WIP

blocks/api/raw-handling/embedded-content-reducer.js

@@ -25,6 +25,7 @@ export default function( node ) {
 	}
 
 	let nodeToInsert = node;
+
 	// if the embedded is an image and its parent is an anchor with just the image
 	// take the anchor out instead of just the image
 	if (
@@ -41,7 +42,13 @@ export default function( node ) {
 		wrapper = wrapper.parentElement;
 	}
 
+	const figure = document.createElement( 'figure' );
+
 	if ( wrapper ) {
-		wrapper.parentNode.insertBefore( nodeToInsert, wrapper );
+		wrapper.parentNode.insertBefore( figure, wrapper );
+	} else {
+		nodeToInsert.parentNode.insertBefore( figure, nodeToInsert );
 	}
+
+	figure.appendChild( nodeToInsert );
 }

blocks/api/raw-handling/index.js

@@ -1,7 +1,6 @@
 /**
  * External dependencies
  */
-import { compact } from 'lodash';
 import showdown from 'showdown';
 
 /**
@@ -11,20 +10,15 @@ import { createBlock, getBlockTransforms, findTransform } from '../factory';
 import { getBlockType, getUnknownTypeHandlerName } from '../registration';
 import { getBlockAttributes, parseWithGrammar } from '../parser';
 import normaliseBlocks from './normalise-blocks';
-import stripAttributes from './strip-attributes';
 import specialCommentConverter from './special-comment-converter';
-import commentRemover from './comment-remover';
-import createUnwrapper from './create-unwrapper';
 import isInlineContent from './is-inline-content';
-import formattingTransformer from './formatting-transformer';
+import phrasingContentReducer from './phrasing-content-reducer';
 import msListConverter from './ms-list-converter';
 import listReducer from './list-reducer';
 import imageCorrector from './image-corrector';
 import blockquoteNormaliser from './blockquote-normaliser';
-import tableNormaliser from './table-normaliser';
-import inlineContentConverter from './inline-content-converter';
 import embeddedContentReducer from './embedded-content-reducer';
-import { deepFilterHTML, isInvalidInline, isNotWhitelisted, isPlain, isInline } from './utils';
+import { deepFilterHTML, isPlain, getContentSchema, getPhrasingContentSchema, removeInvalidHTML } from './utils';
 import shortcodeConverter from './shortcode-converter';
 import slackMarkdownVariantCorrector from './slack-markdown-variant-corrector';
 
@@ -94,14 +88,8 @@ export default function rawHandler( { HTML, plainText = '', mode = 'AUTO', tagNa
 
 	// Return filtered HTML if condition is true
 	if ( mode === 'INLINE' || isAutoModeInline ) {
-		HTML = deepFilterHTML( HTML, [
-			// Add semantic formatting before attributes are stripped.
-			formattingTransformer,
-			stripAttributes,
-			specialCommentConverter,
-			commentRemover,
-			createUnwrapper( ( node ) => ! isInline( node, tagName ) ),
-		] );
+		HTML = deepFilterHTML( HTML, [ phrasingContentReducer ] );
+		HTML = removeInvalidHTML( HTML, getPhrasingContentSchema() );
 
 		// Allows us to ask for this information when we get a report.
 		window.console.log( 'Processed inline HTML:\n\n', HTML );
@@ -116,31 +104,17 @@ export default function rawHandler( { HTML, plainText = '', mode = 'AUTO', tagNa
 			return [ ...accu, piece ];
 		}
 
-		// Context dependent filters. Needs to run before we remove nodes.
 		piece = deepFilterHTML( piece, [
 			msListConverter,
-		] );
-
-		piece = deepFilterHTML( piece, compact( [
 			listReducer,
 			imageCorrector,
-			// Add semantic formatting before attributes are stripped.
-			formattingTransformer,
-			stripAttributes,
+			phrasingContentReducer,
 			specialCommentConverter,
-			commentRemover,
-			! canUserUseUnfilteredHTML && createUnwrapper( ( element ) => element.nodeName === 'IFRAME' ),
 			embeddedContentReducer,
-			createUnwrapper( isNotWhitelisted ),
 			blockquoteNormaliser,
-			tableNormaliser,
-			inlineContentConverter,
-		] ) );
-
-		piece = deepFilterHTML( piece, [
-			createUnwrapper( isInvalidInline ),
 		] );
 
+		piece = removeInvalidHTML( piece, getContentSchema( { iframe: canUserUseUnfilteredHTML } ) );
 		piece = normaliseBlocks( piece );
 
 		// Allows us to ask for this information when we get a report.

blocks/api/raw-handling/list-reducer.js

@@ -1,8 +1,17 @@
+/**
+ * WordPress dependencies
+ */
+import { unwrap } from '@wordpress/utils';
+
 /**
  * Browser dependencies
  */
 const { ELEMENT_NODE } = window.Node;
 
+function isList( node ) {
+	return node.nodeName === 'OL' || node.nodeName === 'UL';
+}
+
 function shallowTextContent( element ) {
 	return [ ...element.childNodes ]
 		.map( ( { nodeValue = '' } ) => nodeValue )
@@ -14,9 +23,7 @@ export default function( node ) {
 		return;
 	}
 
-	const type = node.nodeName;
-
-	if ( type !== 'OL' && type !== 'UL' ) {
+	if ( ! isList( node ) ) {
 		return;
 	}
 
@@ -28,7 +35,7 @@ export default function( node ) {
 	// * There is only one list item.
 	if (
 		prevElement &&
-		prevElement.nodeName === type &&
+		prevElement.nodeName === node.nodeName &&
 		list.children.length === 1
 	) {
 		prevElement.appendChild( list.firstChild );
@@ -56,4 +63,15 @@ export default function( node ) {
 			parentList.parentNode.removeChild( parentList );
 		}
 	}
+
+	// Invalid: OL/UL > OL/UL.
+	if ( parentElement && isList( parentElement ) ) {
+		const prevListItem = node.previousElementSibling;
+
+		if ( prevListItem ) {
+			prevListItem.appendChild( node );
+		} else {
+			unwrap( node );
+		}
+	}
 }

blocks/api/raw-handling/phrasing-content-reducer.js

@@ -0,0 +1,47 @@
+/**
+ * WordPress dependencies
+ */
+import { unwrap, replaceTag } from '@wordpress/utils';
+
+/**
+ * Internal dependencies
+ */
+import { isPhrasingContent, isBlockContent } from './utils';
+
+/**
+ * Browser dependencies
+ */
+const { ELEMENT_NODE } = window.Node;
+
+export default function( node ) {
+	if ( node.nodeType !== ELEMENT_NODE ) {
+		return;
+	}
+
+	if ( node.nodeName === 'SPAN' ) {
+		const fontWeight = node.style.fontWeight;
+		const fontStyle = node.style.fontStyle;
+
+		if ( fontWeight === 'bold' || fontWeight === '700' ) {
+			node = replaceTag( node, 'strong' );
+		} else if ( fontStyle === 'italic' ) {
+			node = replaceTag( node, 'em' );
+		}
+	}
+
+	if ( node.nodeName === 'B' ) {
+		node = replaceTag( node, 'strong' );
+	}
+
+	if ( node.nodeName === 'I' ) {
+		node = replaceTag( node, 'em' );
+	}
+
+	if (
+		isPhrasingContent( node ) &&
+		node.hasChildNodes() &&
+		Array.from( node.childNodes ).some( isBlockContent )
+	) {
+		unwrap( node );
+	}
+}