HebrewMCIM Class Reference

#include <hebrewmcim.h>

Inheritance diagram for HebrewMCIM:

Collaboration diagram for HebrewMCIM:

Public Member Functions
virtual void	clearState ()
virtual int	getState ()
	HebrewMCIM ()
int *	translate (char in)

Protected Member Functions
virtual void	setState (int state)

Private Member Functions
void	init ()

Private Attributes
map< int, int * >	multiChars
int	subst [255]
map< int, int >	subst2 [12]

Detailed Description

Title: Keyboard mapping for Michigan-Claremont Hebrew input Description: Copyright: Copyright (c) 2001 CrossWire Bible Society under the terms of the GNU GPL Company:

Author

Troy A. Griffitts

Version

1.0

Definition at line 42 of file hebrewmcim.h.

Constructor & Destructor Documentation

HebrewMCIM::HebrewMCIM

(

)

Definition at line 26 of file hebrewmcim.cpp.

        :SWInputMethod() {

   init();
}

Member Function Documentation

void SWInputMethod::clearState ( )

virtualinherited

Definition at line 37 of file swinputmeth.cpp.

                               {
   state = 0;
}

int SWInputMethod::getState ( )

virtualinherited

Definition at line 33 of file swinputmeth.cpp.

                            {
   return state;
}

void HebrewMCIM::init ( )

private

Definition at line 83 of file hebrewmcim.cpp.

                      {
    memset(subst, 0, 255);

    subst[')'] = 1488;
    subst['B'] = 1489;
    subst['G'] = 1490;
    subst['D'] = 1491;
    subst['H'] = 1492;
    subst['W'] = 1493;
    subst['Z'] = 1494;
    subst['X'] = 1495;
    subst['+'] = 1496;
    subst['Y'] = 1497;

    subst['k'] = 1498;  // finals
    subst['m'] = 1501;
    subst['n'] = 1503;
    subst['c'] = 1509;

    subst['P'] = 1508;
    subst['K'] = 1499;
    subst['L'] = 1500;
    subst['M'] = 1502;
    subst['N'] = 1504;
    subst['S'] = 1505;
    subst['('] = 1506;
    subst['p'] = 1507;
    subst['C'] = 1510;
    subst['Q'] = 1511;
    subst['R'] = 1512;
    subst['#'] = 1513;

    // special multiChars
    subst['&'] = 50;
    subst['$'] = 50;

    static int x[] = {1513, 1474};
    multiChars['&'] = x;
    static int y[] = {1513, 1473};
    multiChars['$'] = y;

    subst['T'] = 1514;

    // VOWELS
    subst['A'] = 1463;
    subst['F'] = 1464;
    subst['E'] = 1462;
    subst['"'] = 1461;
    subst['I'] = 1460;
    subst['O'] = 1465;
    subst['U'] = 1467;



    // OTHER DIACRITICS
    subst['.'] = 1468;
    subst['-'] = 1470;
    subst[','] = 1471;

    // Compound input

    // CANTILLATION

    subst[':'] = 2;
    subst2[2]['A'] = 1458;
    subst2[2]['E'] = 1457;
    subst2[2]['F'] = 1459;


    /* Telisha qetana is postpositive as in '04' above. However, Michigan
# code '24' is for a medial telisha. Graphically, there is no
# difference.
    */
    subst['2'] = 5;
    subst2[5]['4'] = 1449;


    /* Note Michigan encoding distinguishes between medial metheg '35' (occuring
# on the left of the vowel), and the ordinary meteg '95' (occuring on the
# right of the vowel). It is also used for silluq.
    */
    subst['3'] = 6;
    subst2[6]['3'] = 1433;
    subst2[6]['5'] = 1469;


    /* The Michigan code of telisha gedola in medial position. Graphically,
# there is no difference.
    */
    subst['4'] = 7;
    subst2[7]['4'] = 1440;

    subst['6'] = 8;
    subst2[8]['0'] = 1451;
    subst2[8]['1'] = 1436;

    subst['1'] = 4;
    subst2[4]['0'] = 1434;

    /* In the poetic books, prepositive dehi occurs; it's unclear whether
# tipeha also occurs in the poetic books. Otherwise, we could simply
# check for what book in the Tanach we are in. Michigan uses the same
# code for each.
    */

    subst2[4]['3'] = 1430;

    /* This is the poetic accent mugrash, which also includes rebia, but is
# encoded separately as '81' in the Michigan text.
    */
    subst2[4]['1'] = 1437;
    subst2[4]['4'] = 1440;


    subst['0'] = 3;
    subst2[3]['0'] = 1475;
    subst2[3]['1'] = 1426;

    /* According to BHS, zarqa and sinnor are both postpositive. However,
# the Michigan encoding uses one code for both. The Unicode zarqa
# (0x0598) is definitely NOT postpositive. And further, the shape of
# the symbol is different in BHS and Uniocde. This needs further
# research to determine what's going on here. For now, we follow BHS
# and use the postpositive Unicode zinor or both accents.
    */

    subst2[3]['2'] = 1454;

    /* Pashta is postpositive, and the Unicode equivalent reflects
# this. However, there is a poetic equivalent -- azla legarmeh --
# which is not postpositive, but no equivalent code point exists in
# Unicode. The Michigan encoding does not distinguish between the two,
# although it could be algorithmically determined.
    */

    subst2[3]['3'] = 1433;
    subst2[3]['4'] = 1449;
    subst2[3]['5'] = 1472;


    /* This is the Unicode Hebrew *accent*; there is also another Hebrew
# *punctuation* called GERSHAYIM 0x05F4. I'm using the more
# traditional rounded marks, rather than the alternate straight
# marks.
    */

    subst2[8]['2'] = 1438;

    // Also known as azla
    subst2[8]['3'] = 1448;
    subst2[8]['4'] = 1452;
    subst2[8]['5'] = 1427;


    subst['8'] = 9;
    subst2[9]['0'] = 1428;
    subst2[9]['1'] = 1431;

    /* Note, this accent is actually sinnorit, but it does not exist as a
# separate glyph in the Unicode standard. The 'ZINOR' Unicode accent
# is postpositive, while sinnorit is not. ZARQA is as close as I can
# get to this.
    */
    subst2[9]['2'] = 1432;

    /* The Unicode form does not match the form used by BHS, but the names
# are the same.
    */
    subst2[9]['3'] = 1441;
    subst2[9]['4'] = 1439;
    subst2[9]['5'] = 1429;

    subst['7'] = 10;
    subst2[10]['0'] = 1444;
    subst2[10]['1'] = 1445;
    subst2[10]['2'] = 1446;
    subst2[10]['3'] = 1430;  // also '13', '73' also is used for majela
    subst2[10]['4'] = 1443;
    subst2[10]['5'] = 1469;  // this is silluq; should appear to the left of the vowel

    subst['9'] = 11;
    subst2[11]['1'] = 1435;
    subst2[11]['2'] = 1425;
    subst2[11]['3'] = 1450;
    subst2[11]['4'] = 1447;
    subst2[11]['5'] = 1469;  // should appear to the right of the vowel

}

void SWInputMethod::setState ( int  state )

protectedvirtualinherited

Definition at line 29 of file swinputmeth.cpp.

                                      {
   this->state = state;
}

int * HebrewMCIM::translate ( char  in )

virtual

Implements SWInputMethod.

Definition at line 33 of file hebrewmcim.cpp.

                                  {
    int retVal = 0;
    static int retString[5];
    int retStringIndex = 0;

    memset(retString, 0, 5);

    if (getState() > 1) {
        if (getState() >= 12) { // serious issue with internal structure
            setState(0);
            retString[retStringIndex++] = in;
            return retString;
        }
        map<int, int>::iterator find = subst2[getState()].find(in);
        if (find != subst2[getState()].end())
            retVal = find->second;
        else retVal = in;

        setState(0);
        retString[retStringIndex++] = retVal;
        return retString;
    }
    else {
        retVal = subst[in];

        if (retVal == 0) {
            setState(0);
            retString[retStringIndex++] = in;
            return retString;
        }
        if (retVal > 100) {
            setState(1);
            retString[retStringIndex++] = retVal;
            return retString;
        }
        if (retVal == 50) {  // multiChar
            setState(1);
            int *chars = multiChars[in];
            if (chars != 0) {
                retString[retStringIndex++] = chars[0];
                retString[retStringIndex++] = chars[1];
                return retString;
            }
        }
    }
    setState(retVal);
    return 0;
}

Member Data Documentation

map<int, int*> HebrewMCIM::multiChars

private

Definition at line 48 of file hebrewmcim.h.

int HebrewMCIM::subst[255]

private

Definition at line 46 of file hebrewmcim.h.

map<int, int> HebrewMCIM::subst2[12]

private

Definition at line 47 of file hebrewmcim.h.

---

The documentation for this class was generated from the following files:

• include/hebrewmcim.h
• src/frontend/im/hebrewmcim.cpp