aboutsummaryrefslogtreecommitdiffhomepage
path: root/reader/encoding/encoding.go
diff options
context:
space:
mode:
Diffstat (limited to 'reader/encoding/encoding.go')
-rw-r--r--reader/encoding/encoding.go38
1 files changed, 38 insertions, 0 deletions
diff --git a/reader/encoding/encoding.go b/reader/encoding/encoding.go
new file mode 100644
index 0000000..7f726ad
--- /dev/null
+++ b/reader/encoding/encoding.go
@@ -0,0 +1,38 @@
+// Copyright 2018 Frédéric Guillot. All rights reserved.
+// Use of this source code is governed by the Apache 2.0
+// license that can be found in the LICENSE file.
+
+package encoding
+
+import (
+ "bytes"
+ "io"
+ "unicode/utf8"
+
+ "golang.org/x/net/html/charset"
+)
+
+// CharsetReader is used when the XML encoding is specified for the input document.
+//
+// The document is converted in UTF-8 only if a different encoding is specified
+// and the document is not already UTF-8.
+//
+// Several edge cases could exists:
+//
+// - Feeds with charset specified only in Content-Type header and not in XML document
+// - Feeds with charset specified in both places
+// - Feeds with charset specified only in XML document and not in HTTP header
+func CharsetReader(label string, input io.Reader) (io.Reader, error) {
+ var buf1, buf2 bytes.Buffer
+ w := io.MultiWriter(&buf1, &buf2)
+ io.Copy(w, input)
+ r := bytes.NewReader(buf2.Bytes())
+
+ if !utf8.Valid(buf1.Bytes()) {
+ // Transform document to UTF-8 from the specified XML encoding.
+ return charset.NewReaderLabel(label, r)
+ }
+
+ // The document is already UTF-8, do not do anything (avoid double-encoding)
+ return r, nil
+}