kolibrios-fun/programs/develop/cedit/SRC/Encodings.ob07

(*
    Copyright 2021, 2022 Anton Krotov

    This file is part of CEdit.

    CEdit is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

    CEdit is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with CEdit. If not, see <http://www.gnu.org/licenses/>.
*)

MODULE Encodings;

CONST
    CP866* = 0; W1251* = 1; UTF8* = 2; UTF8BOM* = 3; (*UTF16LE* = 4;*)

    UNDEF* = -1;

    EOL_CRLF* = 0; EOL_LF* = 1; EOL_CR* = 2;
    CR = 0DX; LF = 0AX;

TYPE
    CP = ARRAY 256 OF INTEGER;

    tEOL* = ARRAY 3 OF WCHAR;

VAR
    cpW1251*, cp866*: CP;
    names*: ARRAY 4, 16 OF WCHAR;
    UNI*: ARRAY 65536, 2 OF INTEGER;

    eol*: ARRAY 3 OF tEOL;
    eolNames*: ARRAY 3, 16 OF WCHAR;


PROCEDURE InitCP (VAR cp: CP);
VAR
    i: INTEGER;
BEGIN
    FOR i := 0H TO 7FH DO
        cp[i] := i
    END
END InitCP;


PROCEDURE Init8 (VAR cp: CP; VAR n: INTEGER; a, b, c, d, e, f, g, h: INTEGER);
BEGIN
    cp[n] := a; INC(n);
    cp[n] := b; INC(n);
    cp[n] := c; INC(n);
    cp[n] := d; INC(n);
    cp[n] := e; INC(n);
    cp[n] := f; INC(n);
    cp[n] := g; INC(n);
    cp[n] := h; INC(n);
END Init8;


PROCEDURE InitW1251 (VAR cp: CP);
VAR
    n, i: INTEGER;
BEGIN
	InitCP(cp);
    FOR i := 0410H TO 044FH DO
        cp[i - 350H] := i
    END;
    n := 80H;
    Init8(cp, n, 0402H, 0403H, 201AH, 0453H, 201EH, 2026H, 2020H, 2021H);
    Init8(cp, n, 20ACH, 2030H, 0409H, 2039H, 040AH, 040CH, 040BH, 040FH);
    Init8(cp, n, 0452H, 2018H, 2019H, 201CH, 201DH, 2022H, 2013H, 2014H);
    Init8(cp, n, UNDEF, 2122H, 0459H, 203AH, 045AH, 045CH, 045BH, 045FH);
    Init8(cp, n, 00A0H, 040EH, 045EH, 0408H, 00A4H, 0490H, 00A6H, 00A7H);
    Init8(cp, n, 0401H, 00A9H, 0404H, 00ABH, 00ACH, 00ADH, 00AEH, 0407H);
    Init8(cp, n, 00B0H, 00B1H, 0406H, 0456H, 0491H, 00B5H, 00B6H, 00B7H);
    Init8(cp, n, 0451H, 2116H, 0454H, 00BBH, 0458H, 0405H, 0455H, 0457H);
END InitW1251;


PROCEDURE InitCP866 (VAR cp: CP);
VAR
    n, i: INTEGER;
BEGIN
	InitCP(cp);
	FOR i := 0410H TO 043FH DO
	   cp[i - 390H] := i
	END;
	FOR i := 0440H TO 044FH DO
	   cp[i - 360H] := i
	END;
	n := 0B0H;
	Init8(cp, n, 2591H, 2592H, 2593H, 2502H, 2524H, 2561H, 2562H, 2556H);
	Init8(cp, n, 2555H, 2563H, 2551H, 2557H, 255DH, 255CH, 255BH, 2510H);
	Init8(cp, n, 2514H, 2534H, 252CH, 251CH, 2500H, 253CH, 255EH, 255FH);
	Init8(cp, n, 255AH, 2554H, 2569H, 2566H, 2560H, 2550H, 256CH, 2567H);
	Init8(cp, n, 2568H, 2564H, 2565H, 2559H, 2558H, 2552H, 2553H, 256BH);
	Init8(cp, n, 256AH, 2518H, 250CH, 2588H, 2584H, 258CH, 2590H, 2580H);
	n := 0F0H;
	Init8(cp, n, 0401H, 0451H, 0404H, 0454H, 0407H, 0457H, 040EH, 045EH);
	Init8(cp, n, 00B0H, 2219H, 00B7H, 221AH, 2116H, 00A4H, 25A0H, 00A0H);
END InitCP866;


PROCEDURE setUNI;
VAR
    i: INTEGER;
BEGIN
    FOR i := 0 TO 65535 DO
        UNI[i, CP866] := UNDEF;
        UNI[i, W1251] := UNDEF
    END;
    FOR i := 0 TO 255 DO
        IF cpW1251[i] # UNDEF THEN
            UNI[cpW1251[i], W1251] := i
        END;
        IF cp866[i] # UNDEF THEN
            UNI[cp866[i], CP866] := i
        END
    END
END setUNI;


BEGIN
    eol[EOL_CRLF] := CR + LF;
    eol[EOL_LF]   := LF;
    eol[EOL_CR]   := CR;
    eolNames[EOL_CRLF] := "CRLF";
    eolNames[EOL_LF]   := "LF";
    eolNames[EOL_CR]   := "CR";
    names[CP866]   := "CP866";
    names[W1251]   := "Windows-1251";
    names[UTF8]    := "UTF-8";
    names[UTF8BOM] := "UTF-8-BOM";
    InitW1251(cpW1251);
    InitCP866(cp866);
    setUNI
END Encodings.