romaji["プ"] = "pu";
romaji["ペ"] = "pe";
romaji["ポ"] = "po";
- romaji["ã\83¼"] = "";
+ romaji["ã\83»"] = " ";
// -- double width letters ------
romaji["A"] = "A";
romaji["$"] = "$";
romaji["%"] = "%";
romaji["&"] = "&";
- romaji["'"] = "'"; // TODO:
- romaji["("] = "(";
- romaji[")"] = ")";
+ romaji["'"] = "'";
romaji["*"] = "*";
romaji["+"] = "+";
romaji[","] = ",";
romaji["?"] = "?";
romaji["@"] = "@";
+ romaji["〔"] = "(";
+ romaji["〕"] = ")";
+ romaji["("] = "(";
+ romaji[")"] = ")";
romaji["["] = "[";
- romaji["\"] = "\\";
romaji["]"] = "]";
+ romaji["【"] = "[";
+ romaji["】"] = "]";
+ romaji["{"] = "{";
+ romaji["}"] = "}";
+ romaji["\"] = "\\";
romaji["^"] = "^";
romaji["_"] = "_";
romaji["`"] = "`";
-
- romaji["{"] = "{";
romaji["|"] = "|";
- romaji["}"] = "}";
romaji["~"] = "~";
-
-
- // don't know where those belong to
+ romaji["ー"] = "";
+ romaji["。"] = ".";
+ romaji["、"] = ",";
romaji["〜"] = "~";
- romaji["、"] = ","; // TODO:
romaji["−"] = "-";
-
- romaji[" "] = " ";
romaji["―"] = "-";
- romaji["・"] = "-"; // FIXME
+ romaji[" "] = " ";
+
+ romaji["ー"] = "\3";
}
void remove_quote_1(
if (trans == romaji.end()) {
rom += ch;
if (ch.size() > 1)
- cout << "Don't know how to translate '" << ch << "' in '" << kana << "' to romaji." << endl;
+ cout << "Don't know how to translate '" << ch << "' in '" << kana << "' to romaji.\n";
}
else
rom += trans->second;
string const pred = rom.substr(pos - 3, 3);
if(pred == "chi" ||
pred == "shi" ||
- pred == "dzi"
+ pred == "dzi" ||
+ pred == "tsu" ||
+ pred == "shi"
) {
remove_quote_1(pos, rom);
+ pos -= 2;
continue;
}
}
- else if (pos > 1) {
+ if (pos > 1) {
string const pred = rom.substr(pos - 2, 2);
if(pred == "ki" ||
// shorten "ji\1y" to "j"
// otherwise remove "\1" and the preceding character
// but not the y
- rom.erase(
- pos - 1,
- (pos + 1 < rom.size()
- && rom[pos + 1] == 'y'
- && pred[0] == 'j')
- ? 3
- : 2);
-
- continue;
+ rom.erase(
+ pos - 1,
+ (pos + 1 < rom.size()
+ && rom[pos + 1] == 'y'
+ && pred[0] == 'j')
+ ? 3
+ : 2);
+ pos -= 2;
+ continue;
}
-
- if(pred == "fu" ||
+ else if(
+ pred == "fu" ||
pred == "de" ||
pred == "te" ||
pred == "vu")
{
remove_quote_1(pos, rom);
- continue;
+ pos -= 2;
+ continue;
+ }
+ else if(
+ pred == "su" ||
+ pred == "zu" ||
+ pred == "te" ||
+ pred == "de" ||
+ pred == "ku" ||
+ pred == "gu" ||
+ pred == "mu"
+ )
+ {
+ rom[pos - 1] = 'w';
+ rom.erase(pos, 1);
+ --pos;
+ continue;
+ }
+ else if(
+ pred == "to" ||
+ pred == "do"
+ )
+ {
+ rom[pos - 1] = 'h';
+ rom.erase(pos, 1);
+ --pos;
+ continue;
}
}
+
+ if (pos > 0) {
+ char const pred = rom[pos - 1];
+
+ switch(pred)
+ {
+ case 'a':
+ case 'i':
+ case 'u':
+ case 'e':
+ case 'o':
+ rom.erase(pos, 1);
+ --pos;
+ break;
+ default:
+ cout << "Encountered a special character in " << kana << " but don't know what to do with it.\n";
+ }
+ }
else
{
- cout << "Encountered a placeholder at the beginning. Just removing it." << endl;
- rom.erase(pos);
+ rom.erase(pos, 1);
+ --pos;
}
}
- // FIXME!
- else if (rom[pos] == '\2' && pos + 1 < rom.size())
+ else if (rom[pos] == '\2')
+ {
+ // two tsu may follow each other, so just remove them
+ if(pos + 1 < rom.size() && rom[pos + 1] != '\2')
rom[pos] = rom[pos + 1];
+ else
+ {
+ rom.erase(pos, 1);
+ --pos;
+ }
+ }
+ else if (rom[pos] == '\3')
+ {
+ if(pos == 0)
+ {
+ if(rom.size() == 1)
+ rom = "chouon";
+ else
+ {
+ cout << "ー is the first letter of " << kana << ". Don't know how to translate this.\n";
+ rom.erase(pos, 1);
+ --pos;
+ }
+ }
+ else
+ rom[pos] = rom[pos-1];
+ }
+
+ for (string::size_type pos = 0; pos < rom.size(); ++pos)
+ switch(rom[pos])
+ {
+ case '\1':
+ case '\2':
+ case '\3':
+ cout << "Failed to translate " << kana << '\n';
+ return;
+ }
}