func Partial(r io.Reader) ([]*html.Node, error) {
b := &html.Node{}
b.Data = "body"
b.DataAtom = atom.Body
b.Type = html.ElementNode
return html.ParseFragment(r, b)
}
// GoodAsText applies some heuristics to make the data look good when displayed
// as simple text. For example, if the data is escaped HTML then other heuristics are
// applied to remove the HTML. For example if the data contains an HTML image tag,
// goodAsText will return the alt text. If nothing good is found then an empty slice is
// returned.
func goodAsText(d []byte) []byte {
unesc := html.UnescapeString(string(d))
nodes, err := html.ParseFragment(strings.NewReader(unesc), bodyNode)
if err != nil {
log.Printf("failed to parse [%s] as HTML: %v", unesc, err)
return d
}
var buf bytes.Buffer
for _, root := range nodes {
walk(root, func(n *html.Node) {
if n.Type == html.TextNode {
buf.WriteString(strings.TrimSpace(n.Data))
return
}
if n := buf.Len(); n > 0 && buf.Bytes()[n-1] != ' ' {
buf.WriteString(" ")
}
if n.DataAtom == atom.Img {
if alt := altTextOrEmpty(n); alt != "" {
buf.WriteString(alt)
}
}
})
}
return buf.Bytes()
}
func strip_html_tags(htm string) string {
reader := strings.NewReader(htm)
root := &html.Node{Type: html.ElementNode, Data: "article", DataAtom: atom.Article}
frags, _ := html.ParseFragment(reader, root)
var txt string
for _, f := range frags {
txt += extract_html_text(f)
}
return txt
}
func (t *minionTransport) ProcessResponse(req *http.Request, resp *http.Response) (*http.Response, error) {
body, err := ioutil.ReadAll(resp.Body)
if err != nil {
// copying the response body did not work
return nil, err
}
bodyNode := &html.Node{
Type: html.ElementNode,
Data: "body",
DataAtom: atom.Body,
}
nodes, err := html.ParseFragment(bytes.NewBuffer(body), bodyNode)
if err != nil {
glog.Errorf("Failed to found <body> node: %v", err)
return resp, err
}
// Define the method to traverse the doc tree and update href node to
// point to correct minion
var updateHRef func(*html.Node)
updateHRef = func(n *html.Node) {
if n.Type == html.ElementNode && n.Data == "a" {
for i, attr := range n.Attr {
if attr.Key == "href" {
Url := &url.URL{
Path: "/proxy/minion/" + req.URL.Host + req.URL.Path + attr.Val,
}
n.Attr[i].Val = Url.String()
break
}
}
}
for c := n.FirstChild; c != nil; c = c.NextSibling {
updateHRef(c)
}
}
newContent := &bytes.Buffer{}
for _, n := range nodes {
updateHRef(n)
err = html.Render(newContent, n)
if err != nil {
glog.Errorf("Failed to render: %v", err)
}
}
resp.Body = ioutil.NopCloser(newContent)
// Update header node with new content-length
// TODO: Remove any hash/signature headers here?
resp.Header.Del("Content-Length")
resp.ContentLength = int64(newContent.Len())
return resp, err
}
func TestCleaner(t *testing.T) {
root := &html.Node{Type: html.ElementNode, Data: "article", DataAtom: atom.Article}
frag := `<p><embed src="http://player.56.com/v_MTAwODQ1MzE2.swf/1030_justin0842.swf" type="application/x-shockwave-flash" width="480" height="405" allowfullscreen="true" allownetworking="all" allowscriptaccess="always"></embed></p> <a href="http://juetuzhi.net/2013/11/jue-tu-9055.html#comments" style="background-color:#FFFFCC; color:#CC0000;" title="看看大家都说了些什么"><strong>已有 0 人发表评论，猛击这里参与讨论</strong></a>。 <hr style="height:1px;"> © 2007-2011 <a href="http://juetuzhi.net/" target="_blank">掘图志</a> | <a href="http://feeds.juetuzhi.net/" target="_blank" title="订阅掘图志，精彩图片送上门">订阅</a> | <a href="http://juetuzhi.net/2013/11/jue-tu-9055.html" target="_blank" title="本文链接">本文链接</a> | <a href="http://weibo.com/geuro" title="关注掘图志的新浪微博">新浪微博</a> | <a href="http://t.qq.com/juetuzhi" title="关注掘图志的腾讯微博">腾讯微博</a><br /><table class="wumii-related-items" cellspacing="0" cellpadding="3" border="0" style="clear: both;"> <tr> <td colspan="5"><b><font size="-1" style="display: block !important; padding: 20px 0 5px !important;">您也许还喜欢：</font></b></td> </tr> <tr> <td width="126" valign="top" style="padding: 5px !important; margin: 0 !important;"> <a target="_blank" title="[视频]年度最佳光棍节歌曲！看，这就是好男人的下场！" style="text-decoration: none !important; cursor: pointer !important;" href="http://app.wumii.com/ext/redirect?url=http%3A%2F%2Fjuetuzhi.net%2F2011%2F09%2Fhao-nan-ren-de-xia-chang.html&from=http%3A%2F%2Fjuetuzhi.net%2F2013%2F11%2Fjue-tu-9055.html"> <img style="margin: 0 !important; padding: 2px !important; border: 1px solid #DDDDDD !important; width: 120px !important; height: 120px !important;" src="http://static.wumii.cn/site_images/ti/jDuuAFV4.jpg?i=MG2m21AQ" width="120px" height="120px" /><br /> <font size="-1" style="display: block !important; line-height: 15px !important; width: 126px !important; font: 12px/15px arial !important; height: 60px !important; margin: 3px 0 0 0 !important; padding: 0 !important; overflow: hidden !important;">[视频]年度最佳光棍节歌曲！看，这就是好男人的下场！</font> </a> </td> <td width="126" valign="top" style="padding: 5px !important; margin: 0 !important; border-left: 1px solid #DDDDDD !important;"> <a target="_blank" title="你就一辈子做宅男吧！" style="text-decoration: none !important; cursor: pointer !important;" href="http://app.wumii.com/ext/redirect?url=http%3A%2F%2Fjuetuzhi.net%2F2012%2F06%2Flei-ren-88.html&from=http%3A%2F%2Fjuetuzhi.net%2F2013%2F11%2Fjue-tu-9055.html"> <img style="margin: 0 !important; padding: 2px !important; border: 1px solid #DDDDDD !important; width: 120px !important; height: 120px !important;" src="http://static.wumii.cn/resources/images/related_item_default/31.jpg" width="120px" height="120px" /><br /> <font size="-1" style="display: block !important; line-height: 15px !important; width: 126px !important; font: 12px/15px arial !important; height: 60px !important; margin: 3px 0 0 0 !important; padding: 0 !important; overflow: hidden !important;">你就一辈子做宅男吧！</font> </a> </td> <td width="126" valign="top" style="padding: 5px !important; margin: 0 !important; border-left: 1px solid #DDDDDD !important;"> <a target="_blank" title="好萌的小车！" style="text-decoration: none !important; cursor: pointer !important;" href="http://app.wumii.com/ext/redirect?url=http%3A%2F%2Fjuetuzhi.net%2F2012%2F09%2Flei-ren-761.html&from=http%3A%2F%2Fjuetuzhi.net%2F2013%2F11%2Fjue-tu-9055.html"> <img style="margin: 0 !important; padding: 2px !important; border: 1px solid #DDDDDD !important; width: 120px !important; height: 120px !important;" src="http://static.wumii.cn/resources/images/related_item_default/18.jpg" width="120px" height="120px" /><br /> <font size="-1" style="display: block !important; line-height: 15px !important; width: 126px !important; font: 12px/15px arial !important; height: 60px !important; margin: 3px 0 0 0 !important; padding: 0 !important; overflow: hidden !important;">好萌的小车！</font> </a> </td> <td width="126" valign="top" style="padding: 5px !important; margin: 0 !important; border-left: 1px solid #DDDDDD !important;"> <a target="_blank" title="好萌的牌子" style="text-decoration: none !important; cursor: pointer !important;" href="http://app.wumii.com/ext/redirect?url=http%3A%2F%2Fjuetuzhi.net%2F2013%2F10%2Fjue-tu-8007.html&from=http%3A%2F%2Fjuetuzhi.net%2F2013%2F11%2Fjue-tu-9055.html"> <img style="margin: 0 !important; padding: 2px !important; border: 1px solid #DDDDDD !important; width: 120px !important; height: 120px !important;" src="http://static.wumii.cn/site_images/ti/8cEhGbW.jpg?i=2VfiBF1v" width="120px" height="120px" /><br /> <font size="-1" style="display: block !important; line-height: 15px !important; width: 126px !important; font: 12px/15px arial !important; height: 60px !important; margin: 3px 0 0 0 !important; padding: 0 !important; overflow: hidden !important;">好萌的牌子</font> </a> </td> <td width="126" valign="top" style="padding: 5px !important; margin: 0 !important; border-left: 1px solid #DDDDDD !important;"> <a target="_blank" title="好！有骨气！！！" style="text-decoration: none !important; cursor: pointer !important;" href="http://app.wumii.com/ext/redirect?url=http%3A%2F%2Fjuetuzhi.net%2F2012%2F09%2Flei-ren-733.html&from=http%3A%2F%2Fjuetuzhi.net%2F2013%2F11%2Fjue-tu-9055.html"> <img style="margin: 0 !important; padding: 2px !important; border: 1px solid #DDDDDD !important; width: 120px !important; height: 120px !important;" src="http://static.wumii.cn/site_images/ti/Omwy91sa.jpg?i=CfjL4Eba" width="120px" height="120px" /><br /> <font size="-1" style="display: block !important; line-height: 15px !important; width: 126px !important; font: 12px/15px arial !important; height: 60px !important; margin: 3px 0 0 0 !important; padding: 0 !important; overflow: hidden !important;">好！有骨气！！！</font> </a> </td> </tr> <tr> <td colspan="5" align="right"> <a style="text-decoration: none !important;" href="http://www.wumii.com/widget/relatedItems" target="_blank" title="无觅关联推荐"> <font size="-1" color="#bbbbbb" style="display: block !important; font-family: arial !important; padding: 5px 0 !important; font-size: 12px !important; color: #bbb !important;">无觅</font> </a> </td> </tr> </table><img width='1' height='1' src='http://juetuzhi.feedsportal.com/c/34719/f/638620/s/33ccfe03/sc/17/mf.gif' border='0'/><br clear='all'/><br/><br/><a href="http://da.feedsportal.com/r/180264446218/u/0/f/638620/c/34719/s/33ccfe03/a2.htm"><img src="http://da.feedsportal.com/r/180264446218/u/0/f/638620/c/34719/s/33ccfe03/a2.img" border="0"/></a><img width="1" height="1" src="http://pi.feedsportal.com/r/180264446218/u/0/f/638620/c/34719/s/33ccfe03/a2t.img" border="0"/>`
nodes, _ := html.ParseFragment(strings.NewReader(frag), root)
for _, n := range nodes {
root.AppendChild(n)
}
article, _, _ := NewExtractor("").MakeFragmentReadable(root)
// log.Println(sum)
print_html_doc(article)
}
func html_create_fragment(fulltext string) (*html.Node, error) {
reader := strings.NewReader(fulltext)
v := &html.Node{Type: html.ElementNode, Data: "article", DataAtom: atom.Article}
frags, err := html.ParseFragment(reader, v)
if err != nil {
return v, err
}
for _, frag := range frags {
v.AppendChild(frag)
}
return v, err
}
func tidyHtml(input []byte) ([]byte, error) {
// tidy
nodes, err := html.ParseFragment(bytes.NewReader(input), nil)
if err != nil {
return nil, err
}
buf := new(bytes.Buffer)
for _, node := range nodes {
err = html.Render(buf, node)
if err != nil {
return nil, err
}
}
return buf.Bytes(), nil
}
// GetBodyNode returns an BODY node nested within an HTML node.
func getBodyNode() *html.Node {
ns, err := html.ParseFragment(strings.NewReader("<html><body></body></html>"), nil)
if err != nil {
panic("error generating context")
}
if len(ns) == 0 {
panic("no nodes generating context")
}
h := ns[0]
if h.Type != html.ElementNode || h.DataAtom != atom.Html {
panic("expected an HTML node, got " + pretty.String(h))
}
b := h.LastChild
if b.Type != html.ElementNode || b.DataAtom != atom.Body {
panic("expected a BODY node, got " + pretty.String(b))
}
return b
}
func ConvertHtmlToMarkdown(in []byte, rewriteUrl UrlRewriter) ([]byte, error) {
// parse it!
body := &html.Node{
Type: html.ElementNode,
DataAtom: atom.Body,
Data: "body",
}
reader := bytes.NewReader(in)
elems, err := html.ParseFragment(reader, body)
if err != nil {
return nil, err
}
if reader.Len() != 0 {
return nil, errors.New("Post couldn't be fully parsed!")
}
// stuff it all into the body node so we have a proper tree.
for _, elem := range elems {
body.AppendChild(elem)
}
// process shortcodes and WP-LaTeX markup.
if err = shortcode.ProcessShortcodes(body); err != nil {
return nil, err
}
shortcode.ProcessWpLatex(body)
// render it back
wr := &writer{RewriteUrl: rewriteUrl}
for elem := body.FirstChild; elem != nil; elem = elem.NextSibling {
err = renderElement(wr, elem, -1)
if err != nil {
return nil, err
}
}
wr.handleDelayedLf()
return wr.Bytes(), nil
}
func (c *Context) htmlParseFragment(call otto.FunctionCall) otto.Value {
fragment := call.Argument(0).String()
var ctx *html.Node
arg1, _ := call.Argument(1).Export()
if c, ok := arg1.(*node); ok {
ctx = c.node
}
nodes, err := html.ParseFragment(strings.NewReader(fragment), ctx)
if err != nil {
c.Errorf("error parsing HTML fragment: %s\n", err)
return otto.Value{}
}
values := make([]*node, len(nodes))
for ii, v := range nodes {
values[ii] = asNode(v, c.vm)
}
val, err := c.vm.ToValue(values)
if err != nil {
panic(err)
}
return val
}
func ParseHTML5(filename string) ([]*html.Node, error) {
reader, err := os.Open(filename)
frag, err := html.ParseFragment(reader, bodyNode)
return frag, err
}
