Remove redundancies

clipperhouse · Aug 30, 2024 · eee045c · eee045c
1 parent 733dd06
commit eee045c
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 19 deletions.
diff --git a/phrases/splitfunc.go b/phrases/splitfunc.go
@@ -137,7 +137,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7
-		if current.is(_AHLetter) && lastExIgnore.is(_MidLetter|_MidNumLetQ|_Ignore) && lastLastExIgnore.is(_AHLetter) {
+		if current.is(_AHLetter) && lastExIgnore.is(_MidLetter|_MidNumLetQ) && lastLastExIgnore.is(_AHLetter) {
 			pos += w
 			continue
 		}
@@ -149,7 +149,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7b
-		if current.is(_DoubleQuote) && lastExIgnore.is(_HebrewLetter|_Ignore) {
+		if current.is(_DoubleQuote) && lastExIgnore.is(_HebrewLetter) {
 			found, more := subsequent(_HebrewLetter, data[pos+w:], atEOF)
 
 			if more {
@@ -164,7 +164,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7c
-		if current.is(_HebrewLetter) && lastExIgnore.is(_DoubleQuote|_Ignore) && lastLastExIgnore.is(_HebrewLetter) {
+		if current.is(_HebrewLetter) && lastExIgnore.is(_DoubleQuote) && lastLastExIgnore.is(_HebrewLetter) {
 			pos += w
 			continue
 		}

diff --git a/sentences/splitfunc.go b/sentences/splitfunc.go
@@ -127,22 +127,14 @@ main:
 
 		// https://unicode.org/reports/tr29/#SB6
 		if current.is(_Numeric) && lastExIgnore.is(_ATerm) {
-			if previous(_ATerm, data[:pos]) {
-				pos += w
-				continue
-			}
+			pos += w
+			continue
 		}
 
-		// Optimization: determine if SB7 can possibly apply
-		maybeSB7 := current.is(_Upper) && last.is(_ATerm|_Ignore) && lastLastExIgnore.is(_Upper|_Lower)
-
 		// https://unicode.org/reports/tr29/#SB7
-		if maybeSB7 {
-			pi := previousIndex(_ATerm, data[:pos])
-			if pi >= 0 && previous(_Upper|_Lower, data[:pi]) {
-				pos += w
-				continue
-			}
+		if current.is(_Upper) && lastExIgnore.is(_ATerm) && lastLastExIgnore.is(_Upper|_Lower) {
+			pos += w
+			continue
 		}
 
 		// Optimization: determine if SB8 can possibly apply

diff --git a/words/splitfunc.go b/words/splitfunc.go
@@ -137,7 +137,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7
-		if current.is(_AHLetter) && lastExIgnore.is(_MidLetter|_MidNumLetQ|_Ignore) && lastLastExIgnore.is(_AHLetter) {
+		if current.is(_AHLetter) && lastExIgnore.is(_MidLetter|_MidNumLetQ) && lastLastExIgnore.is(_AHLetter) {
 			pos += w
 			continue
 		}
@@ -149,7 +149,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7b
-		if current.is(_DoubleQuote) && lastExIgnore.is(_HebrewLetter|_Ignore) {
+		if current.is(_DoubleQuote) && lastExIgnore.is(_HebrewLetter) {
 			found, more := subsequent(_HebrewLetter, data[pos+w:], atEOF)
 
 			if more {
@@ -164,7 +164,7 @@ func SplitFunc(data []byte, atEOF bool) (advance int, token []byte, err error) {
 		}
 
 		// https://unicode.org/reports/tr29/#WB7c
-		if current.is(_HebrewLetter) && lastExIgnore.is(_DoubleQuote|_Ignore) && lastLastExIgnore.is(_HebrewLetter) {
+		if current.is(_HebrewLetter) && lastExIgnore.is(_DoubleQuote) && lastLastExIgnore.is(_HebrewLetter) {
 			pos += w
 			continue
 		}