func (record *Record) processOffJoins(doc *html.HtmlDocument) {
xPath := xpath.NewXPath(doc.DocPtr())
// join the offence with the defendants and verdict
joinPtrs := xPath.Evaluate(doc.Root().NodePtr(),
xpath.Compile("//join[@result='criminalCharge']"))
for _, nodePtr := range joinPtrs {
node := xml.NewNode(nodePtr, doc)
targets := strings.Split(node.Attr("targets"), " ")
var personId, offId, verdictId string
for _, targetId := range targets {
if strings.Contains(targetId, "defend") {
personId = targetId
}
if strings.Contains(targetId, "off") {
offId = targetId
}
if strings.Contains(targetId, "verdict") {
verdictId = targetId
}
}
offence := record.findOffence(offId)
if offence == nil {
panic("couldn't find offence " + offId)
}
person := record.findPerson(personId)
if person != nil {
offence.Defendants = append(offence.Defendants, person)
}
verdict := record.findVerdict(verdictId)
if verdict != nil {
offence.Verdict = verdict
}
}
}
func getNewsInfo(doc *html.HtmlDocument) ([]xml.Node, error) {
xp := "//body/div/div/div/div/div/div/div/div/div/div/div/div/div/div/ul/li"
xps := xpath.Compile(xp)
newDatas, err := doc.Root().Search(xps)
if err != nil {
return nil, err
}
return newDatas, nil
}
func (record *Record) processVerdicts(doc *html.HtmlDocument) {
xPath := xpath.NewXPath(doc.DocPtr())
verdictPtrs := xPath.Evaluate(doc.Root().NodePtr(),
xpath.Compile("//rs[@type='verdictDescription']"))
verdicts := make([]Verdict, len(verdictPtrs))
for i, nodePtr := range verdictPtrs {
node := xml.NewNode(nodePtr, doc)
verdict := Verdict{}
verdict.Id = node.Attr("id")
verdict.Desc = cleanUpContent(node.Content())
verdict.SetType(getInterp(nodePtr, "verdictCategory", doc))
verdicts[i] = verdict
}
record.Verdicts = verdicts
}
func (record *Record) processOffences(doc *html.HtmlDocument) {
xPath := xpath.NewXPath(doc.DocPtr())
offencePtrs := xPath.Evaluate(doc.Root().NodePtr(),
xpath.Compile("//rs[@type='offenceDescription']"))
offences := make([]Offence, len(offencePtrs))
for i, nodePtr := range offencePtrs {
node := xml.NewNode(nodePtr, doc)
offence := Offence{}
offence.Id = node.Attr("id")
offence.Category = getInterp(nodePtr, "offenceCategory", doc)
offence.SubCategory = getInterp(nodePtr, "offenceSubcategory", doc)
offence.Desc = cleanUpContent(node.Content())
offences[i] = offence
}
record.Offences = offences
}
func docSearch(doc *ghtml.HtmlDocument, elementName string, pageName string, xpath string, mustFind bool) []gxml.Node {
elementArray, err := doc.Root().Search(xpath)
if (err != nil || len(elementArray) == 0) && mustFind == false {
return nil
}
if err != nil {
fmt.Fprintf(os.Stderr, "Error locating element \"%s\" in page %s (incorrect xpath?): %v\n", elementName, pageName, err)
//fmt.Fprintf(os.Stderr, " doc=%+v\n", doc)
os.Exit(1)
}
if len(elementArray) == 0 {
fmt.Fprintf(os.Stderr, "Error locating element \"%s\" in page %s (incorrect xpath?): len() == 0\n", elementName, pageName)
//fmt.Fprintf(os.Stderr, " doc=%+v\n", doc)
os.Exit(1)
}
return elementArray
}
func (record *Record) processPersons(doc *html.HtmlDocument) {
xPath := xpath.NewXPath(doc.DocPtr())
personPtrs := xPath.Evaluate(doc.Root().NodePtr(),
xpath.Compile("//persname"))
persons := make([]Person, len(personPtrs))
for i, nodePtr := range personPtrs {
node := xml.NewNode(nodePtr, doc)
person := Person{}
person.Id = node.Attr("id")
person.GivenName = getInterp(nodePtr, "given", doc)
person.Surname = getInterp(nodePtr, "surname", doc)
person.SetType(node.Attr("type"))
person.SetGender(getInterp(nodePtr, "gender", doc))
persons[i] = person
}
record.Persons = persons
}
// 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
}
