Handle RSS entries with Atom links

This commit is contained in:
Frédéric Guillot 2017-11-20 15:48:26 -08:00
parent cf8af56a99
commit 557cf9c21d
3 changed files with 114 additions and 38 deletions

View file

@ -14,7 +14,7 @@ import (
"golang.org/x/net/html/charset" "golang.org/x/net/html/charset"
) )
// Parse returns a normalized feed struct. // Parse returns a normalized feed struct from a RSS feed.
func Parse(data io.Reader) (*model.Feed, error) { func Parse(data io.Reader) (*model.Feed, error) {
feed := new(rssFeed) feed := new(rssFeed)
decoder := xml.NewDecoder(data) decoder := xml.NewDecoder(data)

View file

@ -161,6 +161,51 @@ func TestParseEntryWithoutLink(t *testing.T) {
} }
} }
func TestParseEntryWithAtomLink(t *testing.T) {
data := `<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<link>https://example.org/</link>
<item>
<title>Test</title>
<atom:link href="https://example.org/item" />
</item>
</channel>
</rss>`
feed, err := Parse(bytes.NewBufferString(data))
if err != nil {
t.Error(err)
}
if feed.Entries[0].URL != "https://example.org/item" {
t.Errorf("Incorrect entry link, got: %s", feed.Entries[0].URL)
}
}
func TestParseEntryWithMultipleAtomLinks(t *testing.T) {
data := `<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<link>https://example.org/</link>
<item>
<title>Test</title>
<atom:link rel="payment" href="https://example.org/a" />
<atom:link rel="http://foobar.tld" href="https://example.org/b" />
</item>
</channel>
</rss>`
feed, err := Parse(bytes.NewBufferString(data))
if err != nil {
t.Error(err)
}
if feed.Entries[0].URL != "https://example.org/b" {
t.Errorf("Incorrect entry link, got: %s", feed.Entries[0].URL)
}
}
func TestParseFeedURLWithAtomLink(t *testing.T) { func TestParseFeedURLWithAtomLink(t *testing.T) {
data := `<?xml version="1.0" encoding="utf-8"?> data := `<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"> <rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
@ -489,3 +534,11 @@ func TestParseEntryWithFeedBurnerEnclosures(t *testing.T) {
t.Errorf("Incorrect enclosure length, got: %d", feed.Entries[0].Enclosures[0].Size) t.Errorf("Incorrect enclosure length, got: %d", feed.Entries[0].Enclosures[0].Size)
} }
} }
func TestParseInvalidXml(t *testing.T) {
data := `garbage`
_, err := Parse(bytes.NewBufferString(data))
if err == nil {
t.Error("Parse should returns an error")
}
}

View file

@ -35,12 +35,13 @@ type rssLink struct {
XMLName xml.Name XMLName xml.Name
Data string `xml:",chardata"` Data string `xml:",chardata"`
Href string `xml:"href,attr"` Href string `xml:"href,attr"`
Rel string `xml:"rel,attr"`
} }
type rssItem struct { type rssItem struct {
GUID string `xml:"guid"` GUID string `xml:"guid"`
Title string `xml:"title"` Title string `xml:"title"`
Link string `xml:"link"` Links []rssLink `xml:"link"`
OriginalLink string `xml:"http://rssnamespace.org/feedburner/ext/1.0 origLink"` OriginalLink string `xml:"http://rssnamespace.org/feedburner/ext/1.0 origLink"`
Description string `xml:"description"` Description string `xml:"description"`
Content string `xml:"http://purl.org/rss/1.0/modules/content/ encoded"` Content string `xml:"http://purl.org/rss/1.0/modules/content/ encoded"`
@ -65,9 +66,9 @@ type rssEnclosure struct {
} }
func (r *rssFeed) GetSiteURL() string { func (r *rssFeed) GetSiteURL() string {
for _, elem := range r.Links { for _, element := range r.Links {
if elem.XMLName.Space == "" { if element.XMLName.Space == "" {
return elem.Data return element.Data
} }
} }
@ -75,9 +76,9 @@ func (r *rssFeed) GetSiteURL() string {
} }
func (r *rssFeed) GetFeedURL() string { func (r *rssFeed) GetFeedURL() string {
for _, elem := range r.Links { for _, element := range r.Links {
if elem.XMLName.Space == "http://www.w3.org/2005/Atom" { if element.XMLName.Space == "http://www.w3.org/2005/Atom" {
return elem.Href return element.Href
} }
} }
@ -111,10 +112,10 @@ func (r *rssFeed) Transform() *model.Feed {
return feed return feed
} }
func (i *rssItem) GetDate() time.Time { func (r *rssItem) GetDate() time.Time {
value := i.PubDate value := r.PubDate
if i.Date != "" { if r.Date != "" {
value = i.Date value = r.Date
} }
if value != "" { if value != "" {
@ -130,8 +131,8 @@ func (i *rssItem) GetDate() time.Time {
return time.Now() return time.Now()
} }
func (i *rssItem) GetAuthor() string { func (r *rssItem) GetAuthor() string {
for _, element := range i.Authors { for _, element := range r.Authors {
if element.Name != "" { if element.Name != "" {
return element.Name return element.Name
} }
@ -141,11 +142,11 @@ func (i *rssItem) GetAuthor() string {
} }
} }
return i.Creator return r.Creator
} }
func (i *rssItem) GetHash() string { func (r *rssItem) GetHash() string {
for _, value := range []string{i.GUID, i.Link} { for _, value := range []string{r.GUID, r.GetURL()} {
if value != "" { if value != "" {
return helper.Hash(value) return helper.Hash(value)
} }
@ -154,33 +155,43 @@ func (i *rssItem) GetHash() string {
return "" return ""
} }
func (i *rssItem) GetContent() string { func (r *rssItem) GetContent() string {
if i.Content != "" { if r.Content != "" {
return i.Content return r.Content
} }
return i.Description return r.Description
} }
func (i *rssItem) GetURL() string { func (r *rssItem) GetURL() string {
if i.OriginalLink != "" { if r.OriginalLink != "" {
return i.OriginalLink return r.OriginalLink
} }
return i.Link for _, link := range r.Links {
if link.XMLName.Space == "http://www.w3.org/2005/Atom" && link.Href != "" && isValidLinkRelation(link.Rel) {
return link.Href
}
if link.Data != "" {
return link.Data
}
}
return ""
} }
func (i *rssItem) GetEnclosures() model.EnclosureList { func (r *rssItem) GetEnclosures() model.EnclosureList {
enclosures := make(model.EnclosureList, 0) enclosures := make(model.EnclosureList, 0)
for _, enclosure := range i.Enclosures { for _, enclosure := range r.Enclosures {
length, _ := strconv.Atoi(enclosure.Length) length, _ := strconv.Atoi(enclosure.Length)
enclosureURL := enclosure.URL enclosureURL := enclosure.URL
if i.OrigEnclosureLink != "" { if r.OrigEnclosureLink != "" {
filename := path.Base(i.OrigEnclosureLink) filename := path.Base(r.OrigEnclosureLink)
if strings.Contains(enclosureURL, filename) { if strings.Contains(enclosureURL, filename) {
enclosureURL = i.OrigEnclosureLink enclosureURL = r.OrigEnclosureLink
} }
} }
@ -194,15 +205,15 @@ func (i *rssItem) GetEnclosures() model.EnclosureList {
return enclosures return enclosures
} }
func (i *rssItem) Transform() *model.Entry { func (r *rssItem) Transform() *model.Entry {
entry := new(model.Entry) entry := new(model.Entry)
entry.URL = i.GetURL() entry.URL = r.GetURL()
entry.Date = i.GetDate() entry.Date = r.GetDate()
entry.Author = i.GetAuthor() entry.Author = r.GetAuthor()
entry.Hash = i.GetHash() entry.Hash = r.GetHash()
entry.Content = processor.ItemContentProcessor(entry.URL, i.GetContent()) entry.Content = processor.ItemContentProcessor(entry.URL, r.GetContent())
entry.Title = sanitizer.StripTags(strings.Trim(i.Title, " \n\t")) entry.Title = sanitizer.StripTags(strings.Trim(r.Title, " \n\t"))
entry.Enclosures = i.GetEnclosures() entry.Enclosures = r.GetEnclosures()
if entry.Title == "" { if entry.Title == "" {
entry.Title = entry.URL entry.Title = entry.URL
@ -210,3 +221,15 @@ func (i *rssItem) Transform() *model.Entry {
return entry return entry
} }
func isValidLinkRelation(rel string) bool {
switch rel {
case "", "alternate", "enclosure", "related", "self", "via":
return true
default:
if strings.HasPrefix(rel, "http") {
return true
}
return false
}
}