ICU-8972 parse ppucd.txt scx=Script_Extensions

X-SVN-Rev: 31146

diff --git a/icu4c/source/data/unidata/ppucd.txt b/icu4c/source/data/unidata/ppucd.txt
index e2dd3ed32e0eeef0a45c66fbc1f0072931f7b9c9..c1aa688cf04bf7a6f0234d2b3e932ad8a207a0a3 100644 (file)
--- a/icu4c/source/data/unidata/ppucd.txt
+++ b/icu4c/source/data/unidata/ppucd.txt
@@ -885,7 +885,7 @@ value;WB;NL;Newline
 value;WB;NU;Numeric
 value;WB;XX;Other
 
-defaults;0000..10FFFF;age=NA;bc=L;blk=NB;bmg=<code point>;cf=<code point>;dm=<code point>;dt=None;ea=N;FC_NFKC=<code point>;gc=Cn;GCB=XX;hst=NA;InMC=NA;InSC=Other;jg=No_Joining_Group;jt=U;lb=XX;lc=<slc>;NFC_QC=Y;NFD_QC=Y;NFKC_CF=<code point>;NFKC_QC=Y;NFKD_QC=Y;nt=None;SB=XX;sc=Zzzz;scf=<code point>;scx=Script_Extensions;slc=<code point>;stc=<code point>;suc=<code point>;tc=<stc>;uc=<suc>;WB=XX
+defaults;0000..10FFFF;age=NA;bc=L;blk=NB;bmg=<code point>;cf=<code point>;dm=<code point>;dt=None;ea=N;FC_NFKC=<code point>;gc=Cn;GCB=XX;hst=NA;InMC=NA;InSC=Other;jg=No_Joining_Group;jt=U;lb=XX;lc=<slc>;NFC_QC=Y;NFD_QC=Y;NFKC_CF=<code point>;NFKC_QC=Y;NFKD_QC=Y;nt=None;SB=XX;sc=Zzzz;scf=<code point>;scx=<script>;slc=<code point>;stc=<code point>;suc=<code point>;tc=<stc>;uc=<suc>;WB=XX
 
 block;0000..007F;age=1.1;blk=ASCII;ea=Na;gc=Cc;Gr_Base;lb=AL;sc=Zyyy
 # 0000..007F C0 Controls and Basic Latin (Basic Latin)
@@ -13621,7 +13621,7 @@ cp;31E0;na=CJK STROKE HXWG
 cp;31E1;na=CJK STROKE HZZZG
 cp;31E2;na=CJK STROKE PG
 cp;31E3;na=CJK STROKE Q
-cp;31E4..31EF;age=NA;bc=L;ea=N;gc=Cn;-Gr_Base;lb=XX;sc=Zzzz;scx=Script_Extensions
+cp;31E4..31EF;age=NA;bc=L;ea=N;gc=Cn;-Gr_Base;lb=XX;sc=Zzzz;scx=<script>
 
 block;31F0..31FF;age=3.2;Alpha;blk=Katakana_Ext;ea=W;gc=Lo;Gr_Base;IDC;IDS;lb=CJ;SB=LE;sc=Kana;WB=KA;XIDC;XIDS
 # 31F0..31FF Katakana Phonetic Extensions
@@ -16391,7 +16391,7 @@ cp;A837;gc=So;na=NORTH INDIC PLACEHOLDER MARK
 cp;A838;bc=ET;gc=Sc;lb=PO;na=NORTH INDIC RUPEE MARK
 # Miscellaneous sign
 cp;A839;bc=ET;gc=So;na=NORTH INDIC QUANTITY MARK
-cp;A83A..A83F;age=NA;-Gr_Base;lb=XX;sc=Zzzz;scx=Script_Extensions
+cp;A83A..A83F;age=NA;-Gr_Base;lb=XX;sc=Zzzz;scx=<script>
 
 block;A840..A87F;age=5.0;Alpha;blk=Phags_Pa;gc=Lo;Gr_Base;IDC;IDS;InSC=Consonant;lb=AL;SB=LE;sc=Phag;WB=LE;XIDC;XIDS
 # A840..A87F Phags-pa
@@ -19639,7 +19639,7 @@ block;10100..1013F;age=4.0;blk=Aegean_Numbers;gc=No;Gr_Base;lb=AL;nt=Nu;sc=Zyyy;
 cp;10100;gc=Po;lb=BA;na=AEGEAN WORD SEPARATOR LINE;nt=None
 cp;10101;bc=ON;gc=Po;lb=BA;na=AEGEAN WORD SEPARATOR DOT;nt=None
 cp;10102;gc=Po;lb=BA;na=AEGEAN CHECK MARK;nt=None
-cp;10103..10106;age=NA;gc=Cn;-Gr_Base;lb=XX;nt=None;sc=Zzzz;scx=Script_Extensions
+cp;10103..10106;age=NA;gc=Cn;-Gr_Base;lb=XX;nt=None;sc=Zzzz;scx=<script>
 # Numbers
 cp;10107;na=AEGEAN NUMBER ONE;nv=1
 cp;10108;na=AEGEAN NUMBER TWO;nv=2
@@ -19686,7 +19686,7 @@ cp;10130;na=AEGEAN NUMBER SIXTY THOUSAND;nv=60000
 cp;10131;na=AEGEAN NUMBER SEVENTY THOUSAND;nv=70000
 cp;10132;na=AEGEAN NUMBER EIGHTY THOUSAND;nv=80000
 cp;10133;na=AEGEAN NUMBER NINETY THOUSAND;nv=90000
-cp;10134..10136;age=NA;gc=Cn;-Gr_Base;lb=XX;nt=None;sc=Zzzz;scx=Script_Extensions
+cp;10134..10136;age=NA;gc=Cn;-Gr_Base;lb=XX;nt=None;sc=Zzzz;scx=<script>
 # Measures
 cp;10137;gc=So;na=AEGEAN WEIGHT BASE UNIT;nt=None
 cp;10138;gc=So;na=AEGEAN WEIGHT FIRST SUBUNIT;nt=None

diff --git a/icu4c/source/tools/toolutil/ppucd.cpp b/icu4c/source/tools/toolutil/ppucd.cpp
index dd59afdbd4e9b201b9b6391f3baea9ece641cce3..320e38add4ae95749be0863b4912ed81c2078277 100644 (file)
--- a/icu4c/source/tools/toolutil/ppucd.cpp
+++ b/icu4c/source/tools/toolutil/ppucd.cpp
@@ -339,9 +339,33 @@ PreparsedUCD::parseProperty(UniProps &props, const char *field, UnicodeSet &newV
         } else {
             props.intProps[prop-UCHAR_INT_START]=value;
         }
-    } else if(*v=='<' && lineType==DEFAULTS_LINE) {
-        // Ignore default values like <code point>.
-        return TRUE;
+    } else if(*v=='<') {
+        // Do not parse default values like <code point>, just set null values.
+        switch(prop) {
+        case UCHAR_BIDI_MIRRORING_GLYPH:
+            props.bmg=U_SENTINEL;
+            break;
+        case UCHAR_CASE_FOLDING:
+            props.cf.remove();
+            break;
+        case UCHAR_SCRIPT_EXTENSIONS:
+            props.scx.clear();
+            break;
+        case UCHAR_LOWERCASE_MAPPING:
+        case UCHAR_SIMPLE_CASE_FOLDING:
+        case UCHAR_SIMPLE_LOWERCASE_MAPPING:
+        case UCHAR_SIMPLE_TITLECASE_MAPPING:
+        case UCHAR_SIMPLE_UPPERCASE_MAPPING:
+        case UCHAR_TITLECASE_MAPPING:
+        case UCHAR_UPPERCASE_MAPPING:
+            // Ignore unhandled properties.
+            break;
+        default:
+            fprintf(stderr,
+                    "error in preparsed UCD: '%s' is not a valid default value on line %ld\n",
+                    field, (long)lineNumber);
+            errorCode=U_PARSE_ERROR;
+        }
     } else {
         char c;
         switch(prop) {
@@ -369,6 +393,9 @@ PreparsedUCD::parseProperty(UniProps &props, const char *field, UnicodeSet &newV
         case PPUCD_NAME_ALIAS:
             props.nameAlias=v;
             break;
+        case UCHAR_SCRIPT_EXTENSIONS:
+            parseScriptExtensions(v, props.scx, errorCode);
+            break;
         default:
             // Ignore unhandled properties.
             return TRUE;
@@ -450,4 +477,46 @@ PreparsedUCD::parseString(const char *s, UnicodeString &uni, UErrorCode &errorCo
     }
 }
 
+void
+PreparsedUCD::parseScriptExtensions(const char *s, UnicodeSet &scx, UErrorCode &errorCode) {
+    if(U_FAILURE(errorCode)) { return; }
+    scx.clear();
+    CharString scString;
+    for(;;) {
+        const char *scs;
+        const char *scLimit=strchr(s, ' ');
+        if(scLimit!=NULL) {
+            scs=scString.clear().append(s, (int32_t)(scLimit-s), errorCode).data();
+            if(U_FAILURE(errorCode)) { return; }
+        } else {
+            scs=s;
+        }
+        int32_t script=pnames->getPropertyValueEnum(UCHAR_SCRIPT, scs);
+        if(script==UCHAR_INVALID_CODE) {
+            fprintf(stderr,
+                    "error in preparsed UCD: '%s' is not a valid script code on line %ld\n",
+                    scs, (long)lineNumber);
+            errorCode=U_PARSE_ERROR;
+            return;
+        } else if(scx.contains(script)) {
+            fprintf(stderr,
+                    "error in preparsed UCD: scx has duplicate '%s' codes on line %ld\n",
+                    scs, (long)lineNumber);
+            errorCode=U_PARSE_ERROR;
+            return;
+        } else {
+            scx.add(script);
+        }
+        if(scLimit!=NULL) {
+            s=scLimit+1;
+        } else {
+            break;
+        }
+    }
+    if(scx.isEmpty()) {
+        fprintf(stderr, "error in preparsed UCD: empty scx= on line %ld\n", (long)lineNumber);
+        errorCode=U_PARSE_ERROR;
+    }
+}
+
 U_NAMESPACE_END

diff --git a/icu4c/source/tools/toolutil/ppucd.h b/icu4c/source/tools/toolutil/ppucd.h
index 371a8e94eea96179b3f081c0e58ddb03c5d8d9fc..f7d22c70909117d56eb54dfcf69b877cbf1af0cd 100644 (file)
--- a/icu4c/source/tools/toolutil/ppucd.h
+++ b/icu4c/source/tools/toolutil/ppucd.h
@@ -52,6 +52,7 @@ struct U_TOOLUTIL_API UniProps {
     const char *name;
     const char *nameAlias;
     UnicodeString cf;
+    UnicodeSet scx;
 };
 
 class U_TOOLUTIL_API PreparsedUCD {
@@ -145,6 +146,7 @@ private:
     UChar32 parseCodePoint(const char *s, UErrorCode &errorCode);
     UBool parseCodePointRange(const char *s, UChar32 &start, UChar32 &end, UErrorCode &errorCode);
     void parseString(const char *s, UnicodeString &uni, UErrorCode &errorCode);
+    void parseScriptExtensions(const char *s, UnicodeSet &scx, UErrorCode &errorCode);
 
     static const int32_t kNumLineBuffers=3;