Avoid regexes, use private api

packages/block-editor/src/components/rich-text/use-paste-handler.js

@@ -5,7 +5,7 @@ import { useRef } from '@wordpress/element';
 import { useRefEffect } from '@wordpress/compose';
 import {
 	pasteHandler,
-	getClipboardEventData,
+	privateApis,
 	findTransform,
 	getBlockTransforms,
 } from '@wordpress/blocks';
@@ -17,6 +17,9 @@ import { isURL } from '@wordpress/url';
  */
 import { addActiveFormats } from './utils';
 import { splitValue } from './split-value';
+import { unlock } from '../../lock-unlock';
+
+const { getClipboardEventData } = unlock( privateApis );
 
 /** @typedef {import('@wordpress/rich-text').RichTextValue} RichTextValue */
 

packages/block-editor/src/components/writing-flow/use-clipboard-handler.js

@@ -4,7 +4,7 @@
 import {
 	serialize,
 	pasteHandler,
-	getClipboardEventData,
+	privateApis,
 	createBlock,
 	findTransform,
 	getBlockTransforms,
@@ -22,6 +22,9 @@ import { useRefEffect } from '@wordpress/compose';
  */
 import { store as blockEditorStore } from '../../store';
 import { useNotifyCopy } from '../../utils/use-notify-copy';
+import { unlock } from '../../lock-unlock';
+
+const { getClipboardEventData } = unlock( privateApis );
 
 export default function useClipboardHandler() {
 	const {

packages/blocks/README.md

@@ -463,6 +463,10 @@ _Returns_
 
 -   `Array|string`: A list of blocks or a string, depending on `handlerMode`.
 
+### privateApis
+
+Private @wordpress/blocks APIs.
+
 ### rawHandler
 
 Converts an HTML string to known blocks.

packages/blocks/src/api/raw-handling/get-clipboard-event-data.js

@@ -4,37 +4,12 @@
 import { getFilesFromDataTransfer } from '@wordpress/dom';
 
 /**
- * Normalizes a given string of HTML to remove the Windows-specific "Fragment"
- * comments and any preceding and trailing content.
- *
- * @param {string} html the html to be normalized
- * @return {string} the normalized html
+ * Internal dependencies
  */
-export function removeWindowsFragments( html ) {
-	// Strip Windows markers.
-	return html
-		.replace(
-			/^\s*<html[^>]*>\s*<body[^>]*>(?:\s*<!--\s*StartFragment\s*-->)?/i,
-			''
-		)
-		.replace(
-			/(?:<!--\s*EndFragment\s*-->\s*)?<\/body>\s*<\/html>\s*$/i,
-			''
-		);
-}
-
-/**
- * Removes the charset meta tag inserted by Chromium.
- * See:
- * - https://github.com/WordPress/gutenberg/issues/33585
- * - https://bugs.chromium.org/p/chromium/issues/detail?id=1264616#c4
- *
- * @param {string} html the html to be stripped of the meta tag.
- * @return {string} the cleaned html
- */
-export function removeMetaTags( html ) {
-	return html.replace( /<meta[^>]+>/gi, '' );
-}
+import { deepFilterHTML } from './utils';
+import wrapperRemover from './wrapper-remover';
+import msFragmentRemover from './ms-fragment-remover';
+import metaRemover from './meta-remover';
 
 export function getClipboardEventData( event ) {
 	const { clipboardData } = event;
@@ -58,8 +33,11 @@ export function getClipboardEventData( event ) {
 		}
 	}
 
-	html = removeWindowsFragments( html );
-	html = removeMetaTags( html );
+	html = deepFilterHTML( html, [
+		wrapperRemover,
+		msFragmentRemover,
+		metaRemover,
+	] );
 
 	const files = getFilesFromDataTransfer( clipboardData );
 

packages/blocks/src/api/raw-handling/meta-remover.js

@@ -0,0 +1,13 @@
+/**
+ * Removes the charset meta tag inserted by Chromium, along with any other ones.
+ * See:
+ * - https://github.com/WordPress/gutenberg/issues/33585
+ * - https://bugs.chromium.org/p/chromium/issues/detail?id=1264616#c4
+ *
+ * @param {Node} node The node to be processed.
+ */
+export default function metaRemover( node ) {
+	if ( node.nodeName === 'META' ) {
+		node.remove();
+	}
+}

packages/blocks/src/api/raw-handling/ms-fragment-remover.js

@@ -0,0 +1,18 @@
+/**
+ * WordPress dependencies
+ */
+import { remove } from '@wordpress/dom';
+
+/**
+ * Windows-specific "Fragment" comments.
+ *
+ * @param {Node} node The node to be processed.
+ */
+export default function msFragmentRemover( node ) {
+	if (
+		node.nodeType === node.COMMENT_NODE &&
+		[ 'StartFragment', 'EndFragment' ].includes( node.textContent.trim() )
+	) {
+		remove( node );
+	}
+}

packages/blocks/src/api/raw-handling/paste-handler.js

@@ -32,10 +32,9 @@ import brRemover from './br-remover';
 import { deepFilterHTML, isPlain, getBlockContentSchema } from './utils';
 import emptyParagraphRemover from './empty-paragraph-remover';
 import slackParagraphCorrector from './slack-paragraph-corrector';
-import {
-	removeMetaTags,
-	removeWindowsFragments,
-} from './get-clipboard-event-data';
+import wrapperRemover from './wrapper-remover';
+import msFragmentRemover from './ms-fragment-remover';
+import metaRemover from './meta-remover';
 
 /**
  * Browser dependencies
@@ -94,8 +93,11 @@ export function pasteHandler( {
 	tagName,
 	preserveWhiteSpace,
 } ) {
-	HTML = removeMetaTags( HTML );
-	HTML = removeWindowsFragments( HTML );
+	HTML = deepFilterHTML( HTML, [
+		wrapperRemover,
+		msFragmentRemover,
+		metaRemover,
+	] );
 
 	// If we detect block delimiters in HTML, parse entirely as blocks.
 	if ( mode !== 'INLINE' ) {

packages/blocks/src/api/raw-handling/wrapper-remover.js

@@ -0,0 +1,10 @@
+/**
+ * WordPress dependencies
+ */
+import { unwrap } from '@wordpress/dom';
+
+export default function wrapperRemover( node ) {
+	if ( [ 'BODY', 'HTML' ].includes( node.nodeName ) ) {
+		unwrap( node );
+	}
+}

packages/blocks/src/index.js

@@ -11,3 +11,4 @@
 export { store } from './store';
 export * from './api';
 export * from './deprecated';
+export { privateApis } from './private-apis';

packages/blocks/src/private-apis.js

@@ -0,0 +1,13 @@
+/**
+ * Internal dependencies
+ */
+import { lock } from './lock-unlock';
+import { getClipboardEventData } from './api/raw-handling/get-clipboard-event-data';
+
+/**
+ * Private @wordpress/blocks APIs.
+ */
+export const privateApis = {};
+lock( privateApis, {
+	getClipboardEventData,
+} );

test/integration/__snapshots__/blocks-raw-handling.test.js.snap

@@ -26,6 +26,8 @@ exports[`Blocks raw handling pasteHandler iframe-embed 1`] = `""`;
 
 exports[`Blocks raw handling pasteHandler markdown 1`] = `"This is a heading with <em>italic</em><br>This is a paragraph with a <a href="https://w.org/">link</a>, <strong>bold</strong>, and <del>strikethrough</del>.<br>Preserve<br>line breaks please.<br>Lists<br>A<br>Bulleted Indented<br>List<br>One<br>Two<br>Three<br>Table<br>First Header<br>Second Header<br>Content from cell 1<br>Content from cell 2<br>Content in the first column<br>Content in the second column<br><br><br><br>Table with empty cells.<br>Quote<br>First<br>Second<br>Code<br>Inline <code>code</code> tags should work.<br><code>This is a code block.</code>"`;
 
+exports[`Blocks raw handling pasteHandler ms-excel 1`] = `"<br><br><br>Cell 1<br>Cell 2<br>Cell 3<br>Cell 4"`;
+
 exports[`Blocks raw handling pasteHandler ms-word 1`] = `"This is a title<br>&nbsp;<br>This is a subtitle<br>&nbsp;<br>This is a heading level 1<br>&nbsp;<br>This is a heading level 2<br>&nbsp;<br>This is a <strong>paragraph</strong> with a <a href="https://w.org/">link</a>.<br>&nbsp;<br>A<br>Bulleted<br>Indented<br>List<br>&nbsp;<br>One<br>Two<br>Three<br>&nbsp;<br>One<br>Two<br>Three<br>1<br>2<br>3<br>I<br>II<br>III<br>&nbsp;<br>An image:<br>&nbsp;<br><img width="451" height="338" src="file:LOW-RES.png"><br><a href="#anchor">This is an anchor link</a> that leads to the next paragraph.<br><a id="anchor">This is the paragraph with the anchor.</a><br><a href="#nowhere">This is an anchor link</a> that leads nowhere.<br><a>This is a paragraph with an anchor with no link pointing to it.</a><br>This is a reference to a footnote<a href="#_ftn1" id="_ftnref1">[1]</a>.<br>This is a reference to an endnote<a href="#_edn1" id="_ednref1">[i]</a>.<br><br><br><a href="#_ftnref1" id="_ftn1">[1]</a> This is a footnote.<br><br><br><a href="#_ednref1" id="_edn1">[i]</a> This is an endnote."`;
 
 exports[`Blocks raw handling pasteHandler ms-word-list 1`] = `"<a>This is a headline?</a><br>This is a text:<br>One<br>Two<br>Three<br><a>Lorem Ipsum.</a><br>&nbsp;"`;

test/integration/blocks-raw-handling.test.js

@@ -382,6 +382,7 @@ describe( 'Blocks raw handling', () => {
 			'google-docs-table-with-rowspan',
 			'google-docs-table-with-comments',
 			'google-docs-with-comments',
+			'ms-excel',
 			'ms-word',
 			'ms-word-list',
 			'ms-word-styled',

test/integration/fixtures/documents/ms-excel-in.html

@@ -0,0 +1,96 @@
+<html xmlns:o="urn:schemas-microsoft-com:office:office"
+xmlns:x="urn:schemas-microsoft-com:office:excel"
+xmlns="http://www.w3.org/TR/REC-html40">
+
+<head>
+<meta http-equiv=Content-Type content="text/html; charset=utf-8">
+<meta name=ProgId content=Excel.Sheet>
+<meta name=Generator content="Microsoft Excel 15">
+<link id=Main-File rel=Main-File
+href="file:///C:/Users/TETSUA~1.HAM/AppData/Local/Temp/msohtmlclip1/01/clip.htm">
+<link rel=File-List
+href="file:///C:/Users/TETSUA~1.HAM/AppData/Local/Temp/msohtmlclip1/01/clip_filelist.xml">
+<style>
+<!--table
+	{mso-displayed-decimal-separator:"\.";
+	mso-displayed-thousand-separator:"\,";}
+@page
+	{margin:.75in .7in .75in .7in;
+	mso-header-margin:.3in;
+	mso-footer-margin:.3in;}
+.font5
+	{color:windowtext;
+	font-size:6.0pt;
+	font-weight:400;
+	font-style:normal;
+	text-decoration:none;
+	font-family:游ゴシック, monospace;
+	mso-font-charset:128;}
+tr
+	{mso-height-source:auto;
+	mso-ruby-visibility:none;}
+col
+	{mso-width-source:auto;
+	mso-ruby-visibility:none;}
+br
+	{mso-data-placement:same-cell;}
+td
+	{padding-top:1px;
+	padding-right:1px;
+	padding-left:1px;
+	mso-ignore:padding;
+	color:black;
+	font-size:11.0pt;
+	font-weight:400;
+	font-style:normal;
+	text-decoration:none;
+	font-family:游ゴシック, monospace;
+	mso-font-charset:128;
+	mso-number-format:General;
+	text-align:general;
+	vertical-align:bottom;
+	border:none;
+	mso-background-source:auto;
+	mso-pattern:auto;
+	mso-protection:locked visible;
+	white-space:nowrap;
+	mso-rotate:0;}
+.xl65
+	{font-family:"Noto Sans JP", sans-serif;
+	mso-font-charset:128;}
+ruby
+	{ruby-align:left;}
+rt
+	{color:windowtext;
+	font-size:6.0pt;
+	font-weight:400;
+	font-style:normal;
+	text-decoration:none;
+	font-family:游ゴシック, monospace;
+	mso-font-charset:128;
+	mso-char-type:katakana;
+	display:none;}
+-->
+</style>
+</head>
+
+<body link="#0563C1" vlink="#954F72">
+
+<table border=0 cellpadding=0 cellspacing=0 width=140 style='border-collapse:
+ collapse;width:106pt'>
+<!--StartFragment-->
+ <col width=70 span=2 style='width:53pt'>
+ <tr height=24 style='height:18.0pt'>
+  <td height=24 class=xl65 width=70 style='height:18.0pt;width:53pt'>Cell 1</td>
+  <td class=xl65 width=70 style='width:53pt'>Cell 2</td>
+ </tr>
+ <tr height=24 style='height:18.0pt'>
+  <td height=24 class=xl65 style='height:18.0pt'>Cell 3</td>
+  <td class=xl65>Cell 4</td>
+ </tr>
+<!--EndFragment-->
+</table>
+
+</body>
+
+</html>

test/integration/fixtures/documents/ms-excel-out.html

@@ -0,0 +1,3 @@
+<!-- wp:table -->
+<figure class="wp-block-table"><table><tbody><tr><td>Cell 1</td><td>Cell 2</td></tr><tr><td>Cell 3</td><td>Cell 4</td></tr></tbody></table></figure>
+<!-- /wp:table -->