Moving CharacterSet class into own header so can be used by more

lexers.
Updated LexCPP to match changes.
Added use of CharacterSet to LexHTML.

3 files changed, 83 insertions, 58 deletions

diff --git a/src/CharacterSet.h b/src/CharacterSet.h
new file mode 100644
index 000000000..4e8ffbdf6
--- /dev/null
+++ b/src/CharacterSet.h
@@ -0,0 +1,58 @@
+// Scintilla source code edit control
+/** @file CharacterSet.h
+ ** Encapsulates a set of characters. Used to test if a character is within a set.
+ **/
+// Copyright 2007 by Neil Hodgson <neilh@scintilla.org>
+// The License.txt file describes the conditions under which this software may be distributed.
+
+class CharacterSet {
+	int size;
+	bool valueAfter;
+	bool *bset;
+public:
+	enum setBase {
+		setNone=0,
+		setLower=1,
+		setUpper=2,
+		setDigits=4,
+		setAlpha=setLower|setUpper,
+		setAlphaNum=setAlpha|setDigits
+	};
+	CharacterSet(setBase base=setNone, const char *initialSet="", int size_=0x80, bool valueAfter_=false) {
+		size = size_;
+		valueAfter = valueAfter_;
+		bset = new bool[size];
+		for (int i=0; i < size; i++) {
+			bset[i] = false;
+		}
+		AddString(initialSet);
+		if (base & setLower)
+			AddString("abcdefghijklmnopqrstuvwxyz");
+		if (base & setUpper)
+			AddString("ABCDEFGHIJKLMNOPQRSTUVWXYZ");
+		if (base & setDigits)
+			AddString("0123456789");
+	}
+	~CharacterSet() {
+		delete []bset;
+		bset = 0;
+		size = 0;
+	}
+	void Add(int val) {
+		PLATFORM_ASSERT(val >= 0);
+		PLATFORM_ASSERT(val < size);
+		bset[val] = true;
+	}
+	void AddString(const char *CharacterSet) {
+		for (const char *cp=CharacterSet; *cp; cp++) {
+			int val = static_cast<unsigned char>(*cp);
+			PLATFORM_ASSERT(val >= 0);
+			PLATFORM_ASSERT(val < size);
+			bset[val] = true;
+		}
+	}
+	bool Contains(int val) const {
+		PLATFORM_ASSERT(val >= 0);
+		return (val < size) ? bset[val] : valueAfter;
+	}
+};
diff --git a/src/LexCPP.cxx b/src/LexCPP.cxx
index bc3177cb3..8c00acb7b 100644
--- a/src/LexCPP.cxx
+++ b/src/LexCPP.cxx
@@ -19,50 +19,12 @@
 #include "KeyWords.h"
 #include "Scintilla.h"
 #include "SciLexer.h"
+#include "CharacterSet.h"
 
 #ifdef SCI_NAMESPACE
 using namespace Scintilla;
 #endif
 
-#define SET_LOWER "abcdefghijklmnopqrstuvwxyz"
-#define SET_UPPER "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
-#define SET_DIGITS "0123456789"
-
-class SetOfCharacters {
-	int size;
-	bool valueAfter;
-	bool *bset;
-public:
-	SetOfCharacters(const char *setOfCharacters, int size_=0x80, bool valueAfter_=false) {
-		size = size_;
-		valueAfter = valueAfter_;
-		bset = new bool[size];
-		for (int i=0; i < size; i++) {
-			bset[i] = false;
-		}
-		for (const char *cp=setOfCharacters; *cp; cp++) {
-			int val = static_cast<unsigned char>(*cp);
-			PLATFORM_ASSERT(val >= 0);
-			PLATFORM_ASSERT(val < size);
-			bset[val] = true;
-		}
-	}
-	~SetOfCharacters() {
-		delete []bset;
-		bset = 0;
-		size = 0;
-	}
-	void Add(int val) {
-		PLATFORM_ASSERT(val >= 0);
-		PLATFORM_ASSERT(val < size);
-		bset[val] = true;
-	}
-	bool Contains(int val) {
-		PLATFORM_ASSERT(val >= 0);
-		return (val < size) ? bset[val] : valueAfter;
-	}
-};
-
 static bool IsSpaceEquiv(int state) {
 	return (state <= SCE_C_COMMENTDOC) ||
 		// including SCE_C_DEFAULT, SCE_C_COMMENT, SCE_C_COMMENTLINE
@@ -80,12 +42,12 @@ static void ColouriseCppDoc(unsigned int startPos, int length, int initStyle, Wo
 
 	bool stylingWithinPreprocessor = styler.GetPropertyInt("styling.within.preprocessor") != 0;
 
-	SetOfCharacters setOKBeforeRE("(=,");
+	CharacterSet setOKBeforeRE(CharacterSet::setNone, "(=,");
 
-	SetOfCharacters setDoxygen("$@\\&<>#{}[]" SET_LOWER);
+	CharacterSet setDoxygen(CharacterSet::setLower, "$@\\&<>#{}[]");
 
-	SetOfCharacters setWordStart("_" SET_LOWER SET_UPPER, 0x80, true);
-	SetOfCharacters setWord("._" SET_LOWER SET_UPPER SET_DIGITS, 0x80, true);
+	CharacterSet setWordStart(CharacterSet::setAlpha, "_", 0x80, true);
+	CharacterSet setWord(CharacterSet::setAlphaNum, "._", 0x80, true);
 	if (styler.GetPropertyInt("lexer.cpp.allow.dollars", 1) != 0) {
 		setWordStart.Add('$');
 		setWord.Add('$');
diff --git a/src/LexHTML.cxx b/src/LexHTML.cxx
index 877f0881a..eebf5ca1d 100644
--- a/src/LexHTML.cxx
+++ b/src/LexHTML.cxx
@@ -19,6 +19,7 @@
 #include "KeyWords.h"
 #include "Scintilla.h"
 #include "SciLexer.h"
+#include "CharacterSet.h"
 
 #ifdef SCI_NAMESPACE
 using namespace Scintilla;
@@ -106,16 +107,16 @@ static script_type ScriptOfState(int state) {
 }
 
 static int statePrintForState(int state, script_mode inScriptType) {
-	int StateToPrint;
-
-	if ((state >= SCE_HP_START) && (state <= SCE_HP_IDENTIFIER)) {
-		StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_PYTHON);
-	} else if ((state >= SCE_HB_START) && (state <= SCE_HB_STRINGEOL)) {
-		StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_VBS);
-	} else if ((state >= SCE_HJ_START) && (state <= SCE_HJ_REGEX)) {
-		StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_JS);
-	} else {
-		StateToPrint = state;
+	int StateToPrint = state;
+
+	if (state >= SCE_HJ_START) {
+		if ((state >= SCE_HP_START) && (state <= SCE_HP_IDENTIFIER)) {
+			StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_PYTHON);
+		} else if ((state >= SCE_HB_START) && (state <= SCE_HB_STRINGEOL)) {
+			StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_VBS);
+		} else if ((state >= SCE_HJ_START) && (state <= SCE_HJ_REGEX)) {
+			StateToPrint = state + ((inScriptType == eNonHtmlScript) ? 0 : SCE_HA_JS);
+		}
 	}
 
 	return StateToPrint;
@@ -521,6 +522,10 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 	const bool foldCompact = styler.GetPropertyInt("fold.compact", 1) != 0;
 	const bool caseSensitive = styler.GetPropertyInt("html.tags.case.sensitive", 0) != 0;
 
+	const CharacterSet setHTMLWord(CharacterSet::setAlphaNum, ".-_:!#", 0x80, true);
+	const CharacterSet setTagContinue(CharacterSet::setAlphaNum, ".-_:!#[", 0x80, true);
+	const CharacterSet setAttributeContinue(CharacterSet::setAlphaNum, ".-_:!#/", 0x80, true);
+
 	int levelPrev = styler.LevelAt(lineCurrent) & SC_FOLDLEVELNUMBERMASK;
 	int levelCurrent = levelPrev;
 	int visibleChars = 0;
@@ -932,7 +937,7 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 				}
 				// find the length of the word
 				int size = 1;
-				while (ishtmlwordchar(styler.SafeGetCharAt(i + size)))
+				while (setHTMLWord.Contains(styler.SafeGetCharAt(i + size)))
 					size++;
 				styler.ColourTo(i + size - 1, StateToPrint);
 				i += size - 1;
@@ -1020,7 +1025,7 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 			}
 			break;
 		case SCE_H_TAGUNKNOWN:
-			if (!ishtmlwordchar(ch) && !((ch == '/') && (chPrev == '<')) && ch != '[') {
+			if (!setTagContinue.Contains(ch) && !((ch == '/') && (chPrev == '<'))) {
 				int eClass = classifyTagHTML(styler.GetStartSegment(),
 					i - 1, keywords, styler, tagDontFold, caseSensitive, isXml);
 				if (eClass == SCE_H_SCRIPT) {
@@ -1072,7 +1077,7 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 			}
 			break;
 		case SCE_H_ATTRIBUTE:
-			if (!ishtmlwordchar(ch) && ch != '/' && ch != '-') {
+			if (!setAttributeContinue.Contains(ch)) {
 				if (inScriptType == eNonHtmlScript) {
 					int scriptLanguagePrev = scriptLanguage;
 					clientScript = segIsScriptingIndicator(styler, styler.GetStartSegment(), i - 1, scriptLanguage);
@@ -1145,7 +1150,7 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 				i++;
 				ch = chNext;
 				state = SCE_H_DEFAULT;
-			} else if (ishtmlwordchar(ch)) {
+			} else if (setHTMLWord.Contains(ch)) {
 				styler.ColourTo(i - 1, StateToPrint);
 				state = SCE_H_ATTRIBUTE;
 			}
@@ -1169,7 +1174,7 @@ static void ColouriseHyperTextDoc(unsigned int startPos, int length, int initSty
 			}
 			break;
 		case SCE_H_VALUE:
-			if (!ishtmlwordchar(ch)) {
+			if (!setHTMLWord.Contains(ch)) {
 				if (ch == '\"' && chPrev == '=') {
 					// Should really test for being first character
 					state = SCE_H_DOUBLESTRING;