<?php
parse_tab(<<<EOF
Bases: unicode (hex) sjis jis0208
HIRAGANA 12353 0x3041 0x829F 0x2421 (note jis0208 only uses 0x21..0x7E)
KATAKANA 12449 0x30A1 0x8340 0x2521
Base index and increment (lines beginning with space are comments)
000+2: a i u e o
001+2: A I U E O
Input map. 0 represents a vowel (AIUEO), X its capital
replacement, x its lowercase replacement.
= 0: X
= WH0: U x
010+2: KA KI KU KE KO
= KY0: KI yx
= K0: KX
011+2: GA GI GU GE GO
= GY0: GI yx
= G0: GX
020+2: SA SI SU SE SO
= SHI: SI
= SH0: SI yx
= SY0: SI yx
= SW0: SU x
= CI: SI
= S0: SX
021+2: ZA ZI ZU ZE ZO
= ZY0: ZI yx
= JI: ZI
= JY0: ZI yx
= J0: ZI yx
= Z0: ZX
= ZH0: ZI yx
030+2: TA TI
035+2: TU TE TO
= CHI: TI
= CH0: TI yx
= TY0: TI yx
= CY0: TI yx
= TSU: TU
= TS0: TU yx
= TH0: TE yx
= TW0: TO x
= T0: TX
031+2: DA DI
036+2: DU DE DO
= DY0: DI yx
= DH0: DE yx
= DW0: DO x
= DZI: DI
= D0: DX
034: tu
041+1: NA NI NU NE NO
= N0: NX
= NY0: NI yx
046+3: HA HI HU HE HO
= HY0: HI yx
= H0: HX
= FY0: HU yx
= FU: HU
= FW0: HU x
= F0: HU x
047+3: BA BI BU BE BO
= BY0: BI yx
= B0: BX
note these use katakana! vu = U with "
= VY0: vu yx
= VU: vu
= V0: vu x
048+3: PA PI PU PE PO
= PY0: PI yx
= P0: PX
061+1: MA MI MU ME MO
= MY0: MI yx
= M0: MX
066+2: ya yu yo
067+2: YA YU YO
= YA: YA
= YI: I
= YU: YU
= YE: I e
= YO: YO
072+1: RA RI RU RE RO
= RY0: RI yx
= LY0: RI yx
= R0: RX
= L0: RX
077+1: wa WA WI WE WO
(What is small wa? I don't know)
= WA: WA
= WI: U i
= WYI: WI
= WYE: WE
= WE: U e
= WO: WO
082: N
= N: N
179: vu ka ke
(katakana)
187: -
(katakana prolonged voice mark)
EOF
);
function parse_tab($str)
{
global $trtab; $trtab = array();
global $uniHtab; $uniHtab = array(); $uniH = 0;
global $uniKtab; $uniKtab = array(); $uniK = 0;
global $sjisHtab; $sjisHtab = array(); $sjisH = 0;
global $sjisKtab; $sjisKtab = array(); $sjisK = 0;
global $jisHtab; $jisHtab = array(); $jisH = $jisH1 = 0;
global $jisKtab; $jisKtab = array(); $jisK = $jisK1 = 0;
foreach(explode("\n", $str) as $s)
{
if($s=="" || $s[0]==' ')continue;
if(preg_match('/^HIRAGANA/i', $s))
{
preg_match('/^[A-Z]* *[0-9]* *([x0-9A-F]*) *([x0-9A-F]*) *([x0-9A-F]*)/i', $s, $ma);
$uniH = hexdec($ma[1]);
$sjisH= hexdec($ma[2]);
$jisH = hexdec($ma[3]);
$jisH1 = (($jisH % 256) - 0x21) + ((int)($jisH/256) - 0x21) * 0x5E;
}
elseif(preg_match('/^KATAKANA/i', $s))
{
preg_match('/^[A-Z]* *[0-9]* *([x0-9A-F]*) *([x0-9A-F]*) *([x0-9A-F]*)/i', $s, $ma);
$uniK = hexdec($ma[1]);
$sjisK= hexdec($ma[2]);
$jisK = hexdec($ma[3]);
$jisK1 = (($jisK % 256) - 0x21) + ((int)($jisK/256) - 0x21) * 0x5E;
}
elseif(preg_match('/^[0-9]/', $s))
{
$tok = sscanf($s, '%d+%d');
$n = (int)$tok[0];
$mul = (int)$tok[1];
$line = preg_replace('/^[^:]*: */', '', $s);
foreach(explode(" ", $line) as $tok)
{
$uniHtab[$tok] = $uniH + $n;
$uniKtab[$tok] = $uniK + $n;
$sjisHtab[$tok]= $sjisH + $n;
$sjisKtab[$tok]= $sjisK + $n;
$jisHtab[$tok] = $jisH1+$n;
$jisKtab[$tok] = $jisK1+$n;
$n += $mul;
}
}
elseif(preg_match('/[^ ]* .*:/', $s))
{
preg_match('/([^ ]*) *([^:]*): *(.*)/', $s, $ma);
$pat = $ma[1]; /* Currently contains '=' */
$key = strtolower($ma[2]);
$res = preg_replace('/ *\(.*/', '', $ma[3]);
if(!preg_match('/0/', $key))
$trtab[$key] = $res;
else foreach(array('a','i','u','e','o')as $v)
{
$k = str_replace('0', $v, $key);
if(!isset($trtab[$k])) $trtab[$k] = str_replace('X', strtoupper($v), str_replace('x', $v, $res));
}
}
else
{
echo '<p>Unrecognized line in configuration: "', htmlspecialchars($s), '"<p>';
}
}
}
function unihtmldisp($c) { echo '&#', $c, ';'; }
function sjisdisp($c) { echo chr($c/256), chr($c%256); }
function eucjpdisp($c) { echo chr(0xA1 + (int)(($c) / 0x5E)), chr(0xA1 + ($c) % 0x5E); }
function iso2022begin() { echo '$B'; }
function iso2022disp($c) { echo chr(0x21 + (int)(($c) / 0x5E)), chr(0x21 + ($c) % 0x5E); }
function iso2022end() { echo '(B'; }
function disptavut($tab, $iskatakana=false)
{
global $dump;
global $uniHtab, $uniKtab,
$sjisHtab, $sjisKtab,
$jisHtab, $jisKtab;
if($dump == 'iso-2022-jp')iso2022begin();
foreach($tab as $s)
{
if($s == 'yi')$s = 'i';
elseif($s == 'ye')$s = 'e';
elseif(!$uniHtab[$s])
{
if($dump == 'iso-2022-jp')iso2022end();
echo '<p>Unknown syllable: "', htmlspecialchars($s), '"<p>';
if($dump == 'iso-2022-jp')iso2022start();
}
if(strlen($s))
switch($dump)
{
case 'shift_jis': sjisdisp ($iskatakana ? $sjisKtab[$s]:$sjisHtab[$s]); break;
case 'shodouka':
case 'euc-jp': eucjpdisp ($iskatakana ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'iso-2022-jp': iso2022disp($iskatakana ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'text': print $s; break;
default: unihtmldisp($iskatakana ? $uniKtab[$s] : $uniHtab[$s]); break;
}
}
if($dump == 'iso-2022-jp')iso2022end();
}
function tavuta($s, $iskatakana=false)
{
global $trtab, $spacet;
$s = strtolower($s);
$s = str_replace('ô', 'ou', $s);
$s = str_replace('â', 'aa', $s);
$s = str_replace('ê', 'ee', $s);
$s = str_replace('î', 'ii', $s);
$s = str_replace('û', 'uu', $s);
/* The exceptional particles */
$s = str_replace(' o ', ' wo ', $s);
$s = str_replace(' wa ', ' ha ', $s); #$s = str_replace(' wa', ' ha', $s);
$res = '';
$b = strlen($s);
for($a=0; $a<$b; )
{
/**/if(isset($trtab[substr($s, $a, 4)])) { $res .= $trtab[substr($s, $a, 4)].' '; $a += 4; continue; }
elseif(isset($trtab[substr($s, $a, 3)])) { $res .= $trtab[substr($s, $a, 3)].' '; $a += 3; continue; }
elseif(isset($trtab[substr($s, $a, 2)])) { $res .= $trtab[substr($s, $a, 2)].' '; $a += 2; continue; }
elseif(isset($trtab[substr($s, $a, 1)])) { $res .= $trtab[substr($s, $a, 1)].' '; $a += 1; continue; }
else
{
if($s[$a] == ' ' && strlen($res))
{
if(strlen($spacet))
{
if(strlen($res)) { disptavut(explode(' ', $res), $iskatakana); $res = ''; }
echo '<span style="color:#40D060">', htmlspecialchars($spacet), '</span>';
}
$a++;
continue;
}
$tmp = substr(
str_replace('sh', 's',
str_replace('ch', 't',
str_replace('cch', 'tch',
str_replace('ts', 't',
substr($s, $a, 3))))), 0, 2);
if($tmp[0] == $tmp[1])
{
$res .= 'tu ';
$a++;
continue;
}
if(strlen($res)) { disptavut(explode(' ', $res), $iskatakana); $res = ''; }
$res = '';
echo '<em><span style="color:red">', htmlspecialchars($s[$a]), '</span></em>';
$a++;
}
}
if(strlen($res))disptavut(explode(' ', $res), $iskatakana);
return explode(' ', $res);
}
$dump = $_GET['dump'];
$src = $_GET['src'];
$s = $_GET['s'];
/* MAIN START */
if(strlen($dump) && $dump != 'text' && $dump != 'shodouka')
header('Content-type: text/html; charset='.$dump);
if($dump == 'shodouka')
{
include '/WWW/shodouka.php';
}
if($src >= 1 && $src <= 7)
{
echo '<html><head><title>Hiragana and katakana tables</title>',
'</head>',
'<body>';
echo "<small>Character encoding: ", strlen($dump) ? $dump : "default";
echo "<br>Possible choices, depending on what your browser supports: ";
foreach(array('unihtml'=>'html &#codes;',
'iso-2022-jp'=>'iso-2022-jp',
'euc-jp'=>'euc-JP',
'shift_jis'=>'shift_jis',
'text'=>'Syllable codes as ascii',
'shodouka'=>'Shodouka') as $tmp=>$tmp2)
echo '<a href="hiragana.php?src=', $src, '&dump=', $tmp, '">', $tmp2, '</a> ';
echo '</small>';
}
switch($src)
{
case 1:
print '<table width="100%" border=1 cellspacing=0 cellpadding=0><tr><td bgcolor=white>';
highlight_file('hiragana.php');
print '</td></tr></table>';
exit;
case 2:
case 3:
case 4:
case 5:
case 6:
{
print '<table cellpadding=1 cellspacing=0>';
$order = " in alphabetical order";
if($src==2)
{
/* Resort by length */
$trtab2 = array();
foreach($trtab as $s => $s2)
{
if($s != 'n')while(strlen($s) < 3)$s = " $s";
$trtab2[$s] = $s2;
}
}
elseif($src==3)
{
/* Not altered */
$trtab2 = $trtab;
}
elseif($src==4)
{
foreach($trtab as $s => $s2)
if(preg_match('/ /', $s2) || preg_match('/^[lc][aiueo]/', $s))/*
if(preg_match('/^(.[wy]|[tdw]h|[cs]h[aueo]|(ts|f)[aieo]|[jlv]|[yw][ei])/', $s))*/
unset($trtab[$s]);
$trtab2 = $trtab;
}
elseif($src==5 || $src==6)
{
$trtab2 = array();
$c=100;
$k = $src==5
? array('','k','g','s','z','t','d','h','b','p','m','n','r','y')
: array('','k','g','s','z','m','t','d','n','h','b','p','r','w','y');
foreach($k as $k)
foreach(array('a','i','u','e','o') as $v)
{
$s3 = str_replace(array('hu','tu','si','zi','ti'),
array('fu','tsu','shi','ji','chi'), $k.$v);
if(preg_match('/ye|yi|we|wi|wu/', $s3))continue;
$trtab2[str_pad('',$c--) . $s3] = $trtab[$s3];
}
foreach(array('n','wa','wo','wyi','wye') as $k) $trtab2[$k] = $trtab[$k];
$order = "";
}
ksort($trtab2);
for($iskatakana=0; $iskatakana<=1; $iskatakana++)
{
echo "<tr><td colspan=10><h1>Romanized ",
($iskatakana ? "katakana" : "hiragana"),
" syllables$order.",
"</h1>";
if($src==5)
{
echo '<small>';
if($iskatakana)
print '(Used in writing foreign words (English) or emphasis)';
else
print '(Used in writing Japanese words)';
echo '</small>';
}
if($src==3)
{
print '(Note that this is not a pronounciation guide for English speakers. English writing/pronounciation system is way too messed up.<br>See <a href="hiragana.php?src=7">table 7</a>, it explains better.)<p>';
}
echo "</td></tr>";
/* Taulukon tulostus */
$c = 0;
$rivinro = 0;
foreach($trtab2 as $tavu=>$tavut)
{
$style = 'font-size:130%'
.';padding-left:7px'
.';padding-right:7px';
if($src == 5 || $src==6)
{
$style .= ';border-top:1px dashed #002070';
if($src == 6)
{
if(!($c%5))$style .= ';border-left:1px dashed #002070';
if($c == ($rivinro==4 ? 12 : 14))$style .= ';border-right:1px dashed #002070';
if($rivinro > 2)if($c >= ($rivinro==3 ? 13 : 0))
$style .= ';border-bottom:1px dashed #002070';
}
else
{
if(!($c%5))$style .= ';border-left:1px dashed #002070';
if(($rivinro==4 && $c>=10)
|| ($rivinro==5 && $c>=8)
|| ($rivinro==6 && $c>=5)
|| $rivinro==7)$style .= ';border-bottom:1px dashed #002070';
if($c==($rivinro==6 ? 7 : ($rivinro==0||$rivinro==7 ? 4 : ($rivinro==4 ? 14 : 9))))
$style .= ';border-right:1px dashed #002070';
}
}
if(!$c)print '<tr>';
echo '<td style="', htmlspecialchars($style), '">';
if($src != 5)
echo $tavu, '<br>';
else
echo '<small>', trim($tavu), '</small><br>';
if($dump == 'iso-2022-jp')iso2022begin();
foreach(explode(' ', $tavut) as $s)
{
if($s == 'yi')$s = 'i';
elseif($s == 'ye')$s = 'e';
$kk = $iskatakana;
if($s[0]=='v')$kk=0;
switch($dump)
{
case 'shift_jis': sjisdisp ($kk ? $sjisKtab[$s]:$sjisHtab[$s]); break;
case 'shodouka':
case 'euc-jp': eucjpdisp ($kk ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'iso-2022-jp': iso2022disp($kk ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'text': print $s; break;
default: unihtmldisp($kk ? $uniKtab[$s] : $uniHtab[$s]); break;
}
}
if($dump == 'iso-2022-jp')iso2022end();
print '</td>';
$c++;
if($src == 5)
{
$max = 10;
if($rivinro == 0)$max = 5;
elseif($rivinro == 4)$max = 15;
elseif($rivinro == 6)$max = 8;
}
elseif($src == 6)
$max = 15;
else
$max = 10;
if($c == $max){print '</tr>';$c=0;$rivinro++;}
}
print '</tr>';
}
print '</table>';
print '<p>No warranty. Data may be incorrect. If you detect errors, you are encouraged to fix them. See the <a href="hiragana.php?src=1">source code</a>.';
print '<p>Written by <a href="mailto:bisqwit@iki.fi">Bisqwit</a> (<a href="https://iki.fi/bisqwit/">https://iki.fi/bisqwit/</a>).';
echo '</body></html>';
if($dump == 'shodouka')ShodoukaEnd();
exit;
}
case 7:
{
function tab7val($s)
{
return strpos
('A I U E O N a i u e o
KA GA KI GI KU GU KE GE KO GO
SA ZA SI ZI SU ZU SE ZE SO ZO
TA DA TI DI TU DU TE DE TO DO
NA NI NU NE NO
HA BA PA HI BI PI HU BU PU HE BE PE HO BO PO
MA MI MU ME MO
RA RI RU RE RO
WA WI WU WE WO
YA YI YU YE YO
ya yi yu ye yo vu
', $s);
}
function tab7sorter($a, $b)
{
$aa = explode(' ', $a);
$bb = explode(' ', $b);
$m = count($aa); if(count($bb) > $m)$m = count($bb);
for($c=0; $c<$m; $c++)
{
$av = tab7val($aa[$c]);
$bv = tab7val($bb[$c]);
if($av < $bv)return -1;
if($av > $bv)return 1;
}
return 0;
}
for($iskatakana=0; $iskatakana<=1; $iskatakana++)
{
$trtab2 = Array();
foreach($trtab as $r=>$k)
{
if(!$iskatakana)
{
if(false
|| preg_match('/^.w/', $r)
|| preg_match('/^c[^h]/', $r)
|| preg_match('/^[^sc]h/', $r)
|| preg_match('/^ts[^u]/', $r)
|| preg_match('/^dy[ou]/', $r)
|| $r == 'yi'
|| preg_match('/ y[ei]/', $k)
|| preg_match('/ [aiueo]/', $k)
|| preg_match('/^v/', $r)
) continue;
}
$trtab2[$k][] = $r;
}
$trtab2['DI'][] = 'ji';
$trtab2['DU'][] = 'zu';
uksort($trtab2, tab7sorter);
print '<h1>';
print $iskatakana ? 'Katakana (used in foreign words)'
: 'Hiragana (used in native words)';
print ' - different ways to romanize.</h1>';
print '<small>Note that this is not a pronounciation guide for English speakers.</small>';
$rivit = Array();
foreach($trtab2 as $tavut => $r)
{
ob_start();
print '<td align=left>';
if($dump == 'iso-2022-jp')iso2022begin();
foreach(explode(' ', $tavut) as $s)
{
if($s == 'yi')$s = 'i';
elseif($s == 'ye')$s = 'e';
$kk = $iskatakana;
if($s[0]=='v')$kk=0;
switch($dump)
{
case 'shift_jis': sjisdisp ($kk ? $sjisKtab[$s]:$sjisHtab[$s]); break;
case 'shodouka':
case 'euc-jp': eucjpdisp ($kk ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'iso-2022-jp': iso2022disp($kk ? $jisKtab[$s] : $jisHtab[$s]); break;
case 'text': print $s; break;
default: unihtmldisp($kk ? $uniKtab[$s] : $uniHtab[$s]); break;
}
}
if($dump == 'iso-2022-jp')iso2022end();
print '</td>';
sort($r);
if(preg_match('/^[ld]/', $r[0])
&& preg_match('/^[rzj]/', $r[1]) && count($r)==2)$r=Array($r[1],$r[0]);
for($c=0;$c<4;$c++)
echo '<td>', $r[$c], ' </td>';
$rivi = ob_get_contents();
ob_end_clean();
$rivit[] = $rivi;
}
$sarac = 6;
$add = count($rivit)/$sarac;
print '<table cellpadding=1 cellspacing=0 width="100%">';
$lev = 100/$sarac;
for($c=0; $c<=$add; $c++)
{
if(!$c)
{
print '<tr>';
for($s=0;$s<$sarac;$s++)echo '<td colspan=6 width="',$lev,'%"> </td>';
print '</tr>';
}
print '<tr>';
for($s=0; $s<$sarac; $s++)
{
$r = (int)($c + ($add)*$s);
$r2 = (int)( ($add)*($s+1));
if($c+1 <= $add || $r2 > $r)
print $rivit[$r];
else
print '<td></td><td></td><td></td><td></td><td></td>';
print '<td width=50> </td>';
}
print '</tr>';
}
print '</table>';
}
print '<p>No warranty. Data may be incorrect. If you detect errors, you are encouraged to fix them. See the <a href="hiragana.php?src=1">source code</a>.';
print '<p>Written by <a href="mailto:bisqwit@iki.fi">Bisqwit</a> (<a href="https://iki.fi/bisqwit/">https://iki.fi/bisqwit/</a>).';
print '<p>See also: <a href="confuse.php">confusing hiragana and katakana</a>';
echo '</body></html>';
if($dump == 'shodouka')ShodoukaEnd();
exit;
}
}
include '/WWW/headers.php';
headers('Romaji to hiragana converter');
print '<H1>Romaji to hiragana converter</H1>';
NavBar('');
echo
"Note: Type words 'wa' and 'o' as 'ha' and 'wo',
because that's how they are written in japanese.<br>",
'<form method=GET action="hiragana.php">',
'Romaji: <input type=text name=s value="', htmlspecialchars($s), '"><br>',
'Word separator: <input type=text size=6 name=spacet value="',htmlspecialchars($spacet), '"><br>',
'Encoding: <select name=dump>';
foreach(array('unihtml'=>'html &#codes;',
'iso-2022-jp'=>'iso-2022-jp',
'euc-jp'=>'euc-JP',
'shift_jis'=>'shift_jis',
'text'=>'Syllable codes as ascii',
'shodouka'=>'<a href="http://web.lfw.org/shodouka/">Shodouka</a>'
) as $tmp=>$tmp2)
{
echo '<option value="', $tmp, '"';
if($tmp == $dump)echo ' selected';
echo '>', $tmp2, '</option>';
}
echo '</select><br>',
'<input type=submit value="Convert">',
'</form>';
echo "<br>If your text is ENGLISH, try
<a href=\"https://bisqwit.iki.fi/source/deeng.html?engtext=".
htmlspecialchars(urlencode($s))."\">english to romaji converter</a>.<br>
If your text is otherwise NOT standard romaji (but Finnish for example),
use <a href=\"japthing.php?s=",htmlspecialchars(urlencode($s)),
"\">older converter</a> instead.<br>
Those two also offer katakana.<br>
If you want kanji!
Use <a href=\"https://bisqwit.iki.fi/japtools/\">the dictionary</a> instead.
<p>
";
if(strlen($s))
{
echo '<p style="font-size:130%">', htmlspecialchars($s), ": (up:hiragana; down:katakana)<br> <br>\n";
print '<span style="font-size:250%">';
$tmp = tavuta($s, false);
print '</span>';
echo '<hr>';
print '<span style="font-size:250%">';
$tmp = tavuta($s, true);
print '</span>';
echo '</p>';
ob_start();
foreach($tmp as $s)
{
if($s == 'yi') $s = 'i';
elseif($s == 'ye') $s = 'e';
if($s) eucjpdisp($jisHtab[$s]);
}
$s = @iconv('euc-jp', 'utf-8', ob_get_contents());
ob_end_clean();
$url = 'http://babelfish.altavista.com/babelfish/tr?lp=ja_en';
$url .= '&tt=urltext&doit=done&intl=1&urltext='.urlencode($s);
echo '<a href="', htmlspecialchars($url), '">',
'Translate using Babelfish',
'</a>';
}
?>
<p><small>Written by Bisqwit (<a href="https://iki.fi/bisqwit/">https://iki.fi/bisqwit/</a>)</small>
<p><b>
<a href="hiragana.php?src=2">Syllable list</a> is also available.
It lists all the katakana/hiragana syllables.</b><br>
It is available <a href="hiragana.php?src=3">sorted another way</a> too.
A <a href="hiragana.php?src=4">shorter list</a> is there too.
Some <a href="hiragana.php?src=5">other</a>
<a href="hiragana.php?src=6">versions</a> also...
</p>
<?
Epilogue();
?>
Disclaimer: The accuracy of the operation of this converter
is strictly limited to the progress of my Japanese studies.
<p>
<em>Errors, if you find, report to <a href="http://bisqwit.stc.cx/">me</a>
via <a href="mailto:bisqwit@iki.fi">email</a>.</em>
<p><a href="hiragana.php?src=1">Source code</a> is available (php).
It might interest you, as it is short (~340 lines (well, once was)),
but contains hiragana encoding in iso-2022-jp, euc-jp, shift_jis and
unihtml (without external programs), plus a quite elegant parsed table
for romaji to hiragana conversion. It could be easily extended
to support different romajizations, and has a theoretic support
for katakana.
<p>
<br><a href="http://bisqwit.stc.cx/%7Emog/hiragana.gif">These</a>
<a href="http://bisqwit.stc.cx/%7Emog/katakana.gif">images</a> could also be helpful.
<?
Footers();
if($dump == 'shodouka')ShodoukaEnd();
/* MAIN END */
|