ICU-9954 rbbi forward iteration performance improvement. Ensure rules always make some progress; keep engine out of the fail-safe fall back path for rules that do not advance.

X-SVN-Rev: 40461

icu4c/source/data/brkitr/rules/char.txt

@@ -78,6 +78,9 @@ $Prepend [^$Control $CR $LF];
 ^$Prepend* $Regional_Indicator $Regional_Indicator / $Regional_Indicator;
 ^$Prepend* $Regional_Indicator $Regional_Indicator;
 
+# GB 999 Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/line.txt

@@ -335,6 +335,9 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
 
 ## -------------------------------------------------
 

icu4c/source/data/brkitr/rules/line_fi.txt

@@ -344,6 +344,10 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/line_loose.txt

@@ -347,6 +347,9 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
 
 ## -------------------------------------------------
 

icu4c/source/data/brkitr/rules/line_loose_cj.txt

@@ -361,6 +361,10 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/line_loose_fi.txt

@@ -346,6 +346,9 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
 
 ## -------------------------------------------------
 

icu4c/source/data/brkitr/rules/line_normal.txt

@@ -339,6 +339,10 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/line_normal_cj.txt

@@ -345,6 +345,10 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/line_normal_fi.txt

@@ -342,6 +342,10 @@ $RI $CM* $RI $CM* $ZWJ ($ID | $Extended_Pict | $EmojiNRK);
 # LB 30b Do not break between an Emoji Base and an Emoji Modifier
 $EB $CM* $EM;
 
+# LB 31 Break everywhere else.
+#       Match a single code point if no other rule applies.
+.;
+
 ## -------------------------------------------------
 
 !!safe_reverse;

icu4c/source/data/brkitr/rules/word.txt

@@ -194,6 +194,10 @@ $ExtendNumLetEx  $KatakanaEx     {400};    #  (13b)
 $HangulSyllable $HangulSyllable {200};
 $KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found 
 
+# Rule 999
+#     Match a single code point if no other rule applies.
+.;
+
 
 ## -------------------------------------------------
 

icu4c/source/data/brkitr/rules/word_POSIX.txt

@@ -194,6 +194,10 @@ $ExtendNumLetEx  $KatakanaEx     {400};    #  (13b)
 $HangulSyllable $HangulSyllable {200};
 $KanaKanji $KanaKanji {400}; # different rule status if both kana and kanji found
 
+# Rule 999
+#     Match a single code point if no other rule applies.
+.;
+
 
 ## -------------------------------------------------