miniflux/reader/rss/parser.go
Frédéric Guillot 713b38e34c Handle more encoding edge cases
- 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
2018-01-20 13:25:21 -08:00

28 lines
712 B
Go

// Copyright 2017 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 rss
import (
"encoding/xml"
"io"
"github.com/miniflux/miniflux/errors"
"github.com/miniflux/miniflux/model"
"github.com/miniflux/miniflux/reader/encoding"
)
// Parse returns a normalized feed struct from a RSS feed.
func Parse(data io.Reader) (*model.Feed, error) {
feed := new(rssFeed)
decoder := xml.NewDecoder(data)
decoder.CharsetReader = encoding.CharsetReader
err := decoder.Decode(feed)
if err != nil {
return nil, errors.NewLocalizedError("Unable to parse RSS feed: %v.", err)
}
return feed.Transform(), nil
}