/**
* Scans a given file for an exact match to the word target. We have chosen to use equals and
* contains so that sites including the target in the URL get included.
*
* @param url The link to scan
* @param target The word to search for
* @return Whether or not this file contains the target word.
*/
public boolean scan(final URL url, final String target) {
boolean foundTarget = false;
In in = new In(url);
String[] strings;
try {
strings = in.readAllStrings();
} catch (NullPointerException e) {
return false;
}
for (String word : strings) {
// If the word can be identified as a link
if (word.startsWith(LINK_IDENTIFIER + "http")) {
/*
Prepares the link for construction into a URL. First it replaces the link identifier, typically
a string like href=", however this is only valid for the beginning. We later use a split to remove the
" following the actual URL, as well as possibly a tag closure like >. The reason for implementing this
with split, is that we want to preserve the word immediately following the tag closure if applicable.
This is because this may be a word matching target, so we later replace word with the result of that
split where applicable.
*/
String[] splits = word.replace(LINK_IDENTIFIER, "").split("\"(>|)");
if (splits[0] != null) {
splits[0] = splits[0].replace("&", "&");
try {
URL newLink = new URL(splits[0]);
// Don't enqueue the link if we've already discovered it.
if (!discovered.contains(newLink) && discovered.size() < max) {
queue.enqueue(newLink);
discovered.add(newLink);
}
} catch (MalformedURLException ignored) {
// We have already turned the string in question into a valid link. It should not be
// identified
// as a link if it does not conform to the above statements. It could still be saved
// with more
// complex parsing, but we don't wish to expend resources parsing javascripts or by
// analysing
// patterns to determine what to fix.
}
}
if (splits.length > 1 && splits[1] != null) {
word = splits[1];
}
}
// Make sure we check for lower case values when matching words.
if (word.toLowerCase().equals(target) || word.toLowerCase().contains(target)) {
foundTarget = true;
}
}
return foundTarget;
}
public static void main(String[] args) {
In in = new In(args[0]);
String[] a = in.readAllStrings();
System.out.println("Quick 3 way sort:");
Stopwatch timer = new Stopwatch();
Quick3Way.sort(a);
double elapsed = timer.elapsedTime();
System.out.println("Time:" + elapsed);
show(a);
}
/**
* A test client
*
* @param args {@code args[0]} = Input-file
*/
public static void main(String args[]) {
In in = new In(args[0]);
int N = in.readInt();
UnionFindWQU qf = new UnionFindWQU(N);
while (!in.isEmpty()) {
int p = in.readInt();
int q = in.readInt();
if (qf.connected(p, q)) continue;
qf.union(p, q);
System.out.println(p + " " + q);
}
System.out.println(qf.componentCount() + " components");
}
public WordNet(String synsetFileName, String hyponymFileName) {
In synsetFile = new In(synsetFileName);
In hyponymFile = new In(hyponymFileName);
String[] currentLine;
String[] splitNouns;
while (!synsetFile.isEmpty()) {
currentLine = synsetFile.readLine().split(",");
map.put(Integer.parseInt(currentLine[0]), currentLine[1]);
splitNouns = currentLine[1].split(" ");
for (int i = 0; i < splitNouns.length; i += 1) {
if (!nouns.contains(splitNouns[i])) {
nouns.add(splitNouns[i]);
}
}
}
synsets = new Digraph(map.size());
while (!hyponymFile.isEmpty()) {
currentLine = hyponymFile.readLine().split(",");
for (int i = 1; i < currentLine.length; i++) {
synsets.addEdge(Integer.parseInt(currentLine[0]), Integer.parseInt(currentLine[i]));
}
}
}
