)(L'か') + katakana), end, 0, str + index);
break;
case 'g':
kanafill((Rune)((int)(L'か') + katakana), end, 1, str + index);
break;
/* さ family */
case 's':
if(buf[index + 1] == 'h'){
end[0] = buf[index + 2];
end[1] = 0;
}
kanafill((Rune)((int)(L'さ') + katakana), end, 0, str + index);
break;
case 'z':
kanafill((Rune)((int)(L'さ') + katakana), end, 1, str + index);
break;
case 'j':
if(buf[index + 1] != 'y' && buf[index + 1] != 'i'){
end[0] = 'y';
end[1] = buf[index + 1];
kanafill((Rune)((int)(L'さ') + katakana), end, 1, str + index);
} else
kanafill((Rune)((int)(L'さ') + katakana), end, 1, str + index);
break;
/* た family */
case 't':
kanafill((Rune)((int)(L'た') + katakana), end, 0, str + index);
break;
case 'c':
if(buf[index + 1] != 'y' && buf[index + 1] != 'i'){
end[0] = 'y';
end[1] = buf[index + 1];
kanafill((Rune)((int)(L'た') + katakana), end, 0, str + index);
} else
kanafill((Rune)((int)(L'た') + katakana), end, 0, str + index);
break;
case 'd':
kanafill((Rune)((int)(L'た') + katakana), end, 1, str + index);
break;
/* な family (and ん) */
case 'n':
if(buf[1] == 'n')
str[0] = (Rune)((int)(L'ん') + katakana);
else
kanafill((Rune)((int)(L'な') + katakana), end, 0, str + index);
break;
/* は family */
case 'h':
kanafill((Rune)((int)(L'は') + katakana), end, 0, str + index);
break;
case 'f':
end[0] = 'f';
end[1] = buf[index + 1];
kanafill((Rune)((int)(L'は') + katakana), end, 0, str + index);
break;
case 'b':
kanafill((Rune)((int)(L'は') + katakana), end, 1, str + index);
break;
case 'p':
kanafill((Rune)((int)(L'は') + katakana), end, 2, str + index);
break;
/* ま family */
case 'm':
kanafill((Rune)((int)(L'ま') + katakana), end, 0, str + index);
break;
/* や family */
case 'y':
kanafill((Rune)((int)(L'や') + katakana), end, 0, str + index);
break;
/* ら family */
case 'r':
kanafill((Rune)((int)(L'ら') + katakana), end, 0, str + index);
break;
/* わ family */
case 'w':
kanafill((Rune)((int)(L'わ') + katakana), end, 0, str + index);
break;
default:
break;
}
return;
}
/*
* The kanafill function takes in the base kana of the family to be output,
* the character representation of the ending syllabaries, an integer
* representing the type of dakuten (1 for regular, 2 for maru), and a pointer
* to a rune array that is being edited.
*/
void
kanafill(Rune base, char* in, int dakuten, Rune* out)
{
switch(base){
/* Special Boys */
case L'や':
case L'ヤ':
switch(in[0]){
case 'a':
out[0] = base;
break;
case 'u':
out[0] = (Rune)((int)base + 2);
break;
case 'o':
out[0] = (Rune)((int)base + 4);
break;
default:
break;
}
break;
case L'わ':
case L'ワ':
switch(in[0]){
case 'a':
out[0] = base;
break;
case 'i':
out[0] = (Rune)((int)base + 1);
break;
case 'e':
out[0] = (Rune)((int)base + 2);
break;
case 'o':
out[0] = (Rune)((int)base + 3);
break;
default:
break;
}
break;
case L'ん':
case L'ン':
out[0] = L'ん';
break;
/* は family has marudakuten */
case L'は':
case L'ハ':
switch(in[0]){
case 'a':
out[0] = (Rune)((int)base + dakuten);
break;
case 'i':
out[0] = (Rune)((int)base + 3 + dakuten);
break;
case 'u':
out[0] = (Rune)((int)base + 6 + dakuten);
break;
case 'e':
out[0] = (Rune)((int)base + 9 + dakuten);
break;
case 'o':
out[0] = (Rune)((int)base + 12 + dakuten);
break;
case 'y':
out[0] = (Rune)((int)base + 3 + dakuten);
switch(in[1]){
case 'a':
out[1] = L'ゃ';
break;
case 'u':
out[1] = L'ゅ';
break;
case 'o':
out[1] = L'ょ';
break;
default:
break;
}
default:
break;
}
break;
/* families with no dakuten or small forms */
case L'ら':
case L'ラ':
case L'ま':
case L'マ':
case L'な':
case L'ナ':
switch(in[0]){
case 'a':
out[0] = base;
break;
case 'i':
out[0] = (Rune)((int)base + 1);
break;
case 'u':
out[0] = (Rune)((int)base + 2);
break;
case 'e':
out[0] = (Rune)((int)base + 3);
break;
case 'o':
out[0] = (Rune)((int)base + 4);
break;
case 'y':
out[0] = (Rune)((int)base + 1);
switch(in[1]){
case 'a':
out[1] = L'ゃ';
break;
case 'u':
out[1] = L'ゅ';
break;
case 'o':
out[1] = L'ょ';
break;
default:
break;
}
default:
break;
}
break;
/* families with just dakuten */
case L'か':
case L'カ':
case L'さ':
case L'サ':
switch(in[0]){
case 'a':
out[0] = (Rune)((int)base + dakuten);
break;
case 'i':
out[0] = (Rune)((int)base + 2 + dakuten);
break;
case 'u':
out[0] = (Rune)((int)base + 4 + dakuten);
break;
case 'e':
out[0] = (Rune)((int)base + 6 + dakuten);
break;
case 'o':
out[0] = (Rune)((int)base + 8 + dakuten);
break;
case 'y':
out[0] = (Rune)((int)base + 2 + dakuten);
switch(in[1]){
case 'a':
out[1] = L'ゃ';
break;
case 'u':
out[1] = L'ゅ';
break;
case 'o':
out[1] = L'ょ';
break;
default:
break;
}
default:
break;
}
break;
/* た family has small つ and dakuten */
case L'た':
case L'タ':
switch(in[0]){
case 'a':
out[0] = (Rune)((int)base + dakuten);
break;
case 'i':
out[0] = (Rune)((int)base + 2 + dakuten);
break;
case 'u':
out[0] = (Rune)((int)base + 5 + dakuten);
break;
case 'e':
out[0] = (Rune)((int)base + 7 + dakuten);
break;
case 'o':
out[0] = (Rune)((int)base + 9 + dakuten);
break;
case 'y':
out[0] = (Rune)((int)base + 2 + dakuten);
switch(in[1]){
case 'a':
out[1] = L'ゃ';
break;
case 'u':
out[1] = L'ゅ';
break;
case 'o':
out[1] = L'ょ';
break;
default:
break;
}
default:
break;
}
break;
default:
break;
}
return;
}