Fix encoding canonicalization (fixes #156)

ashtuchkin · Apr 7, 2018 · 696be8a · 696be8a
1 parent 67836d5
commit 696be8a
Show file tree

Hide file tree

Showing 2 changed files with 16 additions and 1 deletion.
diff --git a/lib/index.js b/lib/index.js
@@ -65,7 +65,7 @@ iconv.getCodec = function getCodec(encoding) {
         iconv.encodings = require("../encodings"); // Lazy load all encoding definitions.
 
     // Canonicalize encoding name: strip all non-alphanumeric chars and appended year.
-    var enc = (''+encoding).toLowerCase().replace(/[^0-9a-z]|:\d{4}$/g, "");
+    var enc = iconv._canonicalizeEncoding(encoding);
 
     // Traverse iconv.encodings to find actual codec.
     var codecOptions = {};
@@ -108,6 +108,11 @@ iconv.getCodec = function getCodec(encoding) {
     }
 }
 
+iconv._canonicalizeEncoding = function(encoding) {
+    // Canonicalize encoding name: strip all non-alphanumeric chars and appended year.
+    return (''+encoding).toLowerCase().replace(/:\d{4}$|[^0-9a-z]/g, "");
+}
+
 iconv.getEncoder = function getEncoder(encoding, options) {
     var codec = iconv.getCodec(encoding),
         encoder = new codec.encoder(options, codec);

diff --git a/test/main-test.js b/test/main-test.js
@@ -84,3 +84,13 @@ describe("Generic UTF8-UCS2 tests", function() {
         assert.strictEqual(iconv.encode("外国人", "latin1").toString(), "???");
     });
 });
+
+describe("Canonicalize encoding function", function() {
+    it("works with numbers directly", function() {
+        assert.equal(iconv._canonicalizeEncoding(955), "955");
+    });
+
+    it("correctly strips year and non-alpha chars", function() {
+        assert.equal(iconv._canonicalizeEncoding("ISO_8859-5:1988"), "iso88595");
+    });
+});