func (s *Scenario) GetAsset(w *Worker, base *url.URL, node xml.Node, attr string) error {
path, err := url.Parse(node.Attr(attr))
if err != nil {
return w.Fail(nil, err)
}
requestURI := base.ResolveReference(path)
req, res, err := w.SimpleGet(requestURI.String())
if err != nil {
return w.Fail(req, err)
}
if res.StatusCode != 200 {
return w.Fail(res.Request, fmt.Errorf("Response code should be %d, got %d", 200, res.StatusCode))
}
md5sum := calcMD5(res.Body)
defer res.Body.Close()
if expectedMD5, ok := s.ExpectedAssets[requestURI.RequestURI()]; ok {
if md5sum == expectedMD5 {
w.Success(StaticFileScore)
} else {
return w.Fail(res.Request, fmt.Errorf("Expected MD5 checksum is miss match %s, got %s", expectedMD5, md5sum))
}
}
return nil
}
func trNodeToSchedule(scheduleNode xml.Node) (item ScheduleItem, err error) {
results, err := scheduleNode.Search("./td/text()")
if err != nil {
return ScheduleItem{}, err
}
item = ScheduleItem{
TrainNumber: strings.TrimSpace(results[1].String()),
Misc: strings.TrimSpace(results[2].String()),
Class: strings.TrimSpace(results[3].String()),
Relation: strings.TrimSpace(results[4].String()),
StartingStation: strings.TrimSpace(results[5].String()),
CurrentStation: strings.TrimSpace(results[6].String()),
ArrivingTime: strings.TrimSpace(results[7].String()),
DepartingTime: strings.TrimSpace(results[8].String()),
Ls: strings.TrimSpace(results[9].String()),
}
if len(results) > 10 {
item.Status = strings.TrimSpace(results[10].String())
}
stationParts := strings.FieldsFunc(item.Relation, func(r rune) bool {
return r == '-'
})
item.EndStation = stationParts[1] // [ANGKE BOGOR] BOGOR is end station
return
}
func getContent(node gxtml.Node, cssQuery string) string {
result, err := node.Search(toXpath(cssQuery))
if err != nil {
panic(fmt.Errorf("Failed to find %v node", cssQuery))
}
return result[0].Content()
}
func Link(performance xml.Node) string {
anchor, err := performance.Search(".//a")
if err != nil {
fmt.Println(err)
}
return "http://www.bso.org" + anchor[0].Attr("href")
}
func extend(doc *xml.ElementNode, node xml.Node) {
dup := node.Duplicate(1).(*xml.ElementNode)
doc.AddChild(dup)
if strings.ToLower(dup.Name()) != "p" || strings.ToLower(dup.Name()) != "div" {
dup.SetName("div")
}
}
func (s *scrape) MapRegex(node xml.Node) (map[string]string, error) {
if node.IsValid() == false {
return nil, errors.New("Invalid node")
}
m := make(map[string]string, 1)
inner := node.String()
for k, v := range ScrapeRegex {
// remove new line chars
reg, _ := regexp.CompilePOSIX("\r\n|\r|\n")
inner = reg.ReplaceAllString(inner, "")
// get the real data
reg, _ = regexp.CompilePOSIX(v[0])
scraped := reg.FindString(inner)
scraped = reg.ReplaceAllString(scraped, "$1")
if scraped != "" {
m[k] = scraped
}
}
// Skip empty and unwanted
if len(m) > 0 {
if m[ScrapeMeta[IGNOREEMPTY]] != "" {
return m, nil
}
return nil, nil
}
return nil, nil
}
func parseSource(m xml.Node) string {
res, _ := m.Search("source")
if len(res) > 0 {
return res[0].Content()
}
return ""
}
// Walks through the documents elements and populates the buffer.
func (self *Formatter) walk(node xml.Node) {
for c := node.FirstChild(); c != nil; c = c.NextSibling() {
self.walk(c)
}
if node.NodeType() == xml.XML_ELEMENT_NODE {
self.handleNode(node)
}
}
func parseRights(m xml.Node) string {
res, _ := m.Search("rights")
if len(res) > 0 {
return res[0].Content()
}
return ""
}
// Writes code blocks to the buffer.
func (self *Formatter) writeCodeBlock(node xml.Node) {
block := []byte(strings.Trim(node.Content(), "\n\r\v"))
node.SetContent("")
if len(block) == 0 {
return
}
self.buf.Write(block)
self.buf.Write([]byte{'\n', '\n'})
}
func (this *Document) walkElements(node xml.Node, f func(xml.Node) error) error {
f(node)
for child := node.FirstChild(); child != nil; child = child.NextSibling() {
err := this.walkElements(child, f)
if err != nil {
return err
}
}
return nil
}
func parseSubjects(m xml.Node) []string {
subjects := []string{}
res, _ := m.Search("subject")
for _, n := range res {
subjects = append(subjects, n.Content())
}
return subjects
}
func parseDescription(m xml.Node) string {
description := ""
res, _ := m.Search("description")
if len(res) > 0 {
description = res[0].Content()
}
return description
}
func parsePublisher(m xml.Node) string {
publisher := ""
res, _ := m.Search("publisher")
if len(res) > 0 {
publisher = res[0].Content()
}
return publisher
}
func parseTitles(m xml.Node) []string {
titles := []string{}
res, _ := m.Search("title")
for _, n := range res {
titles = append(titles, n.Content())
}
return titles
}
func parseLanguages(m xml.Node) []string {
languages := []string{}
res, _ := m.Search("language")
for _, n := range res {
languages = append(languages, n.Content())
}
return languages
}
func parseIdentifiers(m xml.Node) []*Identifier {
identifiers := []*Identifier{}
res, _ := m.Search("identifier")
for _, n := range res {
identifier := Identifier{Identifier: n.Content(), Scheme: n.Attr("scheme")}
identifiers = append(identifiers, &identifier)
}
return identifiers
}
func Pieces(details xml.Node) string {
pieces, _ := details.Search(".//div[@class='program-media-collapse']/h3")
var piecesString string
piecesString = "<ul class=\"works\">"
for _, piece := range pieces {
piecesString += "<li>"
piecesString += piece.Content()
piecesString += "</li>"
}
piecesString += "</ul>"
return piecesString
}
// unmarshalPath walks down an XML structure looking for wanted
// paths, and calls unmarshal on them.
// The consumed result tells whether XML elements have been consumed
// from the Decoder until start's matching end element, or if it's
// still untouched because start is uninteresting for sv's fields.
func (p *Decoder) unmarshalPath(tinfo *typeInfo, sv reflect.Value, parents []string, start gokoxml.Node) (err error) {
recurse := false
name := start.Name() // For speed
Loop:
for i := range tinfo.fields {
finfo := &tinfo.fields[i]
if finfo.flags&fElement == 0 || len(finfo.parents) < len(parents) {
continue
}
for j := range parents {
if parents[j] != finfo.parents[j] {
continue Loop
}
}
if len(finfo.parents) == len(parents) && finfo.name == name {
// It's a perfect match, unmarshal the field.
return p.unmarshal(sv.FieldByIndex(finfo.idx), start)
}
if len(finfo.parents) > len(parents) && finfo.parents[len(parents)] == name {
// It's a prefix for the field. Break and recurse
// since it's not ok for one field path to be itself
// the prefix for another field path.
recurse = true
// We can reuse the same slice as long as we
// don't try to append to it.
parents = finfo.parents[:len(parents)+1]
break
}
}
if !recurse {
// We have no business with this element.
return nil
}
// The element is not a perfect match for any field, but one
// or more fields have the path to this element as a parent
// prefix. Recurse and attempt to match these.
for cur_node := start.FirstChild(); cur_node != nil; cur_node = cur_node.NextSibling() {
if cur_node.NodeType() != gokoxml.XML_ELEMENT_NODE {
continue
}
if err := p.unmarshalPath(tinfo, sv, parents, cur_node); err != nil {
return err
}
}
// No more XML Nodes.
return nil
}
func linkDensity(node xml.Node) float64 {
links, err := node.Search("a")
if err != nil {
return 0.0
}
llength := 0.0
for _, link := range links {
llength += float64(len(link.Content()))
}
tlength := float64(len(node.Content()))
return llength / tlength
}
// Writes text blocks to the buffer.
func (self *Formatter) writeBlock(node xml.Node, prefix string) {
block := []byte(strings.TrimSpace(node.Content()))
node.SetContent("")
if len(block) == 0 {
return
}
// Position of last space, line break and max length.
sp, br, max := 0, 0, 79-len(prefix)
self.buf.WriteString(prefix)
for i, c := range block {
// Break line if exceeded max length and the position of the last space
// is greater than the position of the last line break. Don't break very
// long words.
if i-br > max && sp > br {
self.buf.WriteByte('\n')
br = sp
// Only the first line is prefixed.
for j := 0; j < len(prefix); j++ {
self.buf.WriteByte(' ')
}
}
if whitespace[c] {
// The last character was a space, so ignore this one.
if sp == i {
sp++
br++
continue
}
// Write the last word to the buffer, append a space and update
// the position of the last space.
if sp > br {
self.buf.WriteByte(' ')
}
self.buf.Write(block[sp:i])
sp = i + 1
}
}
// Write the last word to the buffer.
if sp < len(block) {
if sp > br {
self.buf.WriteByte(' ')
}
self.buf.Write(block[sp:])
}
// Close block with 2 breaks.
self.buf.Write([]byte{'\n', '\n'})
}
/*
* Invokes a MIME type plugin based on current node's type attribute, passing src attribute's value
* as argument. Subcommand's output is piped to Gokogiri through a buffer.
*/
func (gen *Generator) handleMIMETypePlugin(e xml.Node, doc *html.HtmlDocument) (err error) {
src := e.Attribute("src").Value()
typ := e.Attribute("type").Value()
cmd := exec.Command(fmt.Sprintf("m%s%s", ZAS_PREFIX, gen.resolveMIMETypePlugin(typ)), src)
stdout, err := cmd.StdoutPipe()
if err != nil {
return
}
cmd.Stderr = os.Stderr
c := make(chan bufErr)
go func() {
data, err := ioutil.ReadAll(stdout)
c <- bufErr{data, err}
}()
if err = cmd.Start(); err != nil {
return
}
be := <-c
if err = cmd.Wait(); err != nil {
return
}
if be.err != nil {
return be.err
}
parent := e.Parent()
child, err := doc.Coerce(be.buffer)
if err != nil {
return
}
parent.AddChild(child)
e.Remove()
return
}
// parseAppDiv extracts timestamp and blockindex from an appointment div
func parseAppDiv(div xml.Node) (timestamp int64, blockIndex string, err error) {
idValues := idBlockPattern.FindStringSubmatch(div.Attr("id"))
timestamp, err = strconv.ParseInt(idValues[1], 10, 64)
if err != nil {
return
}
blockIndexValues := blockIndexPattern.FindStringSubmatch(div.Content())
if len(blockIndexValues) == 1 {
blockIndex = blockIndexValues[0]
}
return
}
func processNode(node xml.Node, row string) {
row = row + node.Attr("TEXT") + "|"
kids, err := node.Search("node")
if err != nil {
log.Println("Error searching for node:", err)
return
}
if len(kids) > 0 { // has children, not a leaf node
for i := range kids {
processNode(kids[i], row)
}
} else {
fmt.Println(row) // print leaf node
}
}
/*
Not working because:
http://www.sc2ratings.com/players.php?realname=Yang,%20Hee-Soo
is parsed as:
http://www.sc2ratings.com/players.php?realname=Yang, Hee-Soo
*/
func parseLeagues(player xml.Node) []string {
out := []string{}
partialUrl, err := player.Search(".//a/@href")
errorHandler(err)
if len(partialUrl) == 1 {
playerPageUrl := "http://www.sc2ratings.com/" + partialUrl[0].String()
playerPageSource := retrievePageSource(playerPageUrl)
playerPage, err := gokogiri.ParseHtml(playerPageSource)
errorHandler(err)
defer playerPage.Free()
fmt.Println(playerPage)
}
return out
}
func parseDates(m xml.Node) []*Date {
dates := []*Date{}
res, _ := m.Search("date")
for _, n := range res {
date := Date{Date: n.Content(), Event: n.Attr("event")}
dates = append(dates, &date)
}
res, _ = m.Search("meta[@property='dcterms:modified']")
if len(res) > 0 {
date := Date{Date: res[0].Content(), Event: "modified"}
dates = append(dates, &date)
}
return dates
}
func loadInputs(doc xml.Node) ([]*Input, error) {
nodes, e := doc.Search(".//input")
if e != nil {
return nil, e
}
out := []*Input{}
for _, n := range nodes {
i := &Input{
Type: n.Attr("type"),
Name: n.Attr("name"),
Value: n.Attr("value"),
Checked: n.Attr("checked") == "checked",
}
out = append(out, i)
}
return out, nil
}
// ProcessField method fetches data from passed document
func (f *Field) ProcessField(d *html.HtmlDocument) interface{} {
var value interface{}
var node xml.Node
selector := xpath.Compile(f.Selector)
result, _ := d.Root().Search(selector)
if len(result) > 0 {
node = result[0]
} else {
return ""
}
if f.Callback != nil {
value = f.Callback(&node)
} else {
value = node.Content()
}
return value
}
func removeNonAnchorElements(node xml.Node) {
var next xml.Node
for n := node.FirstChild(); n != nil; n = next {
next = n.NextSibling()
if n.NodeType() == xml.XML_ELEMENT_NODE {
removeNonAnchorElements(n)
if strings.ToLower(n.Name()) != "a" {
var chNext xml.Node
for ch := n.FirstChild(); ch != nil; ch = chNext {
chNext = ch.NextSibling()
ch.Unlink()
n.InsertBefore(ch)
}
n.Remove()
}
}
}
}
func parseListApp(div xml.Node) (*Appointment, error) {
values := idListPattern.FindStringSubmatch(div.Attr("id"))
timestamp, err := strconv.ParseInt(values[1], 10, 64)
if err != nil {
return nil, err
}
practitioner, err := strconv.ParseInt(values[2], 10, 64)
if err != nil {
return nil, err
}
return &Appointment{
session: nil,
Timestamp: time.Unix(timestamp, 0),
Practitioner: Practitioner(practitioner),
Status: Booked,
}, nil
}
