thdltools/Jskad
1
0
Fork 0
Code Issues Pull requests Projects Releases Wiki Activity
Jskad/source/org/thdl/tib/text/ttt/ACIPConverter.java

476 lines
24 KiB
Java
Raw Normal View History

ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
/*
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 2003 THDL.
All Rights Reserved.
Contributor(s): ______________________________________.
*/
package org.thdl.tib.text.ttt;
import java.io.*;
import java.util.ArrayList;
import java.util.Stack;
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
import java.awt.Color;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
import org.thdl.util.ThdlDebug;
import org.thdl.util.ThdlOptions;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
import org.thdl.tib.text.TibetanDocument;
import org.thdl.tib.text.TibetanMachineWeb;
import org.thdl.tib.text.DuffCode;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
/**
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
* This class is able to convert an ACIP file into Tibetan Machine Web
* and an ACIP file into TMW. ACIP->Unicode should yield the same
* results as ACIP->TMW followed by TMW->Unicode (FIXME: test it!)
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
* @author David Chandler
*/
public class ACIPConverter {
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
// DLC NOW: (KA)'s info is lost when you convert to Unicode text instead of Unicode RTF. Give an ERROR.
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
/** Command-line converter. Gives error messages on standard
* output about why we can't convert the document perfectly and
* exits with non-zero return code, or is silent otherwise and
* exits with code zero. <p>FIXME: not so efficient; copies the
* whole file into memory first. */
public static void main(String[] args)
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
throws IOException
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
{
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
// We don't want to load the TM or TMW font files ourselves:
ThdlOptions.setUserPreference("thdl.rely.on.system.tmw.fonts", true);
ThdlOptions.setUserPreference("thdl.rely.on.system.tm.fonts", true);
ThdlOptions.setUserPreference("thdl.debug", true);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
TibetanDocument.enableColors();
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
boolean verbose = true;
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
if (args.length != 1) {
System.out.println("Bad args! Need just the name of the ACIP text file.");
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
StringBuffer errors = new StringBuffer();
int maxErrors = 250;
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
ArrayList al = ACIPTshegBarScanner.scanFile(args[0], errors, maxErrors - 1);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (null == al) {
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
System.err.println(maxErrors + " or more lexical errors occurred while scanning ACIP input file; is this");
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
System.err.println("Tibetan or English input?");
System.err.println("");
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
if (false) {
// Nobody wants to see this. FIXME: maybe somebody; have an option.
System.err.println("First " + maxErrors + " lexical errors scanning ACIP input file: ");
System.err.println(errors);
}
System.err.println("Exiting with " + maxErrors + " or more lexical errors; please fix input file and try again.");
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
System.exit(1);
}
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
final boolean abortUponScanningError = false;
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
// DLC NOW: BAo isn't converting.
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (errors.length() > 0) {
System.err.println("Errors scanning ACIP input file: ");
System.err.println(errors);
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
if (abortUponScanningError) {
System.err.println("Exiting; please fix input file and try again.");
System.exit(1);
}
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
String warningLevel = "Most"; // DLC make me configurable.
StringBuffer warnings = null;
boolean putWarningsInOutput = false;
if ("None" != warningLevel) {
warnings = new StringBuffer();
putWarningsInOutput = true;
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
convertToTMW(al, System.out, errors, warnings,
putWarningsInOutput, warningLevel);
int retCode = 0;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (errors.length() > 0) {
System.err.println("Errors converting ACIP input file: ");
System.err.println(errors);
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
System.err.println("The output contains these errors.");
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
System.err.println("Exiting; please fix input file and try again.");
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
retCode = 2;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
if (null != warnings && warnings.length() > 0) {
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
System.err.println("Warnings converting ACIP input file: ");
System.err.println(warnings);
if (putWarningsInOutput)
System.err.println("The output contains these warnings.");
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
retCode = 2;
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (0 == retCode) {
if (verbose) System.err.println("Converted " + args[0] + " perfectly.");
}
System.exit(retCode);
// DLC NOW: tRAStA is not converter correctly to Unicode, and
// no warning is given when converting to TMW.
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
/** Writes TMW/Latin to out. If errors occur in converting a
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
* tsheg bar, then they are written into the output, and also
* appended to errors if errors is non-null. If warnings occur
* in converting a tsheg bar, then they are written into the
* output if writeWarningsToResult is true, and also appended to
* warnings if warnings is non-null. Returns true upon perfect
* success or if there were merely warnings, false if errors
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
* occurred.
* @throws IOException if we cannot write to out
*/
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
public static boolean convertToTMW(ArrayList scan,
OutputStream out,
StringBuffer errors,
StringBuffer warnings,
boolean writeWarningsToResult,
String warningLevel)
throws IOException
{
TibetanDocument tdoc = new TibetanDocument();
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
tdoc.setRomanAttributeSet(ThdlOptions.getStringOption("thdl.acip.to.x.latin.font",
"Courier New"),
ThdlOptions.getIntegerOption("thdl.acip.to.x.latin.font.size",
20));
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
boolean rv
= convertToTMW(scan, tdoc, errors, warnings,
writeWarningsToResult, warningLevel);
tdoc.writeRTFOutputStream(out);
return rv;
}
private static boolean convertToTMW(ArrayList scan,
TibetanDocument tdoc,
StringBuffer errors,
StringBuffer warnings,
boolean writeWarningsToResult,
String warningLevel)
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
throws IOException
{
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
return convertTo(false, scan, null, tdoc, errors, warnings,
writeWarningsToResult, warningLevel);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
// DLC FIXME: sometimes { } is \u0F0B, and sometimes it is a
// space. Treat it as a tsheg only when it appears after a
// syllable or another tsheg.
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
/** Returns UTF-8 encoded Unicode. A bit indirect, so use this
* for testing only if performance is a concern. If errors occur
* in scanning the ACIP or in converting a tsheg bar, then they
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
* are appended to errors if errors is non-null, as well as
* written to the result. If warnings occur in scanning the ACIP
* or in converting a tsheg bar, then they are appended to
* warnings if warnings is non-null, and they are written to the
* result if writeWarningsToResult is true. Returns the
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
* conversion upon perfect success or if there were merely
* warnings, null if errors occurred.
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
*/
public static String convertToUnicode(String acip,
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
StringBuffer errors,
StringBuffer warnings,
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
boolean writeWarningsToResult,
String warningLevel) {
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
ByteArrayOutputStream sw = new ByteArrayOutputStream();
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
ArrayList al = ACIPTshegBarScanner.scan(acip, errors, -1);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
try {
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
if (null != al
&& convertToUnicode(al, sw, errors,
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
warnings, writeWarningsToResult,
warningLevel)) {
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
return sw.toString("UTF-8");
} else {
return null;
}
} catch (Exception e) {
throw new Error(e.toString());
}
}
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
/** Writes Unicode to out. If errors occur in converting a tsheg
* bar, then they are appended to errors if errors is non-null.
* Furthermore, errors are written to out. If writeWarningsToOut
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
* is true, then warnings also will be written to out.
* @return true upon perfect success, false if errors occurred.
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
* @param scan result of ACIPTshegBarScanner.scan(..)
* @param out stream to which to write converted text
* @param errors if non-null, all error messages are appended
* @param warnings if non-null, all warning messages are appended
* to this
* @param writeWarningsToOut if true, then all warning messages
* are written to out in the appropriate places
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
* @throws IOException if we cannot write to out
*/
public static boolean convertToUnicode(ArrayList scan,
OutputStream out,
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
StringBuffer errors,
StringBuffer warnings,
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
boolean writeWarningsToOut,
String warningLevel)
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
throws IOException
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
{
return convertTo(true, scan, out, null, errors, warnings,
writeWarningsToOut, warningLevel);
}
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
private static boolean peekaheadFindsSpacesAndComma(ArrayList /* of ACIPString */ scan,
int pos) {
int sz = scan.size();
while (pos < sz) {
ACIPString s = (ACIPString)scan.get(pos++);
if (s.getType() == ACIPString.TIBETAN_PUNCTUATION && s.getText().equals(" ")) {
// keep going
} else {
if (s.getType() == ACIPString.TIBETAN_PUNCTUATION && s.getText().equals(",")) {
return true;
} else {
return false;
}
}
}
return false;
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
private static boolean convertTo(boolean toUnicode, // else to TMW
ArrayList scan,
OutputStream out, // for toUnicode mode
TibetanDocument tdoc, // for !toUnicode mode
StringBuffer errors,
StringBuffer warnings,
boolean writeWarningsToOut,
String warningLevel)
throws IOException
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
{
int sz = scan.size();
boolean hasErrors = false;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
BufferedWriter writer = null;
if (toUnicode)
writer
= new BufferedWriter(new OutputStreamWriter(out, "UTF-8"));
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
boolean lastGuyWasNonPunct = false;
TStackList lastGuy = null;
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
Color lastColor = Color.BLACK;
Color color = Color.BLACK;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
for (int i = 0; i < sz; i++) {
ACIPString s = (ACIPString)scan.get(i);
int stype = s.getType();
if (stype == ACIPString.ERROR) {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuyWasNonPunct = false;
lastGuy = null;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
hasErrors = true;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
String text = "[#ERROR CONVERTING ACIP DOCUMENT: Lexical error: " + s.getText() + "]";
if (null != writer) writer.write(text);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(text, Color.RED);
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
} else if (stype == ACIPString.WARNING) {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuyWasNonPunct = false;
lastGuy = null;
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
if (writeWarningsToOut) {
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
String text = "[#WARNING CONVERTING ACIP DOCUMENT: Lexical warning: " + s.getText() + "]";
if (null != writer) writer.write(text);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(text, Color.RED);
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
// DLC NOW: Warning: We're going with {'}{R}{DA}, but only because our knowledge of prefix rules says that {'}{R+DA} is not a legal Tibetan tsheg bar ("syllable")
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
if (null != warnings) {
warnings.append("Warning: Lexical warning: ");
warnings.append(s.getText());
warnings.append('\n');
}
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
} else {
if (s.isLatin(stype)) {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuyWasNonPunct = false;
lastGuy = null;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
String text
= (((stype == ACIPString.FOLIO_MARKER) ? "{" : "")
+ s.getText()
+ ((stype == ACIPString.FOLIO_MARKER) ? "}" : ""));
if (null != writer) writer.write(text);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(text, Color.BLACK);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
} else {
String unicode = null;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
DuffCode[] duff = null;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (stype == ACIPString.TIBETAN_NON_PUNCTUATION) {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuyWasNonPunct = true;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
TPairList pl = TPairListFactory.breakACIPIntoChunks(s.getText());
String acipError;
if ((acipError = pl.getACIPError()) != null) {
hasErrors = true;
String errorMessage = "[#ERROR CONVERTING ACIP DOCUMENT: THE TSHEG BAR (\"SYLLABLE\") " + s.getText() + " HAS THESE ERRORS: " + acipError + "]";
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer) writer.write(errorMessage);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(errorMessage, Color.RED);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (null != errors)
errors.append(errorMessage + "\n");
} else {
TParseTree pt = pl.getParseTree();
if (null == pt) {
hasErrors = true;
String errorMessage = "[#ERROR CONVERTING ACIP DOCUMENT: THE TSHEG BAR (\"SYLLABLE\") " + s.getText() + " IS ESSENTIALLY NOTHING.]";
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer) writer.write(errorMessage);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(errorMessage, Color.RED);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (null != errors)
errors.append(errorMessage + "\n");
} else {
TStackList sl = pt.getBestParse();
if (null == sl) {
hasErrors = true;
String errorMessage = "[#ERROR CONVERTING ACIP DOCUMENT: THE TSHEG BAR (\"SYLLABLE\") " + s.getText() + " HAS NO LEGAL PARSES.]";
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer) writer.write(errorMessage);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(errorMessage, Color.RED);
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
if (null != errors)
errors.append(errorMessage + "\n");
} else {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuy = sl;
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
String warning
Jskad's converter now has ACIP-to-Unicode built in. There are known bugs; it is pre-alpha. It's usable, though, and finds tons of errors in ACIP input files, with the user deciding just how pedantic to be. The biggest outstanding bug is the silent one: treating { }, space, as tsheg instead of whitespace when we ought to know better.
2003-08-24 06:40:53 +00:00
= pt.getWarning(warningLevel,
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
pl,
s.getText());
if (null != warning) {
if (writeWarningsToOut) {
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
String text
= ("[#WARNING CONVERTING ACIP DOCUMENT: "
+ warning + "]");
if (null != writer) writer.write(text);
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (null != tdoc) tdoc.appendRoman(text, Color.RED);
TMW->Wylie conversion now takes advantage of prefix rules, the rules that say "ya can take a ga prefix" etc. The ACIP->Unicode converter now gives warnings (optionally, and by default, inline). This converter now produces output even when lexical errors occur, but the output has errors and warnings inline.
2003-08-23 22:03:37 +00:00
}
if (null != warnings) {
warnings.append(warning);
warnings.append('\n');
}
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer) {
unicode = sl.getUnicode();
if (null == unicode) throw new Error("FIXME: make this an assertion 4");
}
if (null != tdoc) {
duff = sl.getDuff();
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
if (sl.isLegalTshegBar(true).isLegal && !sl.isLegalTshegBar(false).isLegal) {
color = Color.GRAY;
} else if (sl.isLegalTshegBar(false).isLegal) {
color = Color.BLACK;
} else {
color = Color.YELLOW;
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (0 == duff.length) {
throw new Error("No DuffCodes for stack list " + sl); // FIXME: make this an assertion
}
}
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
}
}
} else {
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
color = Color.BLACK;
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (stype == ACIPString.START_SLASH) {
if (null != writer) unicode = "\u0F3C";
if (null != tdoc) duff = new DuffCode[] { TibetanMachineWeb.getGlyph("(") };
} else if (stype == ACIPString.END_SLASH) {
if (null != writer) unicode = "\u0F3D";
if (null != tdoc) duff = new DuffCode[] { TibetanMachineWeb.getGlyph(")") };
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
} else if (stype == ACIPString.TIBETAN_PUNCTUATION) {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
// For ACIP, tshegs are used as both
// tshegs and whitespace. We treat a
// space as a tsheg if and only if it
// occurs after TIBETAN_NON_PUNCTUATION.
// But "SHIG ,MDO" is an example of a
// special case, needed because a tsheg is
// not used after a GA in Tibetan
// typesetting.
boolean done = false;
// DLC what about after numbers? marks?
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
TPairList lpl = null;
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
if (s.getText().equals(" ")) {
if (!lastGuyWasNonPunct
|| (null != lastGuy
&& (lpl = lastGuy.get(lastGuy.size() - 1)).size() == 1
&& lpl.get(0).getLeft().equals("G")
&& // it's (G . anything)
// followed by some number
// of spaces (at least one,
// this one) and then a
// comma:
peekaheadFindsSpacesAndComma(scan, i+1))) {
if (null != writer) {
unicode = " ";
done = true;
}
if (null != tdoc) {
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
tdoc.appendRoman(" ", Color.BLACK);
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
continue;
}
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
// DLC AM I DOING THIS? By normal Tibetan & Dzongkha spelling, writing, and input rules
// Tibetan script stacks should be entered and written: 1 headline
// consonant (0F40->0F6A), any subjoined consonant(s) (0F90->
// 0F9C), achung (0F71), shabkyu (0F74), any above headline
// vowel(s) (0F72 0F7A 0F7B 0F7C 0F7D and 0F80) ; any ngaro (0F7E,
// 0F82 and 0F83)
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
}
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
} else if (s.getText().equals(",")
&& lastGuyWasNonPunct
&& null != lastGuy
&& (lpl = lastGuy.get(lastGuy.size() - 1)).size() == 1
&& lpl.get(0).getLeft().equals("NG")) {
DuffCode tshegDuff = TibetanMachineWeb.getGlyph(" ");
if (null == tshegDuff) throw new Error("tsheg duff");
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
tdoc.appendDuffCodes(new DuffCode[] { tshegDuff }, lastColor);
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
}
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
if (!done) {
if (null != writer) unicode = ACIPRules.getUnicodeFor(s.getText(), false);
if (null != tdoc) {
ACIP->TMW and ACIP->Unicode have my pre-stamp of non-approval. Except for (NYAx} and {NYAo}, they're as good as I'll get them without input from experts of the employ of a complementary, syllabary-based approach.
2003-09-04 04:34:18 +00:00
if (s.getText().equals("\r")
|| s.getText().equals("\t")
|| s.getText().equals("\n")
|| s.getText().equals("\r\n")) {
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
tdoc.appendRoman(s.getText(), Color.BLACK);
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
continue;
ACIP->TMW and ACIP->Unicode have my pre-stamp of non-approval. Except for (NYAx} and {NYAo}, they're as good as I'll get them without input from experts of the employ of a complementary, syllabary-based approach.
2003-09-04 04:34:18 +00:00
} else {
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
String wy = ACIPRules.getWylieForACIPOther(s.getText());
if (null == wy) throw new Error("No wylie for ACIP " + s.getText());
duff = new DuffCode[] { TibetanMachineWeb.getGlyph(wy) };
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
}
}
Fixed ACIP->Unicode spaces/tshegs and newlines, especially with shads. "NGA," becomes "NGA-tsheg-," automatically now.
2003-09-05 05:08:47 +00:00
} else {
throw new Error("forgot a case");
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
}
if (null != writer && null == unicode)
throw new Error("FIXME: make this an assertion 1");
if (null != tdoc && (null == duff || 0 == duff.length))
throw new Error("FIXME: make this an assertion 2");
ACIP->TMW and ACIP->Unicode are now smart about when a newline is really a newline and when a space is really a tsheg. The space in {KA ,MDO} is a tsheg, but the space in {GA ,MDO} is not.
2003-09-04 04:04:21 +00:00
lastGuyWasNonPunct = false;
lastGuy = null;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer && null != unicode) writer.write(unicode);
if (null != tdoc) {
if (null != duff && 0 != duff.length) {
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
tdoc.appendDuffCodes(duff, color);
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
// DLC NOW FIXME: use TibTextUtils.getVowel logic to make the output beautiful.
} else {
// this happens when you have an
// [#ERROR]-producing tsheg bar.
// System.err.println("Bad tsheg bar with ACIP {" + s.getText() + "}");
}
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
}
}
Preliminary, untested color support in ACIP->TMW conversion.
2003-09-05 05:54:35 +00:00
lastColor = color;
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
}
Preliminary ACIP->TMW support is in place. {DU} gives you something less beautiful than what Jskad would give, so more work is needed.
2003-08-31 16:06:35 +00:00
if (null != writer) {
writer.close();
}
ACIP->Unicode, without going through TMW, is now possible, so long as \, the Sanskrit virama, is not used. Of the 1370-odd ACIP texts I've got here, about 57% make it through the gauntlet (fewer if you demand a vowel or disambiguator on every stack of a non-Tibetan tsheg bar).
2003-08-18 02:38:54 +00:00
return !hasErrors;
}
}
// DLC FIXME: putting Tibetan in black, Sanskrit in green, and Latin
// in yellow would help you quickly decide if ZHIGN maybe should've
// been ZHING.
Reference in a new issue Copy permalink