Use WordList to lookup HTML void tag, avoid check void tag twice, i…

…ssue #957.
zufuliu · Jan 9, 2025 · bbe02d6 · bbe02d6
1 parent 1396609
commit bbe02d6
Show file tree

Hide file tree

Showing 9 changed files with 67 additions and 78 deletions.
diff --git a/scintilla/lexers/LexHTML.cxx b/scintilla/lexers/LexHTML.cxx
@@ -31,10 +31,11 @@ namespace {
 //KeywordIndex++Autogenerated -- start of section automatically generated
 enum {
 	KeywordIndex_Tag = 0,
-	KeywordIndex_JavaScript = 1,
-	KeywordIndex_VBScript = 2,
-	KeywordIndex_SGML = 3,
-	KeywordIndex_Attribute = 4,
+	KeywordIndex_VoidTag = 1,
+	KeywordIndex_JavaScript = 2,
+	KeywordIndex_VBScript = 3,
+	KeywordIndex_SGML = 4,
+	KeywordIndex_Attribute = 5,
 	MaxKeywordSize = 32,
 };
 //KeywordIndex--Autogenerated -- end of section automatically generated
@@ -185,50 +186,39 @@ bool isHTMLCustomElement(const char *tag, size_t length, bool dashColon) noexcep
 }
 
 int classifyTagHTML(Sci_PositionU end, LexerWordList keywordLists, LexAccessor &styler, bool &tagDontFold, bool isXml, bool allowScripts) {
-	char withSpace[126 + 2];
-	withSpace[0] = ' ';
-	withSpace[1] = '\0';
+	char tag[127 + 1];
 	// Copy after the '<' and stop before space
-	Sci_PositionU i = 1;
+	Sci_PositionU length = 0;
 	bool dashColon = false;
 	const Sci_PositionU start = styler.GetStartSegment();
-	for (Sci_PositionU cPos = start; cPos < end && i < sizeof(withSpace) - 2; cPos++) {
+	for (Sci_PositionU cPos = start; cPos < end && length < sizeof(tag) - 1; cPos++) {
 		const char ch = styler[cPos];
 		if (static_cast<unsigned char>(ch) <= ' ') {
 			break;
 		}
 		if ((ch != '<') && (ch != '/')) {
-			withSpace[i++] = isXml ? ch : MakeLowerCase(ch);
+			tag[length++] = isXml ? ch : MakeLowerCase(ch);
 			if (ch == ':' || ch == '-') {
 				dashColon = true;
 			}
 		}
 	}
 
-	//The following is only a quick hack, to see if this whole thing would work
-	//we first need the tagname with a trailing space...
-	withSpace[i] = ' ';
-	withSpace[i + 1] = '\0';
-
-	// if the current language is XML, I can fold any tag
-	// if the current language is HTML, I don't want to fold certain tags (input, meta, etc.)
-	//...to find it in the list of no-container-tags
-	tagDontFold = (!isXml) && IsHtmlVoidTag(withSpace);
-
-	//now we can remove the trailing space
-	withSpace[i] = '\0';
-	const char * const tag = withSpace + 1;
-
-	// No keywords -> all are known
+	tag[length] = '\0';
 	int chAttr = SCE_H_TAGUNKNOWN;
 	bool customElement = false;
 	if (tag[0] == '!') {
 		chAttr = SCE_H_SGML_DEFAULT;
-	} else if (isXml || keywordLists[KeywordIndex_Tag].InList(tag)) {
-		chAttr = SCE_H_TAG;
-	} else if (!isXml && isHTMLCustomElement(tag, i - 1, dashColon)) {
-		customElement = true;
+	} else if (isXml) {
 		chAttr = SCE_H_TAG;
+	} else {
+		tagDontFold = keywordLists[KeywordIndex_VoidTag].InList(tag);
+		if (tagDontFold || keywordLists[KeywordIndex_Tag].InList(tag)) {
+			chAttr = SCE_H_TAG;
+		} else if (isHTMLCustomElement(tag, length, dashColon)) {
+			customElement = true;
+			chAttr = SCE_H_TAG;
+		}
 	}
 	if (chAttr != SCE_H_TAGUNKNOWN) {
 		styler.ColorTo(end, chAttr);

diff --git a/scintilla/lexers/LexPHP.cxx b/scintilla/lexers/LexPHP.cxx
@@ -61,7 +61,8 @@ enum {
 	KeywordIndex_PredefinedVariable = 4,
 	KeywordIndex_MagicConstant = 5,
 	KeywordIndex_MagicMethod = 6,
-	KeywordIndex_JavaScript = 10,
+	KeywordIndex_VoidTag = 10,
+	KeywordIndex_JavaScript = 11,
 	MaxKeywordSize = 40,
 };
 //KeywordIndex--Autogenerated -- end of section automatically generated
@@ -206,7 +207,7 @@ struct PHPLexer {
 		return sc.Match(hereDocId.c_str()) && !IsIdentifierCharEx(sc.GetRelative(hereDocId.length()));
 	}
 
-	void ClassifyHtmlTag();
+	void ClassifyHtmlTag(LexerWordList keywordLists);
 	bool HandleBlockEnd(HtmlTextBlock block);
 
 	void HandlePHPTag();
@@ -219,29 +220,21 @@ struct PHPLexer {
 	bool ClassifyCssWord();
 };
 
-void PHPLexer::ClassifyHtmlTag() {
+void PHPLexer::ClassifyHtmlTag(LexerWordList keywordLists) {
 	if (sc.state == SCE_H_OTHER) {
 		sc.SetState((tagType == HtmlTagType::Question) ? SCE_H_QUESTION : SCE_H_TAG);
 	} else if (tagType == HtmlTagType::None) {
-		char s[16]{};
-		sc.GetCurrentLowered(s, sizeof(s) - 1);
-		char *p = s + 1;
+		char s[16];
+		sc.GetCurrentLowered(s, sizeof(s));
+		const char * const p = s + 1;
 		if (StrEqual(p, "script")) {
 			tagType = HtmlTagType::Script;
 		} else if (StrEqual(p, "style")) {
 			tagType = HtmlTagType::Style;
 		} else {
 			tagType = HtmlTagType::Normal;
-			const size_t length = sc.LengthCurrent();
-			if (length <= maxHtmlVoidTagLen + 2) {
-				s[length] = ' ';
-				if (*p != '/') {
-					--p;
-				}
-				*p = ' ';
-				if (IsHtmlVoidTag(p)) {
-					tagType = HtmlTagType::Void;
-				}
+			if (keywordLists[KeywordIndex_VoidTag].InList(p)) {
+				tagType = HtmlTagType::Void;
 			}
 		}
 	}
@@ -1001,7 +994,7 @@ void ColourisePHPDoc(Sci_PositionU startPos, Sci_Position lengthDoc, int initSty
 		case SCE_H_QUESTION:
 			if (sc.ch == '>' || sc.Match('/', '>') || IsASpace(sc.ch)
 				|| (lexer.tagType == HtmlTagType::Question && sc.Match('?', '>'))) {
-				lexer.ClassifyHtmlTag();
+				lexer.ClassifyHtmlTag(keywordLists);
 			}
 			break;
 
@@ -1052,7 +1045,7 @@ void ColourisePHPDoc(Sci_PositionU startPos, Sci_Position lengthDoc, int initSty
 
 		case SCE_H_OTHER:
 			if (sc.ch == '>' || sc.Match('/', '>') || (lexer.tagType == HtmlTagType::Question && sc.Match('?', '>'))) {
-				lexer.ClassifyHtmlTag();
+				lexer.ClassifyHtmlTag(keywordLists);
 				break;
 			}
 			if (sc.ch == '<') {

diff --git a/scintilla/lexlib/DocUtils.h b/scintilla/lexlib/DocUtils.h
@@ -4,16 +4,6 @@
 
 namespace Lexilla {
 
-constexpr size_t maxHtmlVoidTagLen = CStrLen("basefont");
-
-inline bool IsHtmlVoidTag(const char *tag) noexcept {
-	return nullptr != strstr(
-	// void elements
-	" area base basefont br col command embed frame hr img input isindex keygen link meta param source track wbr "
-	// end tag can be omitted
-	"p ", tag);
-}
-
 // based on CommonMark Spec 6.6 Raw HTML
 constexpr bool IsHtmlTagStart(int ch) noexcept {
 	return IsAlpha(ch);

diff --git a/src/EditAutoC.cpp b/src/EditAutoC.cpp
@@ -708,8 +708,8 @@ enum {
 	GroovyKeywordIndex_Annotation = 7,
 	GroovyKeywordIndex_GroovyDoc = 9,
 	HTMLKeywordIndex_Tag = 0,
-	HTMLKeywordIndex_Attribute = 4,
-	HTMLKeywordIndex_Value = 5,
+	HTMLKeywordIndex_Attribute = 5,
+	HTMLKeywordIndex_Value = 6,
 	HaxeKeywordIndex_Preprocessor = 1,
 	HaxeKeywordIndex_CommentTag = 8,
 	InnoKeywordIndex_Directive = 4,
@@ -723,7 +723,7 @@ enum {
 	KotlinKeywordIndex_KDoc = 8,
 	NSISKeywordIndex_PredefinedVariable = 5,
 	PHPKeywordIndex_PredefinedVariable = 4,
-	PHPKeywordIndex_Phpdoc = 11,
+	PHPKeywordIndex_Phpdoc = 12,
 	PerlKeywordIndex_Variable = 2,
 	PowerShellKeywordIndex_PredefinedVariable = 4,
 	PythonKeywordIndex_Decorator = 7,

diff --git a/src/EditLexers/stlPHP.cpp b/src/EditLexers/stlPHP.cpp
@@ -387,16 +387,19 @@ static KEYWORDLIST Keywords_PHP = {{
 "param_count params postname protocol_version queryString reconnect recurrences replacement report_mode "
 "server_info server_version sqlstate start stream thread_id type warning_count "
 
-, // 10 JavaScript
+, // 10 void tag
+" area base basefont br col command embed frame hr img input isindex keygen link meta p param source track wbr "
+
+, // 11 JavaScript
 "Infinity NaN arguments async await break case catch class const continue debugger default delete do else export extends "
 "false finally for function get globalThis if import in instanceof let new null of return set static super switch "
 "this throw true try typeof undefined var void while with yield "
 
-, // 11 phpdoc
+, // 12 phpdoc
 "api author category copyright deprecated example filesource global ignore internal license link method "
 "package param property return see since source subpackage throws todo used uses var version "
 
-, nullptr, nullptr, nullptr
+, nullptr, nullptr
 //--Autogenerated -- end of section automatically generated
 
 , // 15 Code Snippet
@@ -455,8 +458,9 @@ EDITLEXER lexPHP = {
 		| KeywordAttr64(7, KeywordAttr_NoLexer) // constant
 		| KeywordAttr64(8, KeywordAttr_NoLexer) // function
 		| KeywordAttr64(9, KeywordAttr_NoLexer) // misc
-		| KeywordAttr64(10, KeywordAttr_PreSorted | KeywordAttr_NoAutoComp) // JavaScript
-		| KeywordAttr64(11, KeywordAttr_NoLexer | KeywordAttr_NoAutoComp) // phpdoc
+		| KeywordAttr64(10, KeywordAttr_PreSorted | KeywordAttr_NoAutoComp) // void tag
+		| KeywordAttr64(11, KeywordAttr_PreSorted | KeywordAttr_NoAutoComp) // JavaScript
+		| KeywordAttr64(12, KeywordAttr_NoLexer | KeywordAttr_NoAutoComp) // phpdoc
 		, 0,
 		SCE_PHP_ESCAPECHAR, SCE_PHP_NOWDOC_ID,
 //Settings--Autogenerated -- end of section automatically generated

diff --git a/src/EditLexers/stlXML.cpp b/src/EditLexers/stlXML.cpp
@@ -5,24 +5,27 @@ static KEYWORDLIST Keywords_XML = {{
 //++Autogenerated -- start of section automatically generated
 nullptr
 
-, // 1 JavaScript
+, // 1 void tag
 nullptr
 
-, // 2 VBScript
+, // 2 JavaScript
 nullptr
 
-, // 3 SGML
+, // 3 VBScript
+nullptr
+
+, // 4 SGML
 "ANY ATTLIST CDATA DOCTYPE ELEMENT EMPTY ENTITIES ENTITY FIXED ID IDREF IDREFS IGNORE IMPLIED INCLUDE "
 "NDATA NMTOKEN NMTOKENS NOTATION PCDATA PUBLIC REQUIRED SGML SYSTEM doctype "
 
-, // 4 attribute
+, // 5 attribute
 "alternate charset encoding href id media name standalone title type value version "
 "xml xml-stylesheet xml:lang xmlns xmlns:xsi xsi:noNamespaceSchemaLocation xsi:schemaLocation "
 
-, // 5 value
+, // 6 value
 "ISO-8859-1 UTF-8 false no off on true utf-8 yes "
 
-, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr
+, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr, nullptr
 //--Autogenerated -- end of section automatically generated
 }};
 
@@ -52,9 +55,9 @@ EDITLEXER lexXML = {
 		0,
 		0, 0,
 		0, 0
-		, KeywordAttr32(3, KeywordAttr_PreSorted) // SGML
-		| KeywordAttr32(4, KeywordAttr_NoLexer) // attribute
-		| KeywordAttr32(5, KeywordAttr_NoLexer) // value
+		, KeywordAttr32(4, KeywordAttr_PreSorted) // SGML
+		| KeywordAttr32(5, KeywordAttr_NoLexer) // attribute
+		| KeywordAttr32(6, KeywordAttr_NoLexer) // value
 		, 0,
 		0, 0,
 //Settings--Autogenerated -- end of section automatically generated

diff --git a/src/Helpers.h b/src/Helpers.h
@@ -118,10 +118,9 @@ constexpr bool IsSchemeNameChar(int ch) noexcept {
 
 inline bool IsHtmlVoidTag(const char *tag) noexcept {
 	return nullptr != StrStrIA(
-	// void elements
-	" area base basefont br col command embed frame hr img input isindex keygen link meta param source track wbr "
-	// end tag can be omitted
-	"p ", tag);
+	// void elements, should keep sync with EditLexers/stlHTML.cpp
+	" area base basefont br col command embed frame hr img input isindex keygen link meta p param source track wbr "
+	, tag);
 }
 
 constexpr int ToUpperA(int ch) noexcept {

diff --git a/tools/KeywordCore.py b/tools/KeywordCore.py
@@ -23,6 +23,9 @@
 ColorNameList = set()
 CSharpKeywordMap = {}
 GroovyKeyword = []
+HtmlVoidTagList = """area base basefont br col command embed frame hr
+img input isindex keygen link meta param source track wbr
+p""".split()
 JavaKeywordMap = {}
 JavaScriptKeywordMap = {}
 SGMLKeyword = []
@@ -97,6 +100,9 @@ def BuildKeywordContent(rid, lexer, keywordList, keywordCount=16):
 			length = len(lines) + sum(len(line) for line in lines)
 			if length >= 0xffff:
 				print(rid, comment, 'string exceeds 64 KiB:', length)
+			if attr & KeywordAttr.PrefixSpace:
+				attr &= ~KeywordAttr.PrefixSpace
+				lines[0] = ' ' + lines[0]
 			output.extend('"' + line + ' "' for line in lines)
 		else:
 			output.append('nullptr')
@@ -1299,6 +1305,7 @@ def parse_html_api_file(path):
 	values = set(values) - set(keywords) - set(attributes)
 	return [
 		('tag', keywords, KeywordAttr.Special),
+		('void tag', HtmlVoidTagList, KeywordAttr.NoAutoComp | KeywordAttr.PrefixSpace),
 		('JavaScript', JavaScriptKeywordMap['keywords'], KeywordAttr.NoAutoComp),
 		('VBScript', VBScriptKeyword, KeywordAttr.MakeLower | KeywordAttr.NoAutoComp),
 		('SGML', SGMLKeyword, KeywordAttr.Default),
@@ -1911,6 +1918,7 @@ def parse_php_api_file(path):
 		('constant', keywordMap['constant'], KeywordAttr.NoLexer),
 		('function', keywordMap['function'], KeywordAttr.NoLexer),
 		('misc', keywordMap['misc'], KeywordAttr.NoLexer),
+		('void tag', HtmlVoidTagList, KeywordAttr.NoAutoComp | KeywordAttr.PrefixSpace),
 		('JavaScript', JavaScriptKeywordMap['keywords'], KeywordAttr.NoAutoComp),
 		('phpdoc', keywordMap['phpdoc'], KeywordAttr.NoLexer | KeywordAttr.NoAutoComp | KeywordAttr.Special),
 	]
@@ -2762,6 +2770,7 @@ def parse_xml_api_file(path):
 	SGMLKeyword.extend(keywords)
 	return [
 		('tag', [], KeywordAttr.Default),
+		('void tag', [], KeywordAttr.Default),
 		('JavaScript', [], KeywordAttr.Default),
 		('VBScript', [], KeywordAttr.Default),
 		('SGML', keywords, KeywordAttr.Default),

diff --git a/tools/LexerConfig.py b/tools/LexerConfig.py
@@ -75,6 +75,7 @@ class KeywordAttr(IntFlag):
 	NoLexer = 4		# not used by lexer, listed for auto-completion.
 	NoAutoComp = 8	# don't add to default auto-completion list.
 	Special = 256	# used by context based auto-completion.
+	PrefixSpace = 512	# prefix first item with extra space.
 
 TabSettings_Default = {
 	'tab_width': 4,