fix(routes/pdf): limit outputEncoding qs params to poppler charsets (…

…#1353) * fix(routes/pdf/txt): limit charsets to poppler supported ones only * fix(routes/pdf/html): limit charsets to poppler supported ones only * fix(routes/pdf): add a few extra charsets
Fdawgs · Mar 24, 2023 · 92500d6 · 92500d6
1 parent 3eb0d71
commit 92500d6
Show file tree

Hide file tree

Showing 3 changed files with 51 additions and 7 deletions.
diff --git a/src/plugins/pdf-to-html/index.js b/src/plugins/pdf-to-html/index.js
@@ -150,11 +150,7 @@ async function plugin(server, options) {
 		 * Remove excess title and meta elements left behind by Poppler;
 		 * Poppler appends `-html` to the file name
 		 */
-		const dom = new JSDOM(
-			await fs.readFile(`${tempFile}-html.html`, {
-				encoding: config.pdfToHtmlOptions.outputEncoding,
-			})
-		);
+		const dom = new JSDOM(await fs.readFile(`${tempFile}-html.html`));
 		const titles = dom.window.document.querySelectorAll("title");
 
 		// Overwrite title set by Poppler, which reveals directories

diff --git a/src/routes/pdf/html/schema.js b/src/routes/pdf/html/schema.js
@@ -96,7 +96,31 @@ const pdfToHtmlPostSchema = {
 			S.string()
 				.default("UTF-8")
 				.description("Sets the encoding to use for text output")
-				.pattern(/^[-\w]+$/m)
+				// Encodings supported by Poppler
+				.enum([
+					"ASCII7",
+					"Big5",
+					"Big5ascii",
+					"EUC-CN",
+					"EUC-JP",
+					"GBK",
+					"ISO-2022-CN",
+					"ISO-2022-JP",
+					"ISO-2022-KR",
+					"ISO-8859-6",
+					"ISO-8859-7",
+					"ISO-8859-8",
+					"ISO-8859-9",
+					"KOI8-R",
+					"Latin1",
+					"Latin2",
+					"Shift-JIS",
+					"TIS-620",
+					"UTF-8",
+					"UTF-16",
+					"Windows-1255",
+					"ZapfDingbats",
+				])
 		)
 		.prop(
 			"ownerPassword",

diff --git a/src/routes/pdf/txt/schema.js b/src/routes/pdf/txt/schema.js
@@ -103,7 +103,31 @@ const pdfToTxtPostSchema = {
 			S.string()
 				.default("UTF-8")
 				.description("Sets the encoding to use for text output")
-				.pattern(/^[-\w]+$/m)
+				// Encodings supported by Poppler
+				.enum([
+					"ASCII7",
+					"Big5",
+					"Big5ascii",
+					"EUC-CN",
+					"EUC-JP",
+					"GBK",
+					"ISO-2022-CN",
+					"ISO-2022-JP",
+					"ISO-2022-KR",
+					"ISO-8859-6",
+					"ISO-8859-7",
+					"ISO-8859-8",
+					"ISO-8859-9",
+					"KOI8-R",
+					"Latin1",
+					"Latin2",
+					"Shift-JIS",
+					"TIS-620",
+					"UTF-8",
+					"UTF-16",
+					"Windows-1255",
+					"ZapfDingbats",
+				])
 		)
 		.prop(
 			"ownerPassword",