Java FileIO.readAsString方法代码示例

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public static void ReadFolder(String dirPath) throws Exception
{
	dirPath = FileIO.makeDirPath(dirPath);
	File dir = new File(dirPath);
	if (!dir.isDirectory()) throw new Exception(dirPath + " is not a directory");

	File[] files = dir.listFiles();
	for (File file : files)
	{
		String results = file.getName() + "\r\n";
		results += FileIO.readAsString(file.getPath(), new String[] { "MAE", "MAUE" });
		FileIO.writeString(dirPath + "Results.txt", results, true);
	}
}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public static void collectResults() throws Exception
{
	String mode = params.BATCH_RUN ? DatasetMode.batch.label : params.DATASET_MODE.label;
	String program = methodId + " [" + mode + "]";

	/* Collect result files to specific directory */
	Path source = FileSystems.getDefault().getPath("results.txt");
	Path target = FileSystems.getDefault().getPath(
			FileIO.makeDirectory(AbstractCF.params.RESULTS_DIRECTORY + Dataset.LABEL) + program + "@"
					+ Dates.now() + ".txt");
	Files.copy(source, target);
	if (params.BATCH_RUN && params.numRunMethod > 1 && numRunMethod < params.numRunMethod) FileIO.empty(source
			.toString());

	/* Send email to notify results */
	if (params.EMAIL_NOTIFICATION)
	{
		String text = FileIO.readAsString(source.toString());
		Gmailer notifier = new Gmailer();

		//notifier.getProps().setProperty("mail.to", "[email protected]");
		//notifier.getProps().setProperty("mail.to", "[email protected]");
		notifier.getProps().setProperty("mail.to", "[email protected]");
		//notifier.getProps().setProperty("mail.bcc", "[email protected]");

		notifier.getProps()
				.setProperty("mail.subject", Dataset.LABEL + ": " + program + " From " + Systems.getIP());
		notifier.send(text, target.toString());
	}
}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public void run_dvd_ratings(String url) throws Exception {
	String[] data = url.split(": ");
	String category = data[0];
	String category_url = data[1];
	category_url = category_url.substring(0, category_url.lastIndexOf('_'));
	String categoryID = category_url.substring(category_url
			.lastIndexOf('_') + 1);

	String catPath = FileIO.makeDirPath(desktop, domain, category);
	File Dir = new File(catPath);
	File[] prodDirs = Dir.listFiles();
	int tk = prodDirs.length;
	for (int k = 0; k < tk; k++) {
		File prodDir = prodDirs[k];
		// for each product
		String productID = prodDir.getName();
		if (productID.equals("webPages"))
			continue;
		if (!prodDir.isDirectory())
			continue;

		String prodPath = FileIO.makeDirPath(catPath, productID);
		String reviewPath = FileIO
				.makeDirPath(prodPath, "Detailed_Reviews");
		String dvdPath = prodPath + "dvd-ratings.txt";
		FileIO.deleteFile(dvdPath);

		File reviewDir = new File(reviewPath);
		File[] reviewDirs = reviewDir.listFiles();
		List<String> reviews = new ArrayList<>();
		for (int i = 0; i < reviewDirs.length; i++) {
			// for each review
			File reviewFile = reviewDirs[i];
			String reviewID = reviewFile.getName();
			Logs.debug("{}: {} ({}/{}): {} ({}/{})", category, productID,
					(k + 1), tk, reviewID, (i + 1), reviewDirs.length);

			String dirPath = FileIO.makeDirPath(reviewPath, reviewID);
			String ratingPath = dirPath + reviewID + ".html";
			if (!FileIO.exist(ratingPath))
				continue; // review page is not existing
			String html = FileIO.readAsString(ratingPath);
			Document doc = Jsoup.parse(html);
			Element div = doc.select("div#OH_BingUserInfo").first();
			if (div == null)
				continue; // no user review exists

			// user-info
			Element a = div.select("p.m-reer-usertab.clearfix a.black")
					.first();
			String raw = a.attr("onmousedown");
			raw = raw.substring(raw.indexOf("(") + 1, raw.lastIndexOf(")"));
			String userUrl = raw.replace(",", "").replace("'", "");
			String userID = userUrl.substring(userUrl.lastIndexOf('_') + 1);

			// user-rating value
			Element r = div.select(
					"p.m-reer-usertab.clearfix img.ratingStars").first();
			String rating = r.attr("alt");

			// user-rating date
			div = doc.select("div#OH_BingUserOpinion").first();
			Element date = div
					.select("div.m-reer-opheader.reviewTitle span.m-reer-ddwrap span[property]")
					.first();
			String datetime = date.attr("content");

			// content
			String review = userID + "," + productID + "," + categoryID
					+ "," + reviewID + "," + rating + "," + datetime + ","
					+ userUrl;
			reviews.add(review);
		}
		FileIO.writeList(dvdPath, reviews);
	}
}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public void run_products(String url) throws Exception {
	String[] data = url.split(": ");
	String category = data[0];
	// String link = data[1];

	String dirPath = FileIO.makeDirPath(desktop, domain, category);
	List<String> links = FileIO.readAsList(dirPath + "movies.txt");
	int tk = links.size();
	for (int k = 0; k < tk; k++) {
		String link = links.get(k);
		String[] d = link.split("::");
		String id = d[0];
		String name = d[1];
		String productLink = d[2];
		int idx = productLink.lastIndexOf("/");
		String p1 = productLink.substring(0, idx) + "/Reviews";
		String reviewLink = p1 + productLink.substring(idx);

		// create folder
		String path = FileIO.makeDirectory(dirPath, id);

		// product page
		String html = null;

		String pagePath = path + id + ".html";
		if (!FileIO.exist(pagePath)) {
			html = read_url(productLink);
			FileIO.deleteFile(path + name + ".html");
			FileIO.writeString(pagePath, html);
		}

		// product reviews
		// get first page anyway to identify the maximum pages
		path = FileIO.makeDirectory(path, "Reviews");
		String reviewPath = path + "page_1.html";
		if (FileIO.exist(reviewPath)) {
			html = FileIO.readAsString(reviewPath);
		} else {
			html = read_url(reviewLink);
			FileIO.writeString(reviewPath, html);
		}
		Logs.debug(category + ": " + id + " (" + (k + 1) + "/" + tk + ")"
				+ ": page " + 1);

		Document doc = Jsoup.parse(html);
		Elements nav = doc.select("div#Pagination");

		if (!nav.isEmpty()) {
			int maxPage = 1;

			Elements last = nav.select("li.last");
			if (!last.isEmpty())
				maxPage = Integer.parseInt(last.first().text()); // more
																	// than
																	// 11
																	// pages
			else
				maxPage = Integer.parseInt(nav.select("li").last().text()); // less
																			// or
																			// equal
																			// 11
																			// pages

			for (int i = 2; i <= maxPage; i++) {
				String filePath = path + "page_" + i + ".html";
				if (FileIO.exist(filePath))
					continue;

				reviewLink = reviewLink + "/Start/" + ((i - 1) * 15);
				html = read_url(reviewLink);
				FileIO.writeString(filePath, html);

				Logs.debug(category + ": " + id + " (" + (k + 1) + "/" + tk
						+ ")" + ": page " + i + "/" + maxPage);
			}
		}
	}

}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public static void parseCategoryPages() throws Exception {
	String filePath = FileIO.getResource("dvd.ciao.txt");
	List<String> urls = FileIO.readAsList(filePath);

	String dir = Systems.getDesktop() + "dvd.ciao.co.uk\\";
	for (String url : urls) {
		// each category
		String[] data = url.split(": ");
		String category = data[0];
		String dirCate = FileIO.makeDirPath(dir, category);
		String dirPath = FileIO.makeDirPath(dir, category, "webPages");

		// clear
		FileIO.deleteFile(dirCate + "movies.txt");

		File dirs = new File(dirPath);
		for (File f : dirs.listFiles()) {
			// each web page
			String html = FileIO.readAsString(dirPath + f.getName());
			Document doc = Jsoup.parse(html);

			Logs.debug(category + ": " + f.getName());

			List<String> movies = new ArrayList<>();
			Elements products = doc.select("td.prodInfo");
			for (Element product : products) {
				// each product
				Element prod = product.select("p.prodName").first();
				String name = prod.text();
				String link = prod.select("a").first().attr("href");
				String id = link.substring(link.lastIndexOf("_") + 1);

				// number of user reviews
				prod = product.select("p.prodRating").first();
				String cnt = prod.select(".userReviewsCount").text()
						.replace("(", "").replace(")", "");
				int count = 0;
				if (!cnt.isEmpty())
					count = Integer.parseInt(cnt);

				// do not consider movies without any reviews
				if (count > 0) {
					String movie = id + "::" + name + "::" + link;
					movies.add(movie);
				}
			}

			FileIO.writeList(dirCate + "movies.txt", movies, null, true);
		}
	}

}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public static void getAllTrust() throws Exception {

		Map<String, String> users = new HashMap<>();
		List<String> userLines = FileIO.readAsList(FileIO
				.getResource("users.txt"));
		for (String line : userLines) {
			String[] data = line.split(",");
			users.put(data[1], data[0]);
		}

		String usersPath = FileIO.makeDirPath(desktop, domain,
				"users.ciao.co.uk");
		FileIO.deleteFile(usersPath + "trust.txt");

		File dir = new File(usersPath);
		File[] files = dir.listFiles();

		for (int i = 0, n = files.length; i < n; i++) {
			// for each user
			File userFile = files[i];
			final String userID = userFile.getName();

			String html = null;
			Document doc = null;
			List<String> friends = new ArrayList<>();
			File[] pages = userFile.listFiles();
			for (int j = 0, m = pages.length; j < m; j++) {
				// for each trust page
				File file = pages[j];
				String name = file.getName();
				if (name.startsWith("friends")) {
					html = FileIO.readAsString(file.getPath());
					doc = Jsoup.parse(html);
					Element trustTable = doc.select("form table.trust").first();
					Elements trs = trustTable.select("tbody tr");
					for (Element tr : trs) {
						Element td = tr.select("td").get(1);
						Element a = td.select("a").first();
						String link = a.attr("href");
						if (users.containsKey(link)) {
							friends.add(users.get(link));
						}
					}
				}
			}

			FileIO.writeList(usersPath + "trust.txt", friends,
					new Converter<String, String>() {

						@Override
						public String transform(String friend) {
							return userID + "," + friend + ",1";
						}
					}, true);
		}

	}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public static void parseReviewPages() throws Exception {
	String filePath = FileIO.getResource("dvd.ciao.txt");
	List<String> urls = FileIO.readAsList(filePath);

	String dir = Systems.getDesktop() + "dvd.ciao.co.uk\\";
	for (String url : urls) {
		// each category
		String[] data = url.split(": ");
		String category = data[0];
		String dirCate = FileIO.makeDirPath(dir, category);

		File dirs = new File(dirCate);
		for (File f : dirs.listFiles()) {
			// each product folder
			if (f.getName().equals("webPages"))
				continue;
			if (!f.isDirectory())
				continue;

			String prodPath = FileIO.makeDirPath(dirCate, f.getName());
			String revwPath = FileIO.makeDirPath(prodPath, "Reviews");

			List<String> reviews = new ArrayList<>();
			File reviewDirs = new File(revwPath);

			for (File rf : reviewDirs.listFiles()) {
				// each review page
				String html = FileIO.readAsString(rf.getPath());
				Document doc = Jsoup.parse(html);

				Elements es = doc.select("div.m-shortReviewSnippet");
				for (Element e : es) {
					Element a = e.select(
							"p.m-shet-review-title a.ReviewTitle").first();
					if (a == null)
						continue; // some reviews do not have specific link
									// to detailed contents

					// url
					String link = a.attr("href");

					int idx = link.lastIndexOf("_");
					String id = link.substring(idx + 1);

					reviews.add(id + "::" + link);
				}
			}

			FileIO.writeList(prodPath + "reviews.txt", reviews, null, false);
		}

	}

}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public void crawl_comments(String url) throws Exception
{
	String html = read_url(url);
	Document doc = Jsoup.parse(html);
	String name = doc.select("div.detail_head_name h1").first().text();
	name = Strings.filterWebString(name, '_');

	String val = doc.select("#detail_nav li a").first().attr("href");
	String id = val.substring(val.lastIndexOf("/") + 1);

	String dirPath = dir + name + "/comments/";
	FileIO.makeDirectory(dirPath);

	// save rating pages
	int max = 1;
	boolean maxSet = false;
	url = url + "/commentlist";

	for (int k = 0; k <= max; k++)
	{
		String page_file = dirPath + "page_" + (k + 1) + ".html";
		Logs.debug(name + " comments with page: " + (k + 1) + "/" + (max + 1));

		String contents = null;
		if (!FileIO.exist(page_file))
		{

			String link = "http://www.gewara.com/ajax/common/qryComment.xhtml?pageNumber="
					+ k
					+ "&relatedid="
					+ id
					+ "&title=&issue=false&hasMarks=true&tag=movie&isPic=true&isVideo=false&pages=true&maxCount=20&userLogo=";

			contents = read_url(link);
			FileIO.writeString(page_file, contents);// new String(contents.getBytes("utf-8"), "utf-8"));
		} else
		{
			contents = FileIO.readAsString(page_file);
		}

		// find the maximum page num;
		if (!maxSet)
		{
			Document doc2 = Jsoup.parse(contents);
			Elements es = doc2.select("div#page a");
			Element e = es.get(es.size() - 2);
			max = Integer.parseInt(e.attr("lang"));
			maxSet = true;
		}

	}
}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public void run_reviews(String url) throws Exception
{
	url = url.trim();
	String html = read_url(url);
	Document doc = Jsoup.parse(html);
	String name = doc.select("span[property=v:itemreviewed]").text();
	name = Strings.filterWebString(name, '_');

	String dirPath = dir + name + "/reviews/";
	FileIO.makeDirectory(dirPath);

	// save rating pages
	int k = 0;
	url = url + "reviews";
	String link = url;
	while (true)
	{
		k++;
		String page = null;
		String path = dirPath + "page_" + k + ".html";
		if (!FileIO.exist(path))
		{
			page = read_url(link);
			FileIO.writeString(path, page);
			Logs.debug(name + " reviews with page: " + k);
		} else
		{
			page = FileIO.readAsString(path);
		}

		// find the next page link;
		Document doc2 = Jsoup.parse(page);
		Elements es = doc2.select("div#paginator a.next");
		if (es == null || es.size() == 0)
		{
			break;
		} else
		{
			link = url + es.first().attr("href");
		}
	}
}
 

import happy.coding.io.FileIO; //导入方法依赖的package包/类
public void run_comments(String url) throws Exception
{
	url = url.trim();
	String html = read_url(url);
	Document doc = Jsoup.parse(html);
	String name = doc.select("span[property=v:itemreviewed]").text();
	name = Strings.filterWebString(name, '_');

	String dirPath = dir + name + "/comments/";
	FileIO.makeDirectory(dirPath);

	// save rating pages
	int k = 0;
	url = url + "comments";
	String link = url;
	while (true)
	{
		k++;
		String page_file = dirPath + "page_" + k + ".html";

		String contents = null;
		if (!FileIO.exist(page_file))
		{
			contents = read_url(link);
			FileIO.writeString(page_file, contents);
			Logs.debug(name + " comments with page: " + k);
		} else
		{
			contents = FileIO.readAsString(page_file);
		}

		// find the next page link;
		Document doc2 = Jsoup.parse(contents);
		Elements es = doc2.select("div#paginator a.next");
		if (es == null || es.size() == 0)
		{
			break;
		} else
		{
			link = url + es.first().attr("href");
		}
	}
}