174 lines
6.3 KiB
Java
174 lines
6.3 KiB
Java
/*
|
|
The contents of this file are subject to the THDL Open Community License
|
|
Version 1.0 (the "License"); you may not use this file except in compliance
|
|
with the License. You may obtain a copy of the License on the THDL web site
|
|
(http://www.thdl.org/).
|
|
|
|
Software distributed under the License is distributed on an "AS IS" basis,
|
|
WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
|
|
License for the specific terms governing rights and limitations under the
|
|
License.
|
|
|
|
The Initial Developer of this software is the Tibetan and Himalayan Digital
|
|
Library (THDL). Portions created by the THDL are Copyright 2001-2003 THDL.
|
|
All Rights Reserved.
|
|
|
|
Contributor(s): ______________________________________.
|
|
*/
|
|
|
|
package org.thdl.tib.text.tshegbar;
|
|
|
|
/** Constants and static routines (DLC still?) useful in writing state
|
|
* machines for transforming Unicode input into other forms.
|
|
*
|
|
* @author David Chandler
|
|
*/
|
|
interface UnicodeReadingStateMachineConstants {
|
|
|
|
/** Returns the codepoint class for cp, e.g. {@link #CC_SJC}.
|
|
* @param cp a Unicode codepoint, which MUST be nondecomposable
|
|
* if it is in the Tibetan range but can be from outside the
|
|
* Tibetan range of Unicode */
|
|
static int getCCForCP(char cp) {
|
|
assert(getNFTHDL(cp) == null);
|
|
if ('\u0F82' == cp) {
|
|
return CC_0F82;
|
|
} else if ('\u0F8A' == cp) {
|
|
return CC_0F8A;
|
|
} else if ('\u0F39' == cp) {
|
|
return CC_0F39;
|
|
} else if ('\u0F71' == cp) {
|
|
return CC_ACHUNG;
|
|
} else if ('\u0F40' <= cp && cp <= '\u0F6A') {
|
|
assert(cp != '\u0F48');
|
|
return CC_CON;
|
|
} else if ('\u0F90' <= cp && cp <= '\u0FBC') {
|
|
assert(cp != '\u0F98');
|
|
return CC_SJC;
|
|
} else if ('\u0F20' <= cp && cp <= '\u0F33') {
|
|
return CC_DIGIT;
|
|
} else if (/* DLC NOW do these combine ONLY with digits, or do CC_CM just NOT combine with digits? */
|
|
'\u0F3E' == cp
|
|
|| '\u0F3F' == cp
|
|
|| '\u0F18' == cp
|
|
|| '\u0F19' == cp) {
|
|
return CC_MCWD;
|
|
} else if ('\u0FC6' == cp
|
|
|| '\u0F87' == cp
|
|
|| '\u0F86' == cp
|
|
|| '\u0F84' == cp
|
|
|| '\u0F83' == cp
|
|
|| '\u0F82' == cp
|
|
|| '\u0F7F' == cp
|
|
|| '\u0F7E' == cp
|
|
|| '\u0F37' == cp /* DLC NOW NORMALIZATION OF 0F10, 11 to 0F0F ??? */
|
|
|| '\u0F35' == cp) {
|
|
return CC_CM;
|
|
} else if ('\u0F72' == cp
|
|
|| '\u0F74' == cp
|
|
|| '\u0F7A' == cp
|
|
|| '\u0F7B' == cp
|
|
|| '\u0F7C' == cp
|
|
|| '\u0F7D' == cp
|
|
|| '\u0F80' == cp) {
|
|
// DLC what about U+0F84 ??? CC_V or CC_CM ?
|
|
return CC_V;
|
|
} else {
|
|
return CC_SIN;
|
|
}
|
|
}
|
|
|
|
// codepoint classes (CC_...) follow. These are mutually
|
|
// exclusive, and their union is the whole of Unicode.
|
|
|
|
/** for everything else, i.e. non-Tibetan characters like U+0E00
|
|
* and also Tibetan characters like U+0FCF and U+0F05 (DLC rename
|
|
* SIN[GLETON] to OTHER as combining marks from outside the
|
|
* Tibetan range count as this) but not U+0F8A */
|
|
static final int CC_SIN = 0;
|
|
|
|
/** for combining marks in the Tibetan range of Unicode that
|
|
* combine with digits alone */
|
|
static final int CC_MCWD = 1;
|
|
|
|
/** for combining marks in the Tibetan range of Unicode, minus
|
|
* CC_MCWD, U+0F82, and U+0F39 */
|
|
static final int CC_CM = 2;
|
|
|
|
/** for combining consonants, i.e. U+0F90-U+0FBC minus U+0F98
|
|
* minus the decomposable entries like U+0F93, U+0F9D, U+0FA2,
|
|
* etc. */
|
|
static final int CC_SJC = 3;
|
|
|
|
/** for noncombining consonants, i.e. U+0F40-U+0F6A minus U+0F48
|
|
* minus the decomposable entries like U+0F43, U+0F4D, U+0F52,
|
|
* etc. */
|
|
static final int CC_CON = 4;
|
|
|
|
/** for simple, nondecomposable vowels, i.e. U+0F72, U+0F74,
|
|
* U+0F7A, U+0F7B, U+0F7C, U+0F7D, U+0F80 */
|
|
static final int CC_V = 5;
|
|
|
|
/** for U+0F8A */
|
|
static final int CC_0F8A = 6;
|
|
|
|
/** for U+0F82, which is treated like {@link #CC_CM} except after
|
|
* U+0F8A */
|
|
static final int CC_0F82 = 7;
|
|
|
|
/** for U+0F39, an integral part of a consonant when it directly
|
|
* follows a member of CM_CONS or CM_SJC */
|
|
static final int CC_0F39 = 8;
|
|
|
|
/** for U+0F71 */
|
|
static final int CC_ACHUNG = 9;
|
|
|
|
/** for digits, i.e. U+0F20-U+0F33 */
|
|
static final int CC_DIGIT = 10;
|
|
|
|
|
|
|
|
// states STATE_...:
|
|
|
|
/** initial state */
|
|
static final int STATE_START = 0;
|
|
|
|
/** ready state, i.e. the state in which some non-empty Unicode
|
|
* String is in the holding area, <i>ready</i> to receive
|
|
* combining marks like U+0F35 */
|
|
static final int STATE_READY = 1;
|
|
|
|
/** digit state, i.e. the state in which some non-empty Unicode
|
|
* String consisting entirely of digits is in the holding area,
|
|
* ready to receive marks that combine only with digits */
|
|
static final int STATE_DIGIT = 2;
|
|
|
|
/** state in which CC_SJC are welcomed and treated as consonants
|
|
* to be subscribed to the GraphemeCluster in holding. */
|
|
static final int STATE_STACKING = 3;
|
|
|
|
/** state in which one or more consonants have been seen and also
|
|
* an achung (U+0F71) has been seen */
|
|
static final int STATE_STACKPLUSACHUNG = 4;
|
|
|
|
/** state that seeing U+0F8A (when that's not an error) puts you
|
|
* in. Needed because U+0F8A is always followed by U+0F82, and
|
|
* we check for the exceptional case that U+0F8A is followed by
|
|
* something else. */
|
|
static final int STATE_PARTIALMARK = 5;
|
|
|
|
/* DLC we should have many error states or none. */
|
|
|
|
|
|
/** the present codepoint marks the start of a new
|
|
* GraphemeCluster */
|
|
static final int ACTION_BEGINS_NEW_GRAPHEME_CLUSTER = 0;
|
|
/** the present codepoint is a continuation of the current
|
|
* GraphemeCluster */
|
|
static final int ACTION_CONTINUES_GRAPHEME_CLUSTER = 1;
|
|
/** there is an error in the input stream, which we are correcting
|
|
* (as we are in error-correcting mode) by starting a new
|
|
* GraphemeCluster with U+0F68 as the first codepoint and the
|
|
* current codepoint as the second */
|
|
static final int ACTION_PREPEND_WITH_0F68 = 2;
|
|
}
|