public void head(Node source, int depth) {
if (source instanceof Element) {
Element sourceEl = (Element) source;
if (whitelist.isSafeTag(sourceEl.tagName())) { // safe, clone and copy safe attrs
ElementMeta meta = createSafeElement(sourceEl);
Element destChild = meta.el;
destination.appendChild(destChild);
numDiscarded += meta.numAttribsDiscarded;
destination = destChild;
} else if (source
!= root) { // not a safe tag, so don't add. don't count root against discarded.
numDiscarded++;
}
} else if (source instanceof TextNode) {
TextNode sourceText = (TextNode) source;
TextNode destText = new TextNode(sourceText.getWholeText(), source.baseUri());
destination.appendChild(destText);
} else if (source instanceof DataNode && whitelist.isSafeTag(source.parent().nodeName())) {
DataNode sourceData = (DataNode) source;
DataNode destData = new DataNode(sourceData.getWholeData(), source.baseUri());
destination.appendChild(destData);
} else { // else, we don't care about comments, xml proc instructions, etc
numDiscarded++;
}
}
/**
* Get the combined data of this element. Data is e.g. the inside of a {@code script} tag.
*
* @return the data, or empty string if none
* @see #dataNodes()
*/
public String data() {
StringBuilder sb = new StringBuilder();
for (Node childNode : childNodes) {
if (childNode instanceof DataNode) {
DataNode data = (DataNode) childNode;
sb.append(data.getWholeData());
} else if (childNode instanceof Element) {
Element element = (Element) childNode;
String elementData = element.data();
sb.append(elementData);
}
}
return sb.toString();
}
public Set<MatchParserObject> parse(WhoScoredMatchParserObject matchParserObject) {
Elements scriptElements = getDocument().getElementsByTag("script");
Pattern matchIdPattern =
Pattern.compile(
".*var liveMatchUpdater = .*parameters: \\{.*id: (\\d*).*\\}.*", Pattern.DOTALL);
Pattern scriptPattern =
Pattern.compile(
"(.*)var initialMatchDataForScrappers = \\[\\[(.*), \\[(.*)", Pattern.DOTALL);
Pattern fixturePattern =
Pattern.compile(
"\\[(\\d*),(\\d*),'(.*?)','(.*?)','(.*?)','.*?',\\d*,'(.*?)',(.*)\\]", Pattern.DOTALL);
Pattern matchEventsPattern = Pattern.compile("\\[\\[(.*?)\\]\\]", Pattern.DOTALL);
Pattern goalPattern =
Pattern.compile(
".*\\['(.*?)',('(.*?)')?,'(goal|owngoal|penalty-goal)','.*?',('OG'|'Pen.')?,(\\d*),(\\d*),(\\d*)\\].*",
Pattern.DOTALL);
Pattern cardPattern =
Pattern.compile(
".*\\['(.*?)',,'(yellow|secondyellow|red)',,,(\\d*),(\\d*),(\\d*)\\].*",
Pattern.DOTALL);
Pattern substitutionPattern =
Pattern.compile(".*\\['(.*?)','(.*?)','subst',,,(\\d*),(\\d*),(\\d*)\\].*", Pattern.DOTALL);
for (Element scriptElement : scriptElements) {
for (DataNode node : scriptElement.dataNodes()) {
Matcher matchIdMatcher = matchIdPattern.matcher(node.toString());
if (matchIdMatcher.matches()) {
int whoScoredId = Integer.parseInt(matchIdMatcher.group(1));
if (matchParserObject.getWhoScoredId() == 0) {
matchParserObject.setWhoScoredId(whoScoredId);
} else if (matchParserObject.getWhoScoredId() != whoScoredId) {
logger.error(
"Provided WhoScoredMatchParserObject had whoScoredId={} but parsed document had whoScoredId={}",
matchParserObject.getWhoScoredId(),
whoScoredId);
}
}
Matcher scriptMatcher = scriptPattern.matcher(node.toString());
if (scriptMatcher.matches()) {
String[] scriptVariables = scriptMatcher.group(2).split("\n, ");
for (String scriptVariable : scriptVariables) {
scriptVariable = scriptVariable.trim();
Matcher fixtureMatcher = fixturePattern.matcher(scriptVariable);
Matcher matchEventsMatcher = matchEventsPattern.matcher(scriptVariable);
if (fixtureMatcher.matches()) {
if (matchParserObject.getHomeTeam() == null) {
matchParserObject.setHomeTeam(
new WhoScoredTeamParserObject(
fixtureMatcher.group(3), Integer.parseInt(fixtureMatcher.group(1))));
}
if (matchParserObject.getAwayTeam() == null) {
matchParserObject.setAwayTeam(
new WhoScoredTeamParserObject(
fixtureMatcher.group(4), Integer.parseInt(fixtureMatcher.group(2))));
}
if (matchParserObject.getDateTime() == null) {
matchParserObject.setDateTime(fixtureMatcher.group(5));
}
if (matchParserObject.getTimeElapsed() == null) {
matchParserObject.setTimeElapsed(fixtureMatcher.group(6));
}
} else if (matchEventsMatcher.matches()) {
for (String eventVariable : matchEventsMatcher.group(1).split("\n")) {
Matcher goalMatcher = goalPattern.matcher(eventVariable);
Matcher cardMatcher = cardPattern.matcher(eventVariable);
Matcher substitutionMatcher = substitutionPattern.matcher(eventVariable);
if (goalMatcher.matches()) {
WhoScoredGoalParserObject goalParserObject =
new WhoScoredGoalParserObject(
goalMatcher.group(1),
Integer.parseInt(goalMatcher.group(7)),
goalMatcher.group(3),
Integer.parseInt(goalMatcher.group(8)),
Integer.parseInt(goalMatcher.group(6)),
goalMatcher.group(4).equalsIgnoreCase("penalty-goal"),
goalMatcher.group(4).equalsIgnoreCase("owngoal"));
WhoScoredTeamParserObject team =
matchParserObject.getTeamForGoal(goalParserObject);
if (team == null) {
logger.error("Could not find team for goal {}.", goalParserObject);
} else {
team.getGoals().add(goalParserObject);
}
} else if (cardMatcher.matches()) {
CardType cardType =
(cardMatcher.group(2).equalsIgnoreCase("yellow")
? CardType.YELLOW
: CardType.RED);
WhoScoredCardParserObject cardParserObject =
new WhoScoredCardParserObject(
cardMatcher.group(1),
Integer.parseInt(cardMatcher.group(4)),
Integer.parseInt(cardMatcher.group(3)),
cardType);
matchParserObject
.getTeamForPlayer(cardParserObject.getPlayerWhoScoredId())
.getCards()
.add(cardParserObject);
} else if (substitutionMatcher.matches()) {
WhoScoredSubstitutionParserObject substitutionParserObject =
new WhoScoredSubstitutionParserObject(
substitutionMatcher.group(1),
Integer.parseInt(substitutionMatcher.group(4)),
substitutionMatcher.group(2),
Integer.parseInt(substitutionMatcher.group(5)),
Integer.parseInt(substitutionMatcher.group(3)));
matchParserObject
.getTeamForPlayer(substitutionParserObject.getPlayerOutWhoScoredId())
.getSubstitutions()
.add(substitutionParserObject);
}
}
}
}
}
}
}
getParserProperties().map(matchParserObject);
Set<MatchParserObject> matchParserObjects = new HashSet<MatchParserObject>();
matchParserObjects.add(matchParserObject);
return matchParserObjects;
}