Merge branch 'release/2.1.5'

Fixes #30

Foreign attributes were handled incorrectly because attribute names
were implemented as pure strings without namespace prefix. This change
keeps the attributes as strings and eliminates any handling in regards
to namespaces.

Spec note about attributes:
"If designed today they would just have a name and value."

https://dom.spec.whatwg.org/#attr

.jazzy.yaml

@@ -1,5 +1,5 @@
 module: HTMLKit
-module_version: 2.1.4
+module_version: 2.1.5
 author: Iskandar Abudiab 
 author_url: https://twitter.com/iabudiab
 github_url: https://github.com/iabudiab/HTMLKit

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Change Log
 
+## [2.1.5](https://github.com/iabudiab/HTMLKit/releases/tag/2.1.5) 
+
+Released on 2018.07.16
+
+### Fixes
+
+- Parser would handle foreign attributes incorrectly (issue #30)
+
+
 ## [2.1.4](https://github.com/iabudiab/HTMLKit/releases/tag/2.1.4) 
 
 Released on 2018.05.01

HTMLKit.podspec

@@ -1,6 +1,6 @@
 Pod::Spec.new do |s|
   s.name               = "HTMLKit"
-  s.version            = "2.1.4"
+  s.version            = "2.1.5"
   s.summary            = "HTMLKit, an Objective-C framework for your everyday HTML needs."
   s.license            = "MIT"
   s.homepage           = "https://github.com/iabudiab/HTMLKit"

HTMLKit.xcodeproj/project.pbxproj

@@ -312,6 +312,9 @@
 		62A95A4D1FB0FBFC0009FF26 /* HTMLSerializationTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62A95A4C1FB0FBFC0009FF26 /* HTMLSerializationTests.m */; };
 		62A95A4E1FB0FBFC0009FF26 /* HTMLSerializationTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62A95A4C1FB0FBFC0009FF26 /* HTMLSerializationTests.m */; };
 		62A95A4F1FB0FBFC0009FF26 /* HTMLSerializationTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62A95A4C1FB0FBFC0009FF26 /* HTMLSerializationTests.m */; };
+		62C82E0D20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62C82E0C20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m */; };
+		62C82E0E20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62C82E0C20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m */; };
+		62C82E0F20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */ = {isa = PBXBuildFile; fileRef = 62C82E0C20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m */; };
 		62D8345A19FB1AC4009205A9 /* HTML5LibTokenizerTest.m in Sources */ = {isa = PBXBuildFile; fileRef = 62D8345819FB1AC4009205A9 /* HTML5LibTokenizerTest.m */; };
 		62D91C231DE218A500BEFADE /* HTMLRange.h in Headers */ = {isa = PBXBuildFile; fileRef = 62D91C211DE218A500BEFADE /* HTMLRange.h */; settings = {ATTRIBUTES = (Public, ); }; };
 		62D91C241DE218A500BEFADE /* HTMLRange.h in Headers */ = {isa = PBXBuildFile; fileRef = 62D91C211DE218A500BEFADE /* HTMLRange.h */; settings = {ATTRIBUTES = (Public, ); }; };
@@ -723,6 +726,7 @@
 		62AE594319F992F30043F069 /* HTMLCommentToken.m */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.objc; path = HTMLCommentToken.m; sourceTree = "<group>"; };
 		62AE594719F9948A0043F069 /* HTMLCharacterToken.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = HTMLCharacterToken.h; path = include/HTMLCharacterToken.h; sourceTree = "<group>"; };
 		62AE594819F9948A0043F069 /* HTMLCharacterToken.m */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.objc; path = HTMLCharacterToken.m; sourceTree = "<group>"; };
+		62C82E0C20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m */ = {isa = PBXFileReference; lastKnownFileType = sourcecode.c.objc; name = HTMLKitParserIssuesTests.m; path = HTMLKitTests/HTMLKitParserIssuesTests.m; sourceTree = "<group>"; };
 		62D8345719FB1AC4009205A9 /* HTML5LibTokenizerTest.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = HTML5LibTokenizerTest.h; path = HTMLKitTests/HTML5LibTokenizerTest.h; sourceTree = "<group>"; };
 		62D8345819FB1AC4009205A9 /* HTML5LibTokenizerTest.m */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.objc; name = HTML5LibTokenizerTest.m; path = HTMLKitTests/HTML5LibTokenizerTest.m; sourceTree = "<group>"; };
 		62D91C211DE218A500BEFADE /* HTMLRange.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; name = HTMLRange.h; path = include/HTMLRange.h; sourceTree = "<group>"; };
@@ -993,6 +997,7 @@
 				625EE45A1CBB171300F2CC8E /* HTMLKitTestUtil.m */,
 				6236738C1AC0CD2400FF89B3 /* Tokenizer */,
 				623975581AC362A5007E26F1 /* Tree Construction */,
+				62C82E0B20FD2FCB008497A8 /* Parser */,
 				624B9FB71AE072CB00646C4C /* DOM */,
 				624B9FB81AE072D500646C4C /* Categories */,
 				624E1A2D1B1D1C8A00E66AAC /* Structures */,
@@ -1097,6 +1102,14 @@
 			name = Tokenizing;
 			sourceTree = "<group>";
 		};
+		62C82E0B20FD2FCB008497A8 /* Parser */ = {
+			isa = PBXGroup;
+			children = (
+				62C82E0C20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m */,
+			);
+			name = Parser;
+			sourceTree = "<group>";
+		};
 		62ECBEDF1C0B671000AF847B /* Parsing */ = {
 			isa = PBXGroup;
 			children = (
@@ -1758,6 +1771,7 @@
 				620AB7B12087F18D00AFCCC7 /* CSSStructuralPseudoSelectors.m in Sources */,
 				62E0BA971E25456700E4D193 /* HTMLCharacterDataTests.m in Sources */,
 				6216ACFD1C28DCC80074CAB4 /* CSSExtensionSelectorsParsingTests.m in Sources */,
+				62C82E0D20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */,
 				62132E5A1C01F83200084175 /* CSSSelectorTest.m in Sources */,
 				6239755B1AC362CA007E26F1 /* HTMLKitTreeConstructionTests.m in Sources */,
 				62F658711BD83C8E0045F137 /* CSSNThExpressionSelectorTests.m in Sources */,
@@ -1902,6 +1916,7 @@
 				620AB7B32087F18D00AFCCC7 /* CSSStructuralPseudoSelectors.m in Sources */,
 				62E0BA991E25456700E4D193 /* HTMLCharacterDataTests.m in Sources */,
 				62857D3B1D39A345008DC254 /* CSSSelectorTest.m in Sources */,
+				62C82E0F20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */,
 				62857D421D39A345008DC254 /* CSSCombinatorSelectorTests.m in Sources */,
 				62857D3C1D39A345008DC254 /* CSSSelectorParserTests.m in Sources */,
 				62857D301D39A339008DC254 /* HTMLKitTokenizerPerformance.m in Sources */,
@@ -1991,6 +2006,7 @@
 				620AB7B22087F18D00AFCCC7 /* CSSStructuralPseudoSelectors.m in Sources */,
 				62E0BA981E25456700E4D193 /* HTMLCharacterDataTests.m in Sources */,
 				6216ACFE1C28DCC80074CAB4 /* CSSExtensionSelectorsParsingTests.m in Sources */,
+				62C82E0E20FD2FFD008497A8 /* HTMLKitParserIssuesTests.m in Sources */,
 				62ECBFCA1C0B6E2E00AF847B /* HTML5LibTokenizerTest.m in Sources */,
 				62ECBFCB1C0B6E2E00AF847B /* HTMLKitTokenizerTests.m in Sources */,
 				62ECBFCC1C0B6E2E00AF847B /* HTMLKitTokenizerPerformance.m in Sources */,

Sources/HTMLKit-Info.plist

@@ -17,7 +17,7 @@
 	<key>CFBundlePackageType</key>
 	<string>FMWK</string>
 	<key>CFBundleShortVersionString</key>
-	<string>2.1.4</string>
+	<string>2.1.5</string>
 	<key>CFBundleSignature</key>
 	<string>????</string>
 	<key>CFBundleVersion</key>

Sources/HTMLNode.m

@@ -127,7 +127,7 @@ NSString * const RemoveChildNode = @"-removeChildNode:";
 
 - (HTMLElement *)nextSiblingElement
 {
-	HTMLNode *node = self.previousSibling;
+	HTMLNode *node = self.nextSibling;
 	while (node && node.nodeType != HTMLNodeElement) {
 		node = node.nextSibling;
 	}

Sources/HTMLParser.m

@@ -1443,7 +1443,6 @@
 	} else if ([tagName isEqualToString:@"math"]) {
 		[self reconstructActiveFormattingElements];
 		AdjustMathMLAttributes(token);
-		AdjustForeignAttributes(token);
 		[self insertForeignElementForToken:token inNamespace:HTMLNamespaceMathML];
 		if (token.isSelfClosing) {
 			[_stackOfOpenElements popCurrentNode];
@@ -1451,7 +1450,6 @@
 	} else if ([tagName isEqualToString:@"svg"]) {
 		[self reconstructActiveFormattingElements];
 		AdjustSVGAttributes(token);
-		AdjustForeignAttributes(token);
 		[self insertForeignElementForToken:token inNamespace:HTMLNamespaceSVG];
 		if (token.isSelfClosing) {
 			[_stackOfOpenElements popCurrentNode];
@@ -2541,7 +2539,6 @@
 					AdjustSVGNameCase(token.asTagToken);
 					AdjustSVGAttributes(token.asTagToken);
 				}
-				AdjustForeignAttributes(token.asTagToken);
 				[self insertForeignElementForToken:token.asTagToken inNamespace:self.adjustedCurrentNode.htmlNamespace];
 				if (token.asTagToken.selfClosing) {
 					[_stackOfOpenElements popCurrentNode];

Sources/include/HTMLElementAdjustment.h

@@ -140,31 +140,3 @@ NS_INLINE void AdjustSVGNameCase(HTMLTagToken *token)
 	NSString *replacement = replacements[token.tagName] ?: token.tagName;
 	token.tagName = replacement;
 }
-
-NS_INLINE void AdjustForeignAttributes(HTMLTagToken *token)
-{
-	if (token.attributes == nil) {
-		return;
-	}
-
-	NSDictionary *replacements = @{ @"xlink:actuate": @"xlink actuate",
-									@"xlink:arcrole": @"xlink arcrole",
-									@"xlink:href": @"xlink href",
-									@"xlink:role": @"xlink role",
-									@"xlink:show": @"xlink show",
-									@"xlink:title": @"xlink title",
-									@"xlink:type": @"xlink type",
-									@"xml:base": @"xml base",
-									@"xml:lang": @"xml lang",
-									@"xml:space": @"xml space",
-									@"xmlns": @"xmlns",
-									@"xmlns:xlink": @"xmlns xlink"};
-
-	HTMLOrderedDictionary *adjusted = [HTMLOrderedDictionary new];
-	for (id key in token.attributes) {
-		NSString *replacement = replacements[key] ?: key;
-		adjusted[replacement] = token.attributes[key];
-	}
-	token.attributes = adjusted;
-}
-

Tests/HTMLKitTests/HTML5LibTreeConstructionTest.m

@@ -9,6 +9,7 @@
 #import "HTML5LibTreeConstructionTest.h"
 #import <XCTest/XCTest.h>
 
+#import "HTMLDOM.h"
 #import "HTMLDocumentType.h"
 #import "HTMLElement.h"
 #import "HTMLText.h"
@@ -289,6 +290,8 @@ NS_INLINE NSArray * parseAttribute(NSString *str)
 	NSRange range = [str rangeOfString:@"=" options:0];
 
 	NSString *key = [str substringToIndex:range.location];
+	key = [key stringByReplacingOccurrencesOfString:@" " withString:@":"];
+
 	NSString *value = [str substringFromIndex:range.location + 2];
 	value = [value substringToIndex:value.length - 1];
 

Tests/HTMLKitTests/HTMLKitParserIssuesTests.m

@@ -0,0 +1,43 @@
+//
+//  HTMLKitParserTests.m
+//  HTMLKit
+//
+//  Created by Iska on 16.07.18.
+//  Copyright © 2018 BrainCookie. All rights reserved.
+//
+
+#import <XCTest/XCTest.h>
+#import "HTMLDOM.h"
+
+@interface HTMLKitParserIssuesTests : XCTestCase
+
+@end
+
+@implementation HTMLKitParserIssuesTests
+
+#pragma mark - Bug Fixes
+
+- (void)testBugFix_Issue_30 {
+	NSString *html =
+	@"<body>"
+	"  <svg id='draw_area' width='600' height='800' xmlns='http://www.w3.org/2000/svg' xmlns:xlink='http://www.w3.org/1999/xlink' version='1.1'>"
+	"    <image id='overlay_img' xlink:href='foo.png' width='600' height='800'/>"
+	"  </svg>"
+	"</body>";
+
+	HTMLDocument* document = [HTMLDocument documentWithString:html];
+	HTMLElement *svg = [document querySelector:@"#draw_area"];
+
+	XCTAssertNil(svg.attributes[@"xlink"]);
+	XCTAssertEqualObjects(svg.attributes[@"xmlns"], @"http://www.w3.org/2000/svg");
+	XCTAssertEqualObjects(svg.attributes[@"xmlns:xlink"], @"http://www.w3.org/1999/xlink");
+
+	HTMLElement *image = [document querySelector:@"#overlay_img"];
+
+	XCTAssertNil(image.attributes[@"xlink"]);
+	XCTAssertNil(image.attributes[@"href"]);
+	XCTAssertEqualObjects(image.attributes[@"xlink:href"], @"foo.png");
+	XCTAssertEqualObjects(image.outerHTML, @"<image id=\"overlay_img\" xlink:href=\"foo.png\" width=\"600\" height=\"800\"></image>");
+}
+
+@end

Tests/HTMLKitTests/HTMLNodeIteratorTests.m

@@ -608,4 +608,19 @@ static HTMLNode * (^ LastDescendant)(HTMLNode *) = ^ HTMLNode * (HTMLNode *node)
     XCTAssertTrue([element.elementId isEqualToString:divId]);
 }
 
+- (void)testBugFix_Issue_28 {
+    HTMLDocument *document = self.document;
+    HTMLNodeIterator *iterator = document.body.nodeIterator;
+    
+    [iterator nextNode]; // Reference node: <body>
+    
+    HTMLElement *span = (HTMLElement*)iterator.nextNode; // <span>
+    NSString *spanTag = @"span";
+    XCTAssertTrue([span.tagName isEqualToString:spanTag]);
+    
+    HTMLElement *paragraph = span.nextSiblingElement; // <p>
+    NSString *paragraphTag = @"p";
+    XCTAssertTrue([paragraph.tagName isEqualToString:paragraphTag]);
+}
+
 @end