From 33fc836e814b19db715bc001a24e88b4754dda07 Mon Sep 17 00:00:00 2001
From: dchandler <dchandler>
Date: Sun, 10 Jul 2005 05:01:03 +0000
Subject: [PATCH] EWTS->Unicode for // now produces \u0f0e as it should.

---
 source/org/thdl/tib/text/ttt/EWTSTest.java    |  4 ++-
 source/org/thdl/tib/text/ttt/EWTSTraits.java  |  1 +
 .../tib/text/ttt/EWTSTshegBarScanner.java     | 28 +++++++++++--------
 3 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/source/org/thdl/tib/text/ttt/EWTSTest.java b/source/org/thdl/tib/text/ttt/EWTSTest.java
index 2dc673d..ed55d1a 100644
--- a/source/org/thdl/tib/text/ttt/EWTSTest.java
+++ b/source/org/thdl/tib/text/ttt/EWTSTest.java
@@ -676,7 +676,9 @@ public class EWTSTest extends TestCase {
         ewts2uni_test(" ", "\u0F0B");
         ewts2uni_test("*", "\u0F0C");
         ewts2uni_test("/", "\u0F0D");
-        if (RUN_FAILING_TESTS) ewts2uni_test("//", "\u0F0E");
+        ewts2uni_test("//", "\u0F0E");
+        ewts2uni_test("////", "\u0F0E\u0f0e");
+        ewts2uni_test("/////", "\u0F0E\u0f0e\u0f0d");
         ewts2uni_test(";", "\u0F0F");
         ewts2uni_test("\\u0F10", "\u0F10");
         ewts2uni_test("|", "\u0F11");
diff --git a/source/org/thdl/tib/text/ttt/EWTSTraits.java b/source/org/thdl/tib/text/ttt/EWTSTraits.java
index 803b52c..d489076 100644
--- a/source/org/thdl/tib/text/ttt/EWTSTraits.java
+++ b/source/org/thdl/tib/text/ttt/EWTSTraits.java
@@ -397,6 +397,7 @@ public final class EWTSTraits implements TTraits {
             if ("R".equals(l)) return "\u0f6a";
             if ("Y".equals(l)) return "\u0f61";
             if ("W".equals(l)) return "\u0f5d";
+            if ("//".equals(l)) return "\u0f0e";
             
             if (!TibetanMachineWeb.isKnownHashKey(l)) {
 //                 System.err.println("Getting unicode for the following is hard: '"
diff --git a/source/org/thdl/tib/text/ttt/EWTSTshegBarScanner.java b/source/org/thdl/tib/text/ttt/EWTSTshegBarScanner.java
index 5447e89..f6201c7 100644
--- a/source/org/thdl/tib/text/ttt/EWTSTshegBarScanner.java
+++ b/source/org/thdl/tib/text/ttt/EWTSTshegBarScanner.java
@@ -79,7 +79,7 @@ class EWTSTshegBarScanner extends TTshegBarScanner {
     // TODO(DLC)[EWTS->Tibetan]:: '@#', in ewts->tmw, is not working
     // TODO(DLC)[EWTS->Tibetan]:: 'jamX 'jam~X one is not working in ->tmw mode
     // TODO(DLC)[EWTS->Tibetan]:: dzaHsogs is not working
-    for (int i = 0; i < sl; i++) {
+    for (int i = 0; i < sl; i++) {  // i is modified in the loop, also
       if (isValidInsideTshegBar(sb.charAt(i))) {
         StringBuffer tbsb = new StringBuffer();
         for (; i < sl; i++) {
@@ -96,16 +96,22 @@ class EWTSTshegBarScanner extends TTshegBarScanner {
         // NOTE: It's questionable, but we treat
         // \u0f00 like punctuation because it was
         // easier coding that way.
-        if ((sb.charAt(i) >= EWTSTraits.PUA_MIN
-             && sb.charAt(i) <= EWTSTraits.PUA_MAX)
-            || (sb.charAt(i) >= '\u0f00' && sb.charAt(i) <= '\u0f17')
-            || (sb.charAt(i) >= '\u0f1a' && sb.charAt(i) <= '\u0f1f')
-            || (sb.charAt(i) >= '\u0fbe' && sb.charAt(i) <= '\u0fcc')
-            || (sb.charAt(i) >= '\u0fcf' && sb.charAt(i) <= '\u0fd1')
-            || (EWTSTraits.SAUVASTIKA == sb.charAt(i))
-            || (EWTSTraits.SWASTIKA == sb.charAt(i))
-            || (" /;|!:=_@#$%<>()*&\r\n\t\u0f36\u0f38\u0f89\u0f8a\u0f8b".indexOf(sb.charAt(i))
-                >= 0)) {
+        if (i + 1 < sl
+            && sb.charAt(i) == '/'
+            && sb.charAt(i + 1) == '/') {
+          al.add(new TString("EWTS", "//",
+                             TString.TIBETAN_PUNCTUATION));
+          ++i;
+        } else if ((sb.charAt(i) >= EWTSTraits.PUA_MIN
+                    && sb.charAt(i) <= EWTSTraits.PUA_MAX)
+                   || (sb.charAt(i) >= '\u0f00' && sb.charAt(i) <= '\u0f17')
+                   || (sb.charAt(i) >= '\u0f1a' && sb.charAt(i) <= '\u0f1f')
+                   || (sb.charAt(i) >= '\u0fbe' && sb.charAt(i) <= '\u0fcc')
+                   || (sb.charAt(i) >= '\u0fcf' && sb.charAt(i) <= '\u0fd1')
+                   || (EWTSTraits.SAUVASTIKA == sb.charAt(i))
+                   || (EWTSTraits.SWASTIKA == sb.charAt(i))
+                   || (" /;|!:=_@#$%<>()*&\r\n\t\u0f36\u0f38\u0f89\u0f8a\u0f8b".indexOf(sb.charAt(i))
+                       >= 0)) {
           al.add(new TString("EWTS", sb.substring(i, i+1),
                              TString.TIBETAN_PUNCTUATION));
         } else {