lexhtml.cxx.svn-base

Notepad++ is a generic source code editor (it tries to be anyway) and Notepad replacement written in

// Scintilla source code edit control
/** @file LexHTML.cxx
 ** Lexer for HTML.
 **/
// Copyright 1998-2005 by Neil Hodgson <[e-mail]>
// The License.txt file describes the conditions under which this software may be distributed.

#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <stdio.h>
#include <stdarg.h>

#include "Platform.h"

#include "PropSet.h"
#include "Accessor.h"
#include "StyleContext.h"
#include "KeyWords.h"
#include "Scintilla.h"
#include "SciLexer.h"

#define SCE_HA_JS (SCE_HJA_START - SCE_HJ_START)
#define SCE_HA_VBS (SCE_HBA_START - SCE_HB_START)
#define SCE_HA_PYTHON (SCE_HPA_START - SCE_HP_START)

enum script_type { eScriptNone = 0, eScriptJS, eScriptVBS, eScriptPython, eScriptPHP, eScriptXML, eScriptSGML, eScriptSGMLblock };
enum script_mode { eHtml = 0, eNonHtmlScript, eNonHtmlPreProc, eNonHtmlScriptPreProc };

static inline bool IsAWordChar(const int ch) {
	return (ch < 0x80) && (isalnum(ch) || ch == '.' || ch == '_');
}

static inline bool IsAWordStart(const int ch) {
	return (ch < 0x80) && (isalnum(ch) || ch == '_');
}

static inline int MakeLowerCase(int ch) {
	if (ch < 'A' || ch > 'Z')
		return ch;
	else
		return ch - 'A' + 'a';
}

static void GetTextSegment(Accessor &styler, unsigned int start, unsigned int end, char *s, size_t len) {
	size_t i = 0;
	for (; (i < end - start + 1) && (i < len-1); i++) {
		s[i] = static_cast<char>(MakeLowerCase(styler[start + i]));
	}
	s[i] = '\0';
}

static script_type segIsScriptingIndicator(Accessor &styler, unsigned int start, unsigned int end, script_type prevValue) {
	char s[100];
	GetTextSegment(styler, start, end, s, sizeof(s));
	//Platform::DebugPrintf("Scripting indicator [%s]\n", s);
	//Don
	//if (strstr(s, "src"))	// External script
		//return eScriptNone;
	//nod
	if (strstr(s, "vbs"))
		return eScriptVBS;
	if (strstr(s, "pyth"))
		return eScriptPython;
	if (strstr(s, "javas"))
		return eScriptJS;
	if (strstr(s, "jscr"))
		return eScriptJS;
	if (strstr(s, "php"))
		return eScriptPHP;
	if (strstr(s, "xml"))
		return eScriptXML;

	return prevValue;
}

static int PrintScriptingIndicatorOffset(Accessor &styler, unsigned int start, unsigned int end) {
	int iResult = 0;
	char s[100];
	GetTextSegment(styler, start, end, s, sizeof(s));
	if (0 == strncmp(s, "php", 3)) {
		iResult = 3;
	}

	return iResult;
}

static script_type ScriptOfState(int state) {
	if ((state >= SCE_HP_START) && (state <= SCE_HP_IDENTIFIER)) {
		return eScriptPython;
	} else if ((state >= SCE_HB_START) && (state <= SCE_HB_STRINGEOL)) {
		return eScriptVBS;
	} else if ((state >= SCE_HJ_START) && (state <= SCE_HJ_REGEX)) {
		return eScriptJS;
	} else if ((state >= SCE_HPHP_DEFAULT) && (state <= SCE_HPHP_COMMENTLINE)) {
		return eScriptPHP;
	} else if ((state >= SCE_H_SGML_DEFAULT) && (state < SCE_H_SGML_BLOCK_DEFAULT)) {
		return eScriptSGML;
	} else if (state == SCE_H_SGML_BLOCK_DEFAULT) {
		return eScriptSGMLblock;
	} else {
		return eScriptNone;
	}
}

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;
	}

	return StateToPrint;
}

static int stateForPrintState(int StateToPrint) {
	int state;

	if ((StateToPrint >= SCE_HPA_START) && (StateToPrint <= SCE_HPA_IDENTIFIER)) {
		state = StateToPrint - SCE_HA_PYTHON;
	} else if ((StateToPrint >= SCE_HBA_START) && (StateToPrint <= SCE_HBA_STRINGEOL)) {
		state = StateToPrint - SCE_HA_VBS;
	} else if ((StateToPrint >= SCE_HJA_START) && (StateToPrint <= SCE_HJA_REGEX)) {
		state = StateToPrint - SCE_HA_JS;
	} else {
		state = StateToPrint;
	}

	return state;
}

static inline bool IsNumber(unsigned int start, Accessor &styler) {
	return IsADigit(styler[start]) || (styler[start] == '.') ||
	       (styler[start] == '-') || (styler[start] == '#');
}

static inline bool isStringState(int state) {
	bool bResult;

	switch (state) {
	case SCE_HJ_DOUBLESTRING:
	case SCE_HJ_SINGLESTRING:
	case SCE_HJA_DOUBLESTRING:
	case SCE_HJA_SINGLESTRING:
	case SCE_HB_STRING:
	case SCE_HBA_STRING:
	case SCE_HP_STRING:
	case SCE_HP_CHARACTER:
	case SCE_HP_TRIPLE:
	case SCE_HP_TRIPLEDOUBLE:
	case SCE_HPA_STRING:
	case SCE_HPA_CHARACTER:
	case SCE_HPA_TRIPLE:
	case SCE_HPA_TRIPLEDOUBLE:
	case SCE_HPHP_HSTRING:
	case SCE_HPHP_SIMPLESTRING:
	case SCE_HPHP_HSTRING_VARIABLE:
	case SCE_HPHP_COMPLEX_VARIABLE:
		bResult = true;
		break;
	default :
		bResult = false;
		break;
	}
	return bResult;
}

static inline bool stateAllowsTermination(int state) {
	bool allowTermination = !isStringState(state);
	if (allowTermination) {
		switch (state) {
		case SCE_HB_COMMENTLINE:
		case SCE_HPHP_COMMENT:
		case SCE_HP_COMMENTLINE:
		case SCE_HPA_COMMENTLINE:
			allowTermination = false;
		}
	}
	return allowTermination;
}

// not really well done, since it's only comments that should lex the %> and <%
static inline bool isCommentASPState(int state) {
	bool bResult;

	switch (state) {
	case SCE_HJ_COMMENT:
	case SCE_HJ_COMMENTLINE:
	case SCE_HJ_COMMENTDOC:
	case SCE_HB_COMMENTLINE:
	case SCE_HP_COMMENTLINE:
	case SCE_HPHP_COMMENT:
	case SCE_HPHP_COMMENTLINE:
		bResult = true;
		break;
	default :
		bResult = false;
		break;
	}
	return bResult;
}

static void classifyAttribHTML(unsigned int start, unsigned int end, WordList &keywords, Accessor &styler) {
	bool wordIsNumber = IsNumber(start, styler);
	char chAttr = SCE_H_ATTRIBUTEUNKNOWN;
	if (wordIsNumber) {
		chAttr = SCE_H_NUMBER;
	} else {
		char s[100];
		GetTextSegment(styler, start, end, s, sizeof(s));
		if (keywords.InList(s))
			chAttr = SCE_H_ATTRIBUTE;
	}
	if ((chAttr == SCE_H_ATTRIBUTEUNKNOWN) && !keywords)
		// No keywords -> all are known
		chAttr = SCE_H_ATTRIBUTE;
	styler.ColourTo(end, chAttr);
}

static int classifyTagHTML(unsigned int start, unsigned int end,
                           WordList &keywords, Accessor &styler, bool &tagDontFold,
			   bool caseSensitive) {
	char s[30 + 2];
	// Copy after the '<'
	unsigned int i = 0;
	for (unsigned int cPos = start; cPos <= end && i < 30; cPos++) {
		char ch = styler[cPos];
		if ((ch != '<') && (ch != '/')) {
			s[i++] = caseSensitive ? ch : static_cast<char>(MakeLowerCase(ch));
		}
	}

	//The following is only a quick hack, to see if this whole thing would work
	//we first need the tagname with a trailing space...
	s[i] = ' ';
	s[i+1] = '\0';

	//...to find it in the list of no-container-tags
	// (There are many more. We will need a keywordlist in the property file for this)
	tagDontFold = (NULL != strstr("meta link img area br hr input ",s));

	//now we can remove the trailing space
	s[i] = '\0';

	bool isScript = false;
	char chAttr = SCE_H_TAGUNKNOWN;
	if (s[0] == '!') {
		chAttr = SCE_H_SGML_DEFAULT;
	} else if (s[0] == '/') {	// Closing tag
		if (keywords.InList(s + 1))
			chAttr = SCE_H_TAG;
	} else {
		if (keywords.InList(s)) {
			chAttr = SCE_H_TAG;
			isScript = 0 == strcmp(s, "script");
		}
	}
	if ((chAttr == SCE_H_TAGUNKNOWN) && !keywords) {
		// No keywords -> all are known
		chAttr = SCE_H_TAG;
		isScript = 0 == strcmp(s, "script");
	}
	styler.ColourTo(end, chAttr);
	return isScript ? SCE_H_SCRIPT : chAttr;
}

static void classifyWordHTJS(unsigned int start, unsigned int end,
                             WordList &keywords, Accessor &styler, script_mode inScriptType) {
	char chAttr = SCE_HJ_WORD;
	bool wordIsNumber = IsADigit(styler[start]) || (styler[start] == '.');
	if (wordIsNumber)
		chAttr = SCE_HJ_NUMBER;
	else {
		char s[30 + 1];
		unsigned int i = 0;
		for (; i < end - start + 1 && i < 30; i++) {
			s[i] = styler[start + i];
		}
		s[i] = '\0';
		if (keywords.InList(s))
			chAttr = SCE_HJ_KEYWORD;
	}
	styler.ColourTo(end, statePrintForState(chAttr, inScriptType));
}

static int classifyWordHTVB(unsigned int start, unsigned int end, WordList &keywords, Accessor &styler, script_mode inScriptType) {
	char chAttr = SCE_HB_IDENTIFIER;
	bool wordIsNumber = IsADigit(styler[start]) || (styler[start] == '.');
	if (wordIsNumber)
		chAttr = SCE_HB_NUMBER;
	else {
		char s[100];
		GetTextSegment(styler, start, end, s, sizeof(s));
		if (keywords.InList(s)) {
			chAttr = SCE_HB_WORD;
			if (strcmp(s, "rem") == 0)
				chAttr = SCE_HB_COMMENTLINE;
		}
	}
	styler.ColourTo(end, statePrintForState(chAttr, inScriptType));
	if (chAttr == SCE_HB_COMMENTLINE)
		return SCE_HB_COMMENTLINE;
	else
		return SCE_HB_DEFAULT;
}

static void classifyWordHTPy(unsigned int start, unsigned int end, WordList &keywords, Accessor &styler, char *prevWord, script_mode inScriptType) {
	bool wordIsNumber = IsADigit(styler[start]);
	char s[30 + 1];
	unsigned int i = 0;
	for (; i < end - start + 1 && i < 30; i++) {
		s[i] = styler[start + i];
	}
	s[i] = '\0';
	char chAttr = SCE_HP_IDENTIFIER;
	if (0 == strcmp(prevWord, "class"))
		chAttr = SCE_HP_CLASSNAME;
	else if (0 == strcmp(prevWord, "def"))
		chAttr = SCE_HP_DEFNAME;
	else if (wordIsNumber)
		chAttr = SCE_HP_NUMBER;
	else if (keywords.InList(s))
		chAttr = SCE_HP_WORD;
	styler.ColourTo(end, statePrintForState(chAttr, inScriptType));
	strcpy(prevWord, s);
}

// Update the word colour to default or keyword
// Called when in a PHP word
static void classifyWordHTPHP(unsigned int start, unsigned int end, WordList &keywords, Accessor &styler) {
	char chAttr = SCE_HPHP_DEFAULT;
	bool wordIsNumber = IsADigit(styler[start]) || (styler[start] == '.' && start+1 <= end && IsADigit(styler[start+1]));
	if (wordIsNumber)
		chAttr = SCE_HPHP_NUMBER;
	else {
		char s[100];
		GetTextSegment(styler, start, end, s, sizeof(s));
		if (keywords.InList(s))
			chAttr = SCE_HPHP_WORD;
	}
	styler.ColourTo(end, chAttr);
}

static bool isWordHSGML(unsigned int start, unsigned int end, WordList &keywords, Accessor &styler) {
	char s[30 + 1];
	unsigned int i = 0;
	for (; i < end - start + 1 && i < 30; i++) {
		s[i] = styler[start + i];
	}
	s[i] = '\0';
	return keywords.InList(s);
}

static bool isWordCdata(unsigned int start, unsigned int end, Accessor &styler) {
	char s[30 + 1];
	unsigned int i = 0;
	for (; i < end - start + 1 && i < 30; i++) {
		s[i] = styler[start + i];
	}
	s[i] = '\0';
	return (0 == strcmp(s, "[CDATA["));
}

// Return the first state to reach when entering a scripting language
static int StateForScript(script_type scriptLanguage) {
	int Result;
	switch (scriptLanguage) {
	// Modif by Don
	/*
	case eScriptVBS:
		Result = SCE_HB_START;
		break;
	*/
	case eScriptJS:
		Result = SCE_HJ_START;
		break;
	// Fidom by Don
	case eScriptPython:
		Result = SCE_HP_START;
		break;
	case eScriptPHP:
		Result = SCE_HPHP_DEFAULT;
		break;
	case eScriptXML:
		Result = SCE_H_TAGUNKNOWN;
		break;
	case eScriptSGML:
		Result = SCE_H_SGML_DEFAULT;
		break;
	default :
		// Modif by Don
		//Result = SCE_HJ_START;
		Result = SCE_HB_START;
		// Fidom by Don
		break;
	}
	return Result;
}

static inline bool ishtmlwordchar(char ch) {
	return !isascii(ch) ||
		(isalnum(ch) || ch == '.' || ch == '-' || ch == '_' || ch == ':' || ch == '!' || ch == '#');
}

static inline bool issgmlwordchar(char ch) {
	return !isascii(ch) ||
		(isalnum(ch) || ch == '.' || ch == '_' || ch == ':' || ch == '!' || ch == '#' || ch == '[');