//--------------------------------------------------------------------
//  Copyright (C) 1999, International Business Machines
//  Corporation and others.  All Rights Reserved.
//--------------------------------------------------------------------
//  Date        Name        Description
//  11/17/99    aliu        Creation.
//--------------------------------------------------------------------

// Latin-Greek

lgreek {
    Rule {
                // ==============================================
                // Modern Greek Transliteration Rules
                //
                // This transliterates modern Greek characters, but using rules
                // that are traditional for Ancient Greek, and
                // thus more resemble Greek words that have become part
                // of English. It differs from the official Greek
                // transliteration, which is more phonetic (since
                // most modern Greek vowels, for example, have
                // degenerated simply to sound like "ee").
                //
                // There are only a few tricky parts.
                // 1. eta and omega don't map directly to Latin vowels,
                //    so we use a macron on e and o, and some
                //    other combinations if they are accented.
                // 2. The accented, diaeresis i and y are substituted too.
                // 3. Some letters use digraphs, like "ph". While typical,
                //    they need some special handling.
                // 4. A gamma before a gamma or a few other letters is
                //    transliterated as an "n", as in "Anglo"
                // 5. An ypsilon after a vowel is a "u", as in
                //    "Mouseio". Otherwise it is a "y" as in "Physikon"
                // 6. The construction of the rules is made simpler by making sure
                //    that most rules for lowercase letters exactly correspond to the
                //    rules for uppercase letters, *except* for the case of the letters
                //    in the rule itself. That way, after modifying the uppercase rules,
                //    you can just copy, paste, and "set to lowercase" to get
                //    the rules for lowercase letters!
                // ==============================================
            
                // ==============================================
                // Variables, used to make the rules more comprehensible
                // and for conditionals.
                // ==============================================
                
                // Latin Letters
                
                "E-MACRON=\u0112;"
                "e-macron=\u0113;"
                "O-MACRON=\u014C;"
                "o-macron=\u014D;"
                "Y-UMLAUT=\u0178;"
                "y-umlaut=\u00FF;"
                
                //! // with real accents.
                //! + "E-MACRON-ACUTE=\u0112\u0301;"
                //! + "e-macron-acute=\u0113\u0301;"
                //! + "O-MACRON-ACUTE=\u014C\u0301;"
                //! + "o-macron-acute=\u014D\u0301;"
                //! + "y-umlaut-acute=\u00FF\u0301;"
                //! + "\u00ef-acute=\u00ef\u0301;"
                //! + "\u00fc-acute=\u00fc\u0301;"
                //! //
 
                // single letter equivalents
                
                "E-MACRON-ACUTE=\u00CA;"
                "e-macron-acute=\u00EA;"
                "O-MACRON-ACUTE=\u00D4;"
                "o-macron-acute=\u00F4;"
                "y-umlaut-acute=\u0177;"
                "\u00ef-acute=\u00EE;"
                "\u00fc-acute=\u00FB;"       
                
                // Greek Letters

                "ALPHA=\u0391;"
                "BETA=\u0392;"
                "GAMMA=\u0393;"
                "DELTA=\u0394;"
                "EPSILON=\u0395;"
                "ZETA=\u0396;"
                "ETA=\u0397;"
                "THETA=\u0398;"
                "IOTA=\u0399;"
                "KAPPA=\u039A;"
                "LAMBDA=\u039B;"
                "MU=\u039C;"
                "NU=\u039D;"
                "XI=\u039E;"
                "OMICRON=\u039F;"
                "PI=\u03A0;"
                "RHO=\u03A1;"
                "SIGMA=\u03A3;"
                "TAU=\u03A4;"
                "YPSILON=\u03A5;"
                "PHI=\u03A6;"
                "CHI=\u03A7;"
                "PSI=\u03A8;"
                "OMEGA=\u03A9;"

                "ALPHA+=\u0386;"
                "EPSILON+=\u0388;"
                "ETA+=\u0389;"
                "IOTA+=\u038A;"
                "OMICRON+=\u038C;"
                "YPSILON+=\u038E;"
                "OMEGA+=\u038F;"
                "IOTA_DIAERESIS=\u03AA;"
                "YPSILON_DIAERESIS=\u03AB;"

                "alpha=\u03B1;"
                "beta=\u03B2;"
                "gamma=\u03B3;"
                "delta=\u03B4;"
                "epsilon=\u03B5;"
                "zeta=\u03B6;"
                "eta=\u03B7;"
                "theta=\u03B8;"
                "iota=\u03B9;"
                "kappa=\u03BA;"
                "lambda=\u03BB;"
                "mu=\u03BC;"
                "nu=\u03BD;"
                "xi=\u03BE;"
                "omicron=\u03BF;"
                "pi=\u03C0;"
                "rho=\u03C1;"
                "sigma=\u03C3;"
                "tau=\u03C4;"
                "ypsilon=\u03C5;"
                "phi=\u03C6;"
                "chi=\u03C7;"
                "psi=\u03C8;"
                "omega=\u03C9;"

                //forms

                "alpha+=\u03AC;"
                "epsilon+=\u03AD;"
                "eta+=\u03AE;"
                "iota+=\u03AF;"
                "omicron+=\u03CC;"
                "ypsilon+=\u03CD;"
                "omega+=\u03CE;"
                "iota_diaeresis=\u03CA;"
                "ypsilon_diaeresis=\u03CB;"
                "iota_diaeresis+=\u0390;"
                "ypsilon_diaeresis+=\u03B0;"
                "sigma+=\u03C2;"

                // Variables for conditional mappings
                
                // Use lowercase for all variable names, to allow cut/paste below.

                "letter=[~[:Lu:][:Ll:]];"
                "lower=[[:Ll:]];"
                "softener=[eiyEIY];"
                "vowel=[aeiouAEIOU"
                  "{ALPHA}{EPSILON}{ETA}{IOTA}{OMICRON}{YPSILON}{OMEGA}"
                  "{ALPHA+}{EPSILON+}{ETA+}{IOTA+}{OMICRON+}{YPSILON+}{OMEGA+}"
                  "{IOTA_DIAERESIS}{YPSILON_DIAERESIS}"
                  "{alpha}{epsilon}{eta}{iota}{omicron}{ypsilon}{omega}"
                  "{alpha+}{epsilon+}{eta+}{iota+}{omicron+}{ypsilon+}{omega+}"
                  "{iota_diaeresis}{ypsilon_diaeresis}"
                  "{iota_diaeresis+}{ypsilon_diaeresis+}"
                  "];"
                "n-gamma=[GKXCgkxc];"
                "gamma-n=[{GAMMA}{KAPPA}{CHI}{XI}{gamma}{kappa}{chi}{xi}];"
                "pp=[Pp];"

                // ==============================================
                // Rules
                // ==============================================
                // The following are special titlecases, and should
                // not be copied when duplicating the lowercase
                // ==============================================
                
                "Th <> {THETA}({lower};"
                "Ph <> {PHI}({lower};"
                "Ch <> {CHI}({lower};"
              //masked: + "Ps<{PHI}({lower};"
                
                // Because there is no uppercase forms for final sigma,
                // we had to move all the sigma rules up here.
                
                // Remember to insert ' to preserve round trip, for double letters
                // don't need to do this for the digraphs with h,
                // since it is not created when mapping back from greek
                
                // use special form for s
                
                "''S <> ({pp}) {SIGMA} ;" // handle PS
                "S <> {SIGMA};"
                
                // The following are a bit tricky. 's' takes two forms in greek
                // final or non final. 
                // We use ~s to represent the abnormal form: final before letter
                // or non-final before non-letter.
                // We use 's to separate p and s (otherwise ps is one letter)
                // so, we break out the following forms:
                
                "''s < ({pp}) {sigma} ({letter});"
                "s <          {sigma} ({letter});"
                "~s <         {sigma} ;"

                "~s <         {sigma+} ({letter});"
                "''s < ({pp}) {sigma+} ;"
                "s <          {sigma+} ;"

                "~s ({letter})  > {sigma+};"
                "~s             > {sigma};"
                "''s ({letter}) > {sigma};"
                "''s            > {sigma+};"
                "s ({letter})   > {sigma};"
                "s              > {sigma+};"
                
                // because there are no uppercase forms, had to move these up too.
                
                "i\"`>{iota_diaeresis+};"
                "y\"`>{ypsilon_diaeresis+};"
                
                "{\u00ef-acute} <> {iota_diaeresis+};"
                "{\u00fc-acute} <> {vowel}){ypsilon_diaeresis+};"
                "{y-umlaut-acute} <> {ypsilon_diaeresis+};"
                                
                // ==============================================
                // Uppercase Forms.
                // To make lowercase forms, just copy and lowercase below
                // ==============================================
 
                // Typing variants, in case the keyboard doesn't have accents
                
                "A`>{ALPHA+};"
                "E`>{EPSILON+};"
                "EE`>{ETA+};"
                "EE>{ETA};" 
                "I`>{IOTA+};"
                "O`>{OMICRON+};"
                "OO`>{OMEGA+};"
                "OO>{OMEGA};"
                "I\">{IOTA_DIAERESIS};"
                "Y\">{YPSILON_DIAERESIS};"
                
                // Basic Letters
                
                "A<>{ALPHA};"
                "\u00c1<>{ALPHA+};"
                "B<>{BETA};"
                "N ({n-gamma}) <> {GAMMA} ({gamma-n});"
                "G<>{GAMMA};"
                "D<>{DELTA};"
                "''E <> ([Ee]){EPSILON};" // handle EE
                "E<>{EPSILON};"
                "\u00c9<>{EPSILON+};"
                "Z<>{ZETA};"
                "{E-MACRON-ACUTE}<>{ETA+};"
                "{E-MACRON}<>{ETA};"
                "TH<>{THETA};"
                "I<>{IOTA};"
                "\u00cd<>{IOTA+};"
                "\u00cf<>{IOTA_DIAERESIS};"
                "K<>{KAPPA};"
                "L<>{LAMBDA};"
                "M<>{MU};"
                "N'' <> {NU} ({gamma-n});"
                "N<>{NU};"
                "X<>{XI};"
                "''O <> ([Oo]) {OMICRON};" // handle OO
                "O<>{OMICRON};"
                "\u00d3<>{OMICRON+};"
                "PH<>{PHI};" // needs ordering before P
                "PS<>{PSI};" // needs ordering before P
                "P<>{PI};"
                "R<>{RHO};"
                "T<>{TAU};"
                "U <> ({vowel}) {YPSILON};"
                "\u00da <> ({vowel}) {YPSILON+};"
                "\u00dc <> ({vowel}) {YPSILON_DIAERESIS};"
                "Y<>{YPSILON};"
                "\u00dd<>{YPSILON+};"
                "{Y-UMLAUT}<>{YPSILON_DIAERESIS};"
                "CH<>{CHI};"
                "{O-MACRON-ACUTE}<>{OMEGA+};"
                "{O-MACRON}<>{OMEGA};"

                // Extra English Letters. Mapped for completeness
                
                "C({softener})>|S;"
                "C>|K;"
                "F>|PH;"
                "H>|CH;"
                "J>|I;"
                "Q>|K;"
                "V>|U;"
                "W>|U;"
                
                // ==============================================
                // Lowercase Forms. Just copy above and lowercase
                // ==============================================

                // typing variants, in case the keyboard doesn't have accents
                
                "a`>{alpha+};"
                "e`>{epsilon+};"
                "ee`>{eta+};"
                "ee>{eta};" 
                "i`>{iota+};"
                "o`>{omicron+};"
                "oo`>{omega+};"
                "oo>{omega};"
                "i\">{iota_diaeresis};"
                "y\">{ypsilon_diaeresis};"
                
                // basic letters
                
                "a<>{alpha};"
                "\u00e1<>{alpha+};"
                "b<>{beta};"
                "n ({n-gamma}) <> {gamma} ({gamma-n});"
                "g<>{gamma};"
                "d<>{delta};"
                "''e <> ([Ee]){epsilon};" // handle EE
                "e<>{epsilon};"
                "\u00e9<>{epsilon+};"
                "z<>{zeta};"
                "{e-macron-acute}<>{eta+};"
                "{e-macron}<>{eta};"
                "th<>{theta};"
                "i<>{iota};"
                "\u00ed<>{iota+};"
                "\u00ef<>{iota_diaeresis};"
                "k<>{kappa};"
                "l<>{lambda};"
                "m<>{mu};"
                "n'' <> {nu} ({gamma-n});"
                "n<>{nu};"
                "x<>{xi};"
                "''o <> ([Oo]) {omicron};" // handle OO
                "o<>{omicron};"
                "\u00f3<>{omicron+};"
                "ph<>{phi};" // needs ordering before p
                "ps<>{psi};" // needs ordering before p
                "p<>{pi};"
                "r<>{rho};"
                "t<>{tau};"
                "u <> ({vowel}){ypsilon};"
                "\u00fa <> ({vowel}){ypsilon+};"
                "\u00fc <> ({vowel}){ypsilon_diaeresis};"
                "y<>{ypsilon};"
                "\u00fd<>{ypsilon+};"
                "{y-umlaut}<>{ypsilon_diaeresis};"
                "ch<>{chi};"
                "{o-macron-acute}<>{omega+};"
                "{o-macron}<>{omega};"

                // extra english letters. mapped for completeness
                
                "c({softener})>|s;"
                "c>|k;"
                "f>|ph;"
                "h>|ch;"
                "j>|i;"
                "q>|k;"
                "v>|u;"
                "w>|u;"
                
                // ====================================
                // Normal final rule: remove '
                // ====================================
                
                //+ "''>;"
    }
}