From 48eda8bd06a5bc2243743a7985758b0b3ac28b81 Mon Sep 17 00:00:00 2001
From: George Rhoten <grhoten@users.noreply.github.com>
Date: Wed, 9 Apr 2003 00:00:51 +0000
Subject: [PATCH] ICU-1439 implement ucnv_getUnicodeSet() for roundtrippable
 code points.

X-SVN-Rev: 11475
---
 icu4c/source/common/ucnv2022.c     | 64 ++++++++++++++++++++++++++----
 icu4c/source/common/unicode/uset.h |  9 +++++
 icu4c/source/common/uset.cpp       |  5 +++
 3 files changed, 71 insertions(+), 7 deletions(-)

diff --git a/icu4c/source/common/ucnv2022.c b/icu4c/source/common/ucnv2022.c
index d04932eed60..25fec6a4d19 100644
--- a/icu4c/source/common/ucnv2022.c
+++ b/icu4c/source/common/ucnv2022.c
@@ -31,6 +31,7 @@
 #if !UCONFIG_NO_LEGACY_CONVERSION
 
 #include "unicode/ucnv.h"
+#include "unicode/uset.h"
 #include "unicode/ucnv_err.h"
 #include "unicode/ucnv_cb.h"
 #include "ucnv_bld.h"
@@ -116,6 +117,12 @@ _MBCSToUnicodeWithOffsets(UConverterToUnicodeArgs *pArgs,
 
 /* Protos */
 /***************** ISO-2022 ********************************/
+static void
+_ISO_2022_GetUnicodeSet(const UConverter *cnv,
+                    USet *set,
+                    UConverterUnicodeSet which,
+                    UErrorCode *pErrorCode);
+
 static void 
 T_UConverter_toUnicode_ISO_2022(UConverterToUnicodeArgs * args,
                                              UErrorCode * err);
@@ -369,7 +376,8 @@ static const UConverterImpl _ISO2022Impl={
     NULL,
     _ISO2022getName,
     _ISO_2022_WriteSub,
-    _ISO_2022_SafeClone
+    _ISO_2022_SafeClone,
+    _ISO_2022_GetUnicodeSet
 };
 static const UConverterStaticData _ISO2022StaticData={
     sizeof(UConverterStaticData),
@@ -418,7 +426,8 @@ static const UConverterImpl _ISO2022JPImpl={
     NULL,
     _ISO2022getName,
     _ISO_2022_WriteSub,
-    _ISO_2022_SafeClone
+    _ISO_2022_SafeClone,
+    _ISO_2022_GetUnicodeSet
 };
 static const UConverterStaticData _ISO2022JPStaticData={
     sizeof(UConverterStaticData),
@@ -467,7 +476,8 @@ static const UConverterImpl _ISO2022KRImpl={
     NULL,
     _ISO2022getName,
     _ISO_2022_WriteSub,
-    _ISO_2022_SafeClone
+    _ISO_2022_SafeClone,
+    _ISO_2022_GetUnicodeSet
 };
 static const UConverterStaticData _ISO2022KRStaticData={
     sizeof(UConverterStaticData),
@@ -517,7 +527,8 @@ static const UConverterImpl _ISO2022CNImpl={
     NULL,
     _ISO2022getName,
     _ISO_2022_WriteSub,
-    _ISO_2022_SafeClone
+    _ISO_2022_SafeClone,
+    _ISO_2022_GetUnicodeSet
 };
 static const UConverterStaticData _ISO2022CNStaticData={
     sizeof(UConverterStaticData),
@@ -565,7 +576,7 @@ _ISO2022Open(UConverter *cnv, const char *name, const char *locale,uint32_t opti
         myConverterData->isFirstBuffer = TRUE;
         cnv->fromUnicodeStatus =FALSE;
         if(locale){
-            uprv_strcpy(myLocale,locale);
+            uprv_strncpy(myLocale, locale, sizeof(myLocale));
             myConverterData->isLocaleSpecified = TRUE;
         }
         myConverterData->version= 0;
@@ -642,8 +653,8 @@ _ISO2022Open(UConverter *cnv, const char *name, const char *locale,uint32_t opti
             uprv_strcpy(myConverterData->locale,"cn");
 
             if ((options  & UCNV_OPTIONS_VERSION_MASK)==1){
-               myConverterData->version = 1;
-               uprv_strcpy(myConverterData->name,"ISO_2022,locale=cn,version=1");
+                myConverterData->version = 1;
+                uprv_strcpy(myConverterData->name,"ISO_2022,locale=cn,version=1");
             }else{
                 uprv_strcpy(myConverterData->name,"ISO_2022,locale=cn,version=0");
                 myConverterData->version = 0;
@@ -3406,4 +3417,43 @@ _ISO_2022_SafeClone(
     return &localClone->cnv;
 }
 
+static void
+_ISO_2022_GetUnicodeSet(const UConverter *cnv,
+                    USet *set,
+                    UConverterUnicodeSet which,
+                    UErrorCode *pErrorCode)
+{
+    int32_t i;
+    USet *cnvSet;
+    UConverterDataISO2022* cnvData;
+
+    if (U_FAILURE(*pErrorCode)) {
+        return;
+    }
+    if (cnv->sharedData == &_ISO2022Data) {
+        /* We use UTF-8 in this case */
+        uset_addRange(set, 0, 0xd7FF);
+        uset_addRange(set, 0xE000, 0x10FFFF);
+        return;
+    }
+
+    cnvData = (UConverterDataISO2022*)cnv->extraInfo;
+    if (cnv->sharedData == &_ISO2022KRData && cnvData->currentConverter != NULL) {
+        ucnv_getUnicodeSet(cnvData->currentConverter, set, which, pErrorCode);
+        return;
+    }
+
+    cnvSet = uset_open(0, 0);
+    if (!cnvSet) {
+        *pErrorCode =U_MEMORY_ALLOCATION_ERROR;
+        return;
+    }
+
+    for (i=0; (i<UCNV_2022_MAX_CONVERTERS) && cnvData->myConverterArray[i]; i++) {
+        ucnv_getUnicodeSet(cnvData->myConverterArray[i], cnvSet, which, pErrorCode);
+        uset_addAll(set, cnvSet /* pErrorCode */);
+    }
+    uset_close(cnvSet);
+}
+
 #endif /* #if !UCONFIG_NO_LEGACY_CONVERSION */
diff --git a/icu4c/source/common/unicode/uset.h b/icu4c/source/common/unicode/uset.h
index 2cbbabe37ca..ded37ccd7d3 100644
--- a/icu4c/source/common/unicode/uset.h
+++ b/icu4c/source/common/unicode/uset.h
@@ -183,6 +183,15 @@ uset_toPattern(const USet* set,
 U_CAPI void U_EXPORT2
 uset_add(USet* set, UChar32 c);
 
+/**
+ * Adds each of the characters in this string to the set. Thus "ch" => {"c", "h"}
+ * If this set already any particular character, it has no effect on that character.
+ * @param s the source string
+ * @draft ICU 2.4
+ */
+U_CAPI void U_EXPORT2
+uset_addAll(USet* set, const USet *additionalSet);
+
 /**
  * Adds the given range of characters to the given USet.  After this call,
  * uset_contains(set, start, end) will return TRUE.
diff --git a/icu4c/source/common/uset.cpp b/icu4c/source/common/uset.cpp
index 8efc5313ac6..4b405f966be 100644
--- a/icu4c/source/common/uset.cpp
+++ b/icu4c/source/common/uset.cpp
@@ -88,6 +88,11 @@ uset_toPattern(const USet* set,
     return pat.extract(result, resultCapacity, *ec);
 }
 
+U_CAPI void U_EXPORT2
+uset_addAll(USet* set, const USet *additionalSet) {
+    ((UnicodeSet*) set)->addAll(*((const UnicodeSet*)additionalSet));
+}
+
 U_CAPI void U_EXPORT2
 uset_add(USet* set, UChar32 c) {
     ((UnicodeSet*) set)->add(c);