single-file-core 1.3.21 → 1.3.23

package/core/lib/processor-helper.js

@@ -394,7 +394,7 @@ function getProcessorHelperClass(utilInstance) {
 		}
 
 		setMetaCSP(metaElement) {
-			metaElement.content = "default-src 'none'; connect-src 'self'; font-src 'self' data: blob:; img-src 'self' data: blob:; style-src 'self' 'unsafe-inline' data: blob:; frame-src 'self' data: blob:; media-src 'self' data: blob:; script-src 'self' 'unsafe-inline' data: blob:; object-src 'self' data: blob:;";
+			metaElement.content = "default-src 'none'; connect-src 'self' data: blob:; font-src 'self' data: blob:; img-src 'self' data: blob:; style-src 'self' 'unsafe-inline' data: blob:; frame-src 'self' data: blob:; media-src 'self' data: blob:; script-src 'self' 'unsafe-inline' data: blob:; object-src 'self' data: blob:;";
 		}
 
 		removeUnusedStylesheets() {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "single-file-core",
-  "version": "1.3.21",
+  "version": "1.3.23",
   "description": "SingleFile Core",
   "author": "Gildas Lormeau",
   "license": "AGPL-3.0-or-later",

package/processors/compression/compression-extract.js

@@ -70,6 +70,7 @@ async function extract(content, { password, prompt = () => { }, shadowRootScript
 	const REGEXP_MATCH_ROOT_INDEX = /^([0-9_]+\/)?index\.html$/;
 	const REGEXP_MATCH_INDEX = /index\.html$/;
 	const REGEXP_MATCH_FRAMES = /frames\//;
+	const REGEXP_MATCH_MANIFEST = /manifest\.json$/;
 	const CHARSET_UTF8 = ";charset=utf-8";
 	const REGEXP_ESCAPE = /([{}()^$&.*?/+|[\\\\]|\]|-)/g;
 
@@ -105,11 +106,6 @@ async function extract(content, { password, prompt = () => { }, shadowRootScript
 				} else if (filename.match(REGEXP_MATCH_SCRIPT)) {
 					mimeType = "text/javascript" + CHARSET_UTF8;
 				}
-				if (textContent !== undefined) {
-					content = noBlobURL ? await getDataURI(textContent, mimeType) : URL.createObjectURL(new Blob([textContent], { type: mimeType }));
-				} else {
-					content = "data:text/plain,";
-				}
 			}
 		} else {
 			resources.push(resourceInfo);
@@ -127,7 +123,13 @@ async function extract(content, { password, prompt = () => { }, shadowRootScript
 			}
 		}
 		const name = entry.filename.match(/^([0-9_]+\/)?(.*)$/)[2];
+		let prefixPath = "";
+		const prefixPathMatch = filename.match(/(.*\/)[^/]+$/);
+		if (prefixPathMatch && prefixPathMatch[1]) {
+			prefixPath = prefixPathMatch[1];
+		}
 		Object.assign(resourceInfo, {
+			prefixPath,
 			filename: entry.filename,
 			name,
 			url: entry.comment,
@@ -142,43 +144,57 @@ async function extract(content, { password, prompt = () => { }, shadowRootScript
 	textResources.sort(sortByFilenameLengthInc);
 	resources = resources.sort(sortByFilenameLengthDec).concat(...textResources).concat(...indexPages);
 	for (const resource of resources) {
-		let { textContent, mimeType, filename } = resource;
+		const { filename, prefixPath } = resource;
+		let { textContent } = resource;
 		if (textContent !== undefined) {
-			let prefixPath = "";
-			const prefixPathMatch = filename.match(/(.*\/)[^/]+$/);
-			if (prefixPathMatch && prefixPathMatch[1]) {
-				prefixPath = prefixPathMatch[1];
-			}
 			if (filename.match(REGEXP_MATCH_ROOT_INDEX)) {
 				origDocContent = textContent;
 			}
 			if (!filename.match(REGEXP_MATCH_SCRIPT)) {
-				const resourceFilename = filename;
 				resources.forEach(innerResource => {
 					const { filename, parentResources, content } = innerResource;
+					if (filename.startsWith(prefixPath) && filename != resource.filename) {
+						const relativeFilename = filename.substring(prefixPath.length);
+						if (!relativeFilename.match(REGEXP_MATCH_MANIFEST)) {
+							if (textContent.includes(relativeFilename)) {
+								parentResources.push(resource.filename);
+								if (innerResource.textContent === undefined) {
+									textContent = replaceAll(textContent, relativeFilename, content);
+								}
+							}
+						}
+					}
+				});
+				resource.textContent = textContent;
+			}
+		}
+	}
+	for (const resource of resources) {
+		let { textContent, prefixPath, filename } = resource;
+		if (textContent !== undefined) {
+			if (!filename.match(REGEXP_MATCH_SCRIPT)) {
+				const resourceFilename = filename;
+				for (const innerResource of resources) {
+					const { filename } = innerResource;
 					if (filename.startsWith(prefixPath) && filename != resourceFilename) {
 						const relativeFilename = filename.substring(prefixPath.length);
-						if (!relativeFilename.match(/manifest\.json$/)) {
+						if (!relativeFilename.match(REGEXP_MATCH_MANIFEST)) {
 							const position = textContent.indexOf(relativeFilename);
 							if (position != -1) {
-								parentResources.push(resourceFilename);
-								textContent = replaceAll(textContent, relativeFilename, content);
+								innerResource.content = await getContent(innerResource);
+								textContent = replaceAll(textContent, relativeFilename, innerResource.content);
 							}
 						}
 					}
-				});
+				}
 				resource.textContent = textContent;
+				resource.content = await getContent(resource);
 			}
 			if (filename.match(REGEXP_MATCH_INDEX)) {
 				if (shadowRootScriptURL) {
 					resource.textContent = textContent.replace(/<script data-template-shadow-root.*<\/script>/g, "<script data-template-shadow-root src=" + shadowRootScriptURL + "></" + "script>");
 				}
 			}
-			if (filename.match(REGEXP_MATCH_INDEX) || filename.match(REGEXP_MATCH_FRAMES) || noBlobURL) {
-				resource.content = await getDataURI(textContent, mimeType);
-			} else {
-				resource.content = URL.createObjectURL(new Blob([textContent], { type: mimeType }));
-			}
 			if (filename.match(REGEXP_MATCH_ROOT_INDEX)) {
 				docContent = textContent;
 				url = resource.url;
@@ -187,6 +203,10 @@ async function extract(content, { password, prompt = () => { }, shadowRootScript
 	}
 	return { docContent, origDocContent, resources, url };
 
+	async function getContent(resource) {
+		return resource.filename.match(REGEXP_MATCH_FRAMES) || noBlobURL ? await getDataURI(resource.textContent, resource.mimeType) : URL.createObjectURL(new Blob([resource.textContent], { type: resource.mimeType }));
+	}
+
 	async function getDataURI(textContent, mimeType) {
 		const reader = new FileReader();
 		reader.readAsDataURL(new Blob([textContent], { type: mimeType }));

package/processors/compression/compression.js

@@ -52,13 +52,13 @@ const EMBEDDED_IMAGE_DATA_TAGS = [
 	...EXTRA_DATA_TAGS,
 ];
 const EXTRA_DATA_REGEXPS = [
-	[/<\/noscript>/i],
-	[/<\/script>/i],
-	[/<\/xmp>/i],
-	[/<\/plaintext>/i]
+	[/<noscript/i, /<\/noscript>/i],
+	[/<script/i, /<\/script>/i],
+	[/<xmp/i, /<\/xmp>/i],
+	[/<plaintext/i, /<\/plaintext>/i]
 ];
 const EMBEDDED_IMAGE_DATA_REGEXPS = [
-	/-->/i,
+	[/<!--/i, /-->/i],
 	...EXTRA_DATA_REGEXPS,
 ];
 const CRC32_TABLE = new Uint32Array(256).map((_, indexTable) => {
@@ -96,7 +96,7 @@ async function process(pageData, options, lastModDate = new Date()) {
 		await writeData(zipDataWriter.writable, options.embeddedImage.slice(0, PNG_SIGNATURE_LENGTH + PNG_IHDR_LENGTH));
 		if (options.selfExtractingArchive) {
 			const embeddedImageText = embeddedImageData.reduce((text, charCode) => text + String.fromCharCode(charCode), "");
-			const tagIndex = EMBEDDED_IMAGE_DATA_REGEXPS.findIndex(test => !embeddedImageText.match(test));
+			const tagIndex = EMBEDDED_IMAGE_DATA_REGEXPS.findIndex(tests => !embeddedImageText.match(tests[1]));
 			let startTag;
 			[startTag, endTag] = tagIndex == -1 ? ["", ""] : EMBEDDED_IMAGE_DATA_TAGS[tagIndex];
 			const html = getHTMLStartData(pageData, options) + startTag;
@@ -128,8 +128,8 @@ async function process(pageData, options, lastModDate = new Date()) {
 			if (!options.extractDataFromPageTags) {
 				let textContent = "";
 				data.slice(startOffset).forEach(charCode => textContent += String.fromCharCode(charCode));
-				const matchEndTagComment = textContent.match(/-->/i);
-				if (matchEndTagComment) {
+				const matchCommentTags = textContent.match(/<!--/i) || textContent.match(/-->/i);
+				if (matchCommentTags) {
 					return findExtraDataTags(textContent, pageData, options, lastModDate);
 				}
 			}
@@ -296,7 +296,7 @@ function getHTMLStartData(pageData, options) {
 		pageContent += "<meta name=viewport content=" + JSON.stringify(pageData.viewport) + ">";
 	}
 	if (options.insertMetaCSP) {
-		pageContent += "<meta http-equiv=content-security-policy content=\"default-src 'none'; connect-src 'self'; font-src 'self' data: blob:; img-src 'self' data: blob:; style-src 'self' 'unsafe-inline' data: blob:; frame-src 'self' data: blob:; media-src 'self' data: blob:; script-src 'self' 'unsafe-inline' data: blob:; object-src 'self' data: blob:\">";
+		pageContent += "<meta http-equiv=content-security-policy content=\"default-src 'none'; connect-src 'self' data: blob:; font-src 'self' data: blob:; img-src 'self' data: blob:; style-src 'self' 'unsafe-inline' data: blob:; frame-src 'self' data: blob:; media-src 'self' data: blob:; script-src 'self' 'unsafe-inline' data: blob:; object-src 'self' data: blob:\">";
 	}
 	pageContent += "<style>@keyframes display-wait-message{0%{opacity:0}100%{opacity:1}};body{color:transparent};div{color:initial}</style>";
 	pageContent += "<body hidden>";
@@ -308,8 +308,9 @@ function getPageTitle(pageData) {
 }
 
 function findExtraDataTags(textContent, pageData, options, lastModDate, indexExtractDataFromPageTags = 0) {
-	const matchEndTag = textContent.match(EXTRA_DATA_REGEXPS[indexExtractDataFromPageTags]);
-	if (matchEndTag) {
+	const regExpsTag = EXTRA_DATA_REGEXPS[indexExtractDataFromPageTags];
+	const matchTag = textContent.match(regExpsTag[0]) || textContent.match(regExpsTag[1]);
+	if (matchTag) {
 		if (indexExtractDataFromPageTags < EXTRA_DATA_TAGS.length - 1) {
 			return findExtraDataTags(textContent, pageData, options, lastModDate, indexExtractDataFromPageTags + 1);
 		} else {