ISO-2022-JP (JIS) に対応 (#792)

mei23 · Dec 25, 2024 · bcd3994 · bcd3994
1 parent f15d6e5
commit bcd3994
Show file tree

Hide file tree

Showing 4 changed files with 19 additions and 0 deletions.
diff --git a/built/utils/encoding.js b/built/utils/encoding.js
@@ -4,6 +4,7 @@ exports.detectEncoding = detectEncoding;
 exports.toUtf8 = toUtf8;
 const iconv = require("iconv-lite");
 const jschardet = require("jschardet");
+const Encoding = require('encoding-japanese');
 const regCharset = new RegExp(/charset\s*=\s*["']?([\w-]+)/, 'i');
 /**
  * Detect HTML encoding
@@ -30,9 +31,14 @@ function detectEncoding(body) {
     return 'utf-8';
 }
 function toUtf8(body, encoding) {
+    if (encoding === 'ISO-2022-JP') {
+        return Encoding.codeToString(Encoding.convert(body, 'UNICODE', encoding));
+    }
     return iconv.decode(body, encoding);
 }
 function toEncoding(candicate) {
+    if (candicate.toUpperCase() === 'ISO-2022-JP')
+        return 'ISO-2022-JP';
     if (iconv.encodingExists(candicate)) {
         if (['shift_jis', 'shift-jis', 'windows-31j', 'x-sjis'].includes(candicate.toLowerCase()))
             return 'cp932';

diff --git a/package.json b/package.json
@@ -24,6 +24,7 @@
 	},
 	"dependencies": {
 		"cheerio": "1.0.0",
+		"encoding-japanese": "2.2.0",
 		"escape-regexp": "0.0.1",
 		"got": "11.8.6",
 		"h3": "1.13.0",

diff --git a/src/utils/encoding.ts b/src/utils/encoding.ts
@@ -1,5 +1,6 @@
 import * as iconv from 'iconv-lite';
 import * as jschardet from 'jschardet';
+const Encoding = require('encoding-japanese');
 
 const regCharset = new RegExp(/charset\s*=\s*["']?([\w-]+)/, 'i');
 
@@ -29,10 +30,16 @@ export function detectEncoding(body: Buffer): string {
 }
 
 export function toUtf8(body: Buffer, encoding: string): string {
+	if (encoding === 'ISO-2022-JP') {
+		return Encoding.codeToString(Encoding.convert(body, 'UNICODE', encoding))
+	}
+
 	return iconv.decode(body, encoding);
 }
 
 function toEncoding(candicate: string): string | null {
+	if (candicate.toUpperCase() === 'ISO-2022-JP') return 'ISO-2022-JP';
+
 	if (iconv.encodingExists(candicate)) {
 		if (['shift_jis', 'shift-jis', 'windows-31j', 'x-sjis'].includes(candicate.toLowerCase())) return 'cp932';
 		return candicate;

diff --git a/yarn.lock b/yarn.lock
@@ -331,6 +331,11 @@ domutils@^3.0.1, domutils@^3.1.0:
     domelementtype "^2.3.0"
     domhandler "^5.0.3"
 
+[email protected]:
+  version "2.2.0"
+  resolved "https://registry.yarnpkg.com/encoding-japanese/-/encoding-japanese-2.2.0.tgz#0ef2d2351250547f432a2dd155453555c16deb59"
+  integrity sha512-EuJWwlHPZ1LbADuKTClvHtwbaFn4rOD+dRAbWysqEOXRc2Uui0hJInNJrsdH0c+OhJA4nrCBdSkW4DD5YxAo6A==
+
 encoding-sniffer@^0.2.0:
   version "0.2.0"
   resolved "https://registry.yarnpkg.com/encoding-sniffer/-/encoding-sniffer-0.2.0.tgz#799569d66d443babe82af18c9f403498365ef1d5"