func (this *Feed) isAtom(doc *xmlx.Document) bool {
if doc.SelectNode("http://www.w3.org/2005/Atom", "feed") != nil ||
doc.SelectNode("http://purl.org/atom/ns#", "feed") != nil {
return true
}
return false
}
// Returns the type of the feed, ie. "atom" or "rss", and the version number as an array.
// The first item in the array is the major and the second the minor version number.
func (this *Feed) GetVersionInfo(doc *xmlx.Document) (ftype string, fversion [2]int) {
var node *xmlx.Node
if this.isAtom(doc) {
ftype = "atom"
fversion = [2]int{1, 0}
return
}
if this.isRss(doc) {
ftype = "rss"
major := 0
minor := 0
version := node.As("", "version")
p := strings.Index(version, ".")
if p != -1 {
major, _ = strconv.Atoi(version[0:p])
minor, _ = strconv.Atoi(version[p+1 : len(version)])
}
fversion = [2]int{major, minor}
return
}
// issue#5: Some documents have an RDF root node instead of rss.
if node = doc.SelectNode("http://www.w3.org/1999/02/22-rdf-syntax-ns#", "RDF"); node != nil {
ftype = "rss"
fversion = [2]int{1, 1}
return
}
ftype = "unknown"
fversion = [2]int{0, 0}
return
}
func parser(doc *xmlx.Document) *RssXml {
rss := &RssXml{
Version: "2.0",
}
const ns = "*"
node := doc.SelectNode(ns, "rss")
channel := node.SelectNode(ns, "channel")
feed := &RssFeed{}
feed.Title = channel.S(ns, "title")
feed.Link = channel.S(ns, "link")
feed.Description = channel.S(ns, "description")
feed.Language = channel.S(ns, "language")
feed.Copyright = channel.S(ns, "copyright")
feed.PubDate = channel.S(ns, "pubDate")
items := channel.SelectNodes(ns, "item")
for _, i := range items {
item := &RssItem{}
item.Title = i.S(ns, "title")
item.Link = i.S(ns, "link")
item.Description = i.S(ns, "description")
item.PubDate = i.S(ns, "pubDate")
feed.items = append(feed.items, item)
}
rss.Channel = feed
return rss
}
func readRss2(doc *xmlx.Document) (foundChannels []*Channel, err error) {
const ns = "*"
root := doc.SelectNode(ns, "rss")
if root == nil {
root = doc.SelectNode(ns, "RDF")
}
if root == nil {
return foundChannels, errors.New("Failed to find rss/rdf node in XML.")
}
for _, node := range root.SelectNodes(ns, "channel") {
foundChannels = append(foundChannels, readRssChannel(ns, doc, node))
}
return foundChannels, err
}
func GetVersionInfo(doc *xmlx.Document) (string, [2]int) {
if node := doc.SelectNode("http://www.w3.org/2005/Atom", "feed"); node != nil {
return "atom", [2]int{1, 0}
}
if node := doc.SelectNode("", "rss"); node != nil {
version := node.As("", "version")
p := strings.Index(version, ".")
major, _ := strconv.Atoi(version[0:p])
minor, _ := strconv.Atoi(version[p+1 : len(version)])
return "rss", [2]int{major, minor}
}
// issue#5: Some documents have an RDF root node instead of rss.
if node := doc.SelectNode("http://www.w3.org/1999/02/22-rdf-syntax-ns#", "RDF"); node != nil {
return "rss", [2]int{1, 1}
}
return "unknown", [2]int{0, 0}
}
func (this *Feed) readRss2(doc *xmlx.Document) (err error) {
var foundChannels []*Channel
var ch *Channel
var i *Item
var n *xmlx.Node
var list, tl []*xmlx.Node
const ns = "*"
root := doc.SelectNode(ns, "rss")
if root == nil {
root = doc.SelectNode(ns, "RDF")
}
if root == nil {
return errors.New("Failed to find rss/rdf node in XML.")
}
channels := root.SelectNodes(ns, "channel")
for _, node := range channels {
ch = new(Channel)
foundChannels = append(foundChannels, ch)
ch.Title = node.S(ns, "title")
list = node.SelectNodes(ns, "link")
ch.Links = make([]Link, len(list))
for i, v := range list {
if v.Name.Space == "http://www.w3.org/2005/Atom" && v.Name.Local == "link" {
ch.Links[i].Href = v.As("", "href")
ch.Links[i].Rel = v.As("", "rel")
ch.Links[i].Type = v.As("", "type")
ch.Links[i].HrefLang = v.As("", "hreflang")
} else {
ch.Links[i].Href = v.GetValue()
}
}
ch.Description = node.S(ns, "description")
ch.Language = node.S(ns, "language")
ch.Copyright = node.S(ns, "copyright")
ch.ManagingEditor = node.S(ns, "managingEditor")
ch.WebMaster = node.S(ns, "webMaster")
ch.PubDate = node.S(ns, "pubDate")
ch.LastBuildDate = node.S(ns, "lastBuildDate")
ch.Docs = node.S(ns, "docs")
list = node.SelectNodes(ns, "category")
ch.Categories = make([]*Category, len(list))
for i, v := range list {
ch.Categories[i] = new(Category)
ch.Categories[i].Domain = v.As(ns, "domain")
ch.Categories[i].Text = v.GetValue()
}
if n = node.SelectNode(ns, "generator"); n != nil {
ch.Generator = Generator{}
ch.Generator.Text = n.GetValue()
}
ch.TTL = node.I(ns, "ttl")
ch.Rating = node.S(ns, "rating")
list = node.SelectNodes(ns, "hour")
ch.SkipHours = make([]int, len(list))
for i, v := range list {
ch.SkipHours[i] = v.I(ns, "hour")
}
list = node.SelectNodes(ns, "days")
ch.SkipDays = make([]int, len(list))
for i, v := range list {
ch.SkipDays[i] = days[v.GetValue()]
}
if n = node.SelectNode(ns, "image"); n != nil {
ch.Image.Title = n.S(ns, "title")
ch.Image.Url = n.S(ns, "url")
ch.Image.Link = n.S(ns, "link")
ch.Image.Width = n.I(ns, "width")
ch.Image.Height = n.I(ns, "height")
ch.Image.Description = n.S(ns, "description")
}
if n = node.SelectNode(ns, "cloud"); n != nil {
ch.Cloud = Cloud{}
ch.Cloud.Domain = n.As(ns, "domain")
ch.Cloud.Port = n.Ai(ns, "port")
ch.Cloud.Path = n.As(ns, "path")
ch.Cloud.RegisterProcedure = n.As(ns, "registerProcedure")
ch.Cloud.Protocol = n.As(ns, "protocol")
}
if n = node.SelectNode(ns, "textInput"); n != nil {
ch.TextInput = Input{}
ch.TextInput.Title = n.S(ns, "title")
ch.TextInput.Description = n.S(ns, "description")
ch.TextInput.Name = n.S(ns, "name")
ch.TextInput.Link = n.S(ns, "link")
}
list = node.SelectNodes(ns, "item")
if len(list) == 0 {
list = doc.SelectNodes(ns, "item")
}
for _, item := range list {
i = new(Item)
i.Title = item.S(ns, "title")
i.Description = item.S(ns, "description")
i.Latitude = item.S(ns, "lat")
i.Longitude = item.S(ns, "long")
tl = item.SelectNodes(ns, "link")
for _, v := range tl {
lnk := new(Link)
if v.Name.Space == "http://www.w3.org/2005/Atom" && v.Name.Local == "link" {
lnk.Href = v.As("", "href")
lnk.Rel = v.As("", "rel")
lnk.Type = v.As("", "type")
lnk.HrefLang = v.As("", "hreflang")
} else {
lnk.Href = v.GetValue()
}
i.Links = append(i.Links, lnk)
}
if n = item.SelectNode(ns, "author"); n != nil {
i.Author.Name = n.GetValue()
} else if n = item.SelectNode(ns, "creator"); n != nil {
i.Author.Name = n.GetValue()
}
i.Comments = item.S(ns, "comments")
guid := item.S(ns, "guid")
if len(guid) > 0 {
i.Guid = &guid
}
i.PubDate = item.S(ns, "pubDate")
tl = item.SelectNodes(ns, "category")
for _, lv := range tl {
cat := new(Category)
cat.Domain = lv.As(ns, "domain")
cat.Text = lv.GetValue()
i.Categories = append(i.Categories, cat)
}
tl = item.SelectNodes(ns, "enclosure")
for _, lv := range tl {
enc := new(Enclosure)
enc.Url = lv.As(ns, "url")
enc.Length = lv.Ai64(ns, "length")
enc.Type = lv.As(ns, "type")
i.Enclosures = append(i.Enclosures, enc)
}
if src := item.SelectNode(ns, "source"); src != nil {
i.Source = new(Source)
i.Source.Url = src.As(ns, "url")
i.Source.Text = src.GetValue()
}
tl = item.SelectNodes("http://purl.org/rss/1.0/modules/content/", "*")
for _, lv := range tl {
if lv.Name.Local == "encoded" {
i.Content = new(Content)
i.Content.Text = lv.String()
break
}
}
tl = item.SelectNodes(ns, ns)
i.Extensions = make(map[string]map[string][]Extension)
for _, lv := range tl {
getExtensions(&i.Extensions, lv)
}
ch.Items = append(ch.Items, i)
}
x := node.SelectNodes(ns, ns)
ch.Extensions = make(map[string]map[string][]Extension)
for _, v := range x {
if v.Name.Space != "" {
getExtensions(&ch.Extensions, v)
}
}
}
this.Channels = foundChannels
return
}
func (this *Feed) readRss2(doc *xmlx.Document) (err error) {
days := make(map[string]int)
days["Monday"] = 1
days["Tuesday"] = 2
days["Wednesday"] = 3
days["Thursday"] = 4
days["Friday"] = 5
days["Saturday"] = 6
days["Sunday"] = 7
getChan := func(pubdate, title string) *Channel {
for _, c := range this.Channels {
switch {
case len(pubdate) > 0:
if c.PubDate == pubdate {
return c
}
case len(title) > 0:
if c.Title == title {
return c
}
}
}
return nil
}
var ch *Channel
var i *Item
var n *xmlx.Node
var list, tl []*xmlx.Node
const ns = "*"
root := doc.SelectNode(ns, "rss")
if root == nil {
root = doc.SelectNode(ns, "RDF")
}
if root == nil {
return errors.New("Failed to find rss/rdf node in XML.")
}
channels := root.SelectNodes(ns, "channel")
for _, node := range channels {
if ch = getChan(node.S(ns, "pubDate"), node.S(ns, "title")); ch == nil {
ch = new(Channel)
this.Channels = append(this.Channels, ch)
}
ch.Title = node.S(ns, "title")
list = node.SelectNodes(ns, "link")
ch.Links = make([]Link, len(list))
for i, v := range list {
ch.Links[i].Href = v.GetValue()
}
ch.Description = node.S(ns, "description")
ch.Language = node.S(ns, "language")
ch.Copyright = node.S(ns, "copyright")
ch.ManagingEditor = node.S(ns, "managingEditor")
ch.WebMaster = node.S(ns, "webMaster")
ch.PubDate = node.S(ns, "pubDate")
ch.LastBuildDate = node.S(ns, "lastBuildDate")
ch.Docs = node.S(ns, "docs")
list = node.SelectNodes(ns, "category")
ch.Categories = make([]*Category, len(list))
for i, v := range list {
ch.Categories[i] = new(Category)
ch.Categories[i].Domain = v.As(ns, "domain")
ch.Categories[i].Text = v.GetValue()
}
if n = node.SelectNode(ns, "generator"); n != nil {
ch.Generator = Generator{}
ch.Generator.Text = n.GetValue()
}
ch.TTL = node.I(ns, "ttl")
ch.Rating = node.S(ns, "rating")
list = node.SelectNodes(ns, "hour")
ch.SkipHours = make([]int, len(list))
for i, v := range list {
ch.SkipHours[i] = v.I(ns, "hour")
}
list = node.SelectNodes(ns, "days")
ch.SkipDays = make([]int, len(list))
for i, v := range list {
ch.SkipDays[i] = days[v.GetValue()]
}
if n = node.SelectNode(ns, "image"); n != nil {
ch.Image.Title = n.S(ns, "title")
ch.Image.Url = n.S(ns, "url")
ch.Image.Link = n.S(ns, "link")
ch.Image.Width = n.I(ns, "width")
ch.Image.Height = n.I(ns, "height")
ch.Image.Description = n.S(ns, "description")
}
if n = node.SelectNode(ns, "cloud"); n != nil {
ch.Cloud = Cloud{}
ch.Cloud.Domain = n.As(ns, "domain")
ch.Cloud.Port = n.Ai(ns, "port")
ch.Cloud.Path = n.As(ns, "path")
ch.Cloud.RegisterProcedure = n.As(ns, "registerProcedure")
ch.Cloud.Protocol = n.As(ns, "protocol")
}
if n = node.SelectNode(ns, "textInput"); n != nil {
ch.TextInput = Input{}
ch.TextInput.Title = n.S(ns, "title")
ch.TextInput.Description = n.S(ns, "description")
ch.TextInput.Name = n.S(ns, "name")
ch.TextInput.Link = n.S(ns, "link")
}
itemcount := len(ch.Items)
list = node.SelectNodes(ns, "item")
if len(list) == 0 {
list = doc.SelectNodes(ns, "item")
}
for _, item := range list {
i = new(Item)
i.Title = item.S(ns, "title")
i.Description = item.S(ns, "description")
tl = item.SelectNodes(ns, "link")
for _, v := range tl {
lnk := new(Link)
lnk.Href = v.GetValue()
i.Links = append(i.Links, lnk)
}
if n = item.SelectNode(ns, "author"); n != nil {
i.Author = Author{}
i.Author.Name = n.GetValue()
}
if n = item.SelectNode(ns, "creator"); n != nil {
i.Author = Author{Name: n.GetValue()}
}
i.Comments = item.S(ns, "comments")
i.Guid = item.S(ns, "guid")
i.PubDate = item.S(ns, "pubDate")
tl = item.SelectNodes(ns, "category")
for _, lv := range tl {
cat := new(Category)
cat.Domain = lv.As(ns, "domain")
cat.Text = lv.GetValue()
i.Categories = append(i.Categories, cat)
}
tl = item.SelectNodes(ns, "enclosure")
for _, lv := range tl {
enc := new(Enclosure)
enc.Url = lv.As(ns, "url")
enc.Length = lv.Ai64(ns, "length")
enc.Type = lv.As(ns, "type")
i.Enclosures = append(i.Enclosures, enc)
}
if src := item.SelectNode(ns, "source"); src != nil {
i.Source = new(Source)
i.Source.Url = src.As(ns, "url")
i.Source.Text = src.GetValue()
}
tl = item.SelectNodes("http://purl.org/rss/1.0/modules/content/", "*")
for _, lv := range tl {
if lv.Name.Local == "encoded" {
i.Content = new(Content)
i.Content.Text = lv.String()
break
}
}
ch.Items = append(ch.Items, i)
}
if itemcount != len(ch.Items) && this.itemhandler != nil {
this.itemhandler(this, ch, ch.Items[itemcount:])
}
}
return
}
func (this *Feed) isRss(doc *xmlx.Document) bool {
if doc.SelectNode("", "rss") != nil {
return true
}
return false
}