Examples with Element org.jsoup.nodes.Element used on opensource projects

Example 96 with Element

use of org.jsoup.nodes.Element in project Java-Tutorial by gpcodervn.

the class JSoupExample method getAllPostImages.

// Lấy tất cả ảnh đại diện của bài viết
public static void getAllPostImages() throws IOException {
    Document doc = Jsoup.connect("https://gpcoder.com").get();
    Elements images = doc.select("article img[src~=(?i)\\.(png|jpe?g|gif)]");
    for (Element image : images) {
        System.out.println("\nsrc : " + image.attr("src"));
        System.out.println("height : " + image.attr("height"));
        System.out.println("width : " + image.attr("width"));
        System.out.println("alt : " + image.attr("alt"));
    }
}

Example 97 with Element

use of org.jsoup.nodes.Element in project lavaplayer by sedmelluq.

the class YoutubeAudioSourceManager method extractPlaylistTracks.

private String extractPlaylistTracks(Element videoContainer, Element loadMoreContainer, List<AudioTrack> tracks) {
    for (Element video : videoContainer.select(".pl-video")) {
        Elements lengthElements = video.select(".timestamp span");
        // If the timestamp element does not exist, it means the video is private
        if (!lengthElements.isEmpty()) {
            String videoId = video.attr("data-video-id").trim();
            String title = video.attr("data-title").trim();
            String author = video.select(".pl-video-owner a").text().trim();
            long duration = DataFormatTools.durationTextToMillis(lengthElements.first().text());
            tracks.add(buildTrackObject(videoId, title, author, false, duration));
        }
    }
    if (loadMoreContainer != null) {
        Elements more = loadMoreContainer.select(".load-more-button");
        if (!more.isEmpty()) {
            return more.first().attr("data-uix-load-more-href");
        }
    }
    return null;
}

Example 98 with Element

use of org.jsoup.nodes.Element in project lavaplayer by sedmelluq.

the class YoutubeSearchProvider method extractTrackFromResultEntry.

private void extractTrackFromResultEntry(List<AudioTrack> tracks, Element element) {
    Element durationElement = element.select("[class^=video-time]").first();
    Element contentElement = element.select(".yt-lockup-content").first();
    String videoId = element.attr("data-context-item-id");
    if (durationElement == null || contentElement == null || videoId.isEmpty()) {
        return;
    }
    long duration = DataFormatTools.durationTextToMillis(durationElement.text());
    String title = contentElement.select(".yt-lockup-title > a").text();
    String author = contentElement.select(".yt-lockup-byline > a").text();
    tracks.add(sourceManager.buildTrackObject(videoId, title, author, false, duration));
}

Example 99 with Element

use of org.jsoup.nodes.Element in project epadd by ePADD.

the class Highlighter method getHTMLAnnotatedDocumentContents.

/**
 * @param contents is the content to be annotated, typically the text in email body
 * A convenience method to do the bulk job of annotating all the terms in termsToHighlight, termsToHyperlink and entitiesWithId
 * Also hyperlinks any URLs found in the content
 * @param regexToHighlight - the output will highlight all the strings matching regexToHighlight
 * @param showDebugInfo - when set will append to the output some debug info. related to the entities present in the content and passed through entitiesWithId
 *
 * Note: DO not modify any of the objects passed in the parameter
 *       if need to be modified then clone and modify a local copy
 */
// TODO: can also get rid of termsToHyperlink
public static String getHTMLAnnotatedDocumentContents(String contents, Date d, String docId, String regexToHighlight, Set<String> termsToHighlight, Map<String, EmailRenderer.Entity> entitiesWithId, Set<String> termsToHyperlink, boolean showDebugInfo) {
    Set<String> highlightTerms = new LinkedHashSet<>(), hyperlinkTerms = new LinkedHashSet<>();
    if (termsToHighlight != null)
        highlightTerms.addAll(termsToHighlight);
    if (termsToHyperlink != null)
        hyperlinkTerms.addAll(termsToHyperlink);
    if (log.isDebugEnabled())
        log.debug("DocId: " + docId + "; Highlight terms: " + highlightTerms + "; Entities: " + entitiesWithId + "; Hyperlink terms: " + hyperlinkTerms);
    // System.err.println("DocId: " + docId + "; Highlight terms: " + highlightTerms + "; Entities: " + entitiesWithId + "; Hyperlink terms: " + hyperlinkTerms);
    short HIGHLIGHT = 0, HYPERLINK = 1;
    // pp for post process, as we cannot add complex tags which highlighting
    String preHighlightTag = "<span class='hilitedTerm rounded' >", postHighlightTag = "</span>";
    String preHyperlinkTag = "<span data-process='pp'>", postHyperlinkTag = "</span>";
    // since the urls are not tokenized as one token, it is not possible to highlight with Lucene highlighter.
    Pattern p = Pattern.compile("https?://[^\\s\\n]*");
    Matcher m = p.matcher(contents);
    StringBuffer sb = new StringBuffer();
    while (m.find()) {
        String link = m.group();
        String url = link;
        if (d != null) {
            Calendar c = new GregorianCalendar();
            c.setTime(d);
            String archiveDate = c.get(Calendar.YEAR) + String.format("%02d", c.get(Calendar.MONTH)) + String.format("%02d", c.get(Calendar.DATE)) + "120000";
            url = "http://web.archive.org/web/" + archiveDate + "/" + link;
        }
        m.appendReplacement(sb, Matcher.quoteReplacement("<a target=\"_blank\" href=\"" + url + "\">" + link + "</a> "));
    }
    m.appendTail(sb);
    contents = sb.toString();
    if (!Util.nullOrEmpty(regexToHighlight)) {
        contents = annotateRegex(contents, regexToHighlight, preHighlightTag, postHighlightTag);
    }
    List<String> catchTerms = Arrays.asList("class", "span", "data", "ignore");
    Set<String> ignoreTermsForHyperlinking = catchTerms.stream().map(String::toLowerCase).collect(Collectors.toSet());
    // entitiesid stuff is already canonicalized with tokenize used with analyzer
    if (entitiesWithId != null)
        hyperlinkTerms.addAll(entitiesWithId.keySet().stream().filter(term -> !ignoreTermsForHyperlinking.contains(term.trim().toLowerCase())).map(term -> "\"" + term + "\"").collect(Collectors.toSet()));
    // If there are overlapping annotations, then they need to be serialised.
    // This is serialized order for such annotations.
    // map strings to be annotated -> boolean denoting whether to highlight or hyperlink.
    List<Pair<String, Short>> order = new ArrayList<>();
    // should preserve order so that highlight terms are seen before hyperlink
    Set<String> allTerms = new LinkedHashSet<>();
    allTerms.addAll(highlightTerms);
    /*
		 * We ant to assign order in which terms are highlighted or hyperlinked.
		 * for example: if we want to annotate both "Robert" and "Robert Creeley", and if we annotate "Robert" first then we may miss on "Robert Creeley"
		 * so we assign order over strings that share any common words as done in the loop below
		 * TODO:
		 * This test can still miss cases when a regular expression that eventually matches a word already annotated and
		 * when two terms like "Robert Creeley" "Mr Robert" to match a text like: "Mr Robert Creeley".
		 * TODO: Give pref. to highlighter over hyperlink
		 * TODO: remove order and simplify
		 * In such cases one of the terms may not be annotated.
		 * Terms that are added to o are those that just share at-least one word
		 */
    // should preserve order so that highlight terms that are added first stay that way
    Map<Pair<String, Short>, Integer> o = new LinkedHashMap<>();
    // prioritised terms
    // Note that a term can be marked both for highlight and hyperlink
    Set<String> consTermsHighlight = new HashSet<>(), consTermsHyperlink = new HashSet<>();
    for (String at : allTerms) {
        // Catch: if we are trying to highlight terms like class, span e.t.c,
        // we better annotate them first as it may go into span tags and annotate the stuff, causing the highlighter to break
        Set<String> substrs = IndexUtils.computeAllSubstrings(at);
        for (String substr : substrs) {
            if (at.equals(substr) || at.equals("\"" + substr + "\""))
                continue;
            boolean match = catchTerms.contains(substr.toLowerCase());
            int val = match ? Integer.MAX_VALUE : substr.length();
            // The highlight or hyperlink terms may have quotes, specially handling below is for that.. is there a better way?
            if (highlightTerms.contains(substr) || highlightTerms.contains("\"" + substr + "\"")) {
                highlightTerms.remove(substr);
                highlightTerms.remove("\"" + substr + "\"");
                // there should be no repetitions in the order array, else it leads to multiple annotations i.e. two spans around one single element
                if (!consTermsHighlight.contains(substr)) {
                    o.put(new Pair<>(substr, HIGHLIGHT), val);
                    consTermsHighlight.add(substr);
                }
            }
            if (hyperlinkTerms.contains(substr) || hyperlinkTerms.contains("\"" + substr + "\"")) {
                hyperlinkTerms.remove(substr);
                hyperlinkTerms.remove("\"" + substr + "\"");
                if (!consTermsHyperlink.contains(substr)) {
                    o.put(new Pair<>(substr, HYPERLINK), val);
                    consTermsHyperlink.add(substr);
                }
            }
        }
    }
    // now sort the phrases from longest length to smallest length
    List<Pair<Pair<String, Short>, Integer>> os = Util.sortMapByValue(o);
    order.addAll(os.stream().map(pair -> pair.first).collect(Collectors.toSet()));
    // System.err.println(order+" hit: "+highlightTerms+" -- hyt: "+hyperlinkTerms);
    // annotate whatever is left in highlight and hyperlink Terms.
    // String result = contents;
    String result = highlightBatch(contents, highlightTerms.toArray(new String[highlightTerms.size()]), preHighlightTag, postHighlightTag);
    result = highlightBatch(result, hyperlinkTerms.toArray(new String[hyperlinkTerms.size()]), preHyperlinkTag, postHyperlinkTag);
    // now highlight terms in order.
    for (Pair<String, Short> ann : order) {
        short type = ann.second;
        String term = ann.first;
        String preTag = null, postTag = null;
        if (type == HYPERLINK) {
            preTag = preHyperlinkTag;
            postTag = postHyperlinkTag;
        } else if (type == HIGHLIGHT) {
            preTag = preHighlightTag;
            postTag = postHighlightTag;
        }
        try {
            result = highlight(result, term, preTag, postTag);
        } catch (IOException | InvalidTokenOffsetsException | ParseException e) {
            Util.print_exception("Exception while adding html annotation: " + ann.first, e, log);
            e.printStackTrace();
        }
    }
    // do some line breaking and show overflow.
    String[] lines = result.split("\\n");
    StringBuilder htmlResult = new StringBuilder();
    boolean overflow = false;
    for (String line : lines) {
        htmlResult.append(line);
        htmlResult.append("\n<br/>");
    }
    if (overflow) {
        htmlResult.append("</div>\n");
        // the nojog class ensures that the jog doesn't pop up when the more
        // button is clicked
        htmlResult.append("<span class=\"nojog\" style=\"color:#500050;text-decoration:underline;font-size:12px\" onclick=\"muse.reveal(this, false);\">More</span><br/>\n");
    }
    // Now do post-processing to add complex tags that depend on the text inside. title, link and cssclass
    org.jsoup.nodes.Document doc = Jsoup.parse(htmlResult.toString());
    Elements elts = doc.select("[data-process]");
    for (int j = 0; j < elts.size(); j++) {
        Element elt = elts.get(j);
        Element par = elt.parent();
        // Do not touch nested entities
        if (par != null && par.attr("data-process") == null)
            // (preHighlightTag.contains(par.tagName())||preHyperlinkTag.contains(par.tagName())))
            continue;
        String entity = elt.text();
        int span_j = j;
        String link = "browse?adv-search=1&termBody=on&termSubject=on&termAttachments=on&termOriginalBody=on&term=\"" + Util.escapeHTML(entity) + "\"";
        // note &quot here because the quotes have to survive
        // through the html page and reflect back in the URL
        // may need to URI escape docId?
        link += "&initDocId=" + docId;
        String title = "";
        try {
            String cssclass = "";
            EmailRenderer.Entity info = entitiesWithId.get(entity);
            if (info != null) {
                if (info.ids != null) {
                    title += "<div id=\"fast_" + info.ids + "\"></div>";
                    title += "<script>getFastData(\"" + info.ids + "\");</script>";
                    cssclass = "resolved";
                } else {
                    // the last three are the OpenNLPs'
                    // could have defined overlapping sub-classes, which would have reduced code repetitions in css file; but this way more flexibility
                    String[] types = new String[] { "cp", "cl", "co", "person", "org", "place", "acr" };
                    String[] cssclasses = new String[] { "custom-people", "custom-loc", "custom-org", "opennlp-person", "opennlp-org", "opennlp-place", "acronym" };
                    outer: for (String et : info.types) {
                        for (int t = 0; t < types.length; t++) {
                            String type = types[t];
                            if (type.equals(et)) {
                                if (t < 3) {
                                    cssclass += cssclasses[t] + " ";
                                    // consider no other class
                                    continue outer;
                                } else {
                                    cssclass += cssclasses[t] + " ";
                                }
                            }
                        }
                    }
                }
            } else {
                cssclass += " unresolved";
            }
            // enables completion (expansion) of words while browsing of messages.
            if (entity != null) {
                // enable for only few types
                if (cssclass.contains("custom-people") || cssclass.contains("acronym") || cssclass.contains("custom-org") || cssclass.contains("custom-loc")) {
                    // TODO: remove regexs
                    entity = entity.replaceAll("(^\\s+|\\s+$)", "");
                    if (!entity.contains(" ")) {
                        // String rnd = rand.nextInt() + "";
                        // <img src="images/spinner.gif" style="height:15px"/>
                        // <script>expand("" + entity + "\",\"" + StringEscapeUtils.escapeJava(docId) + "\",\"" + rnd + "");</script>
                        // if(info.expandsTo!=null)
                        // title += "<div class=\"resolutions\" id=\"expand_" + rnd + "\"><a href='browse?term=\""+info.expandsTo+"\"'>"+info.expandsTo+"</a></div>";
                        cssclass += " expand";
                    }
                }
            }
            for (int k = j; k <= span_j; k++) {
                elt = elts.get(k);
                // don't annotate nested tags-- double check if the parent tag is highlight related tag or entity related annotation
                if (elt.parent().tag().getName().toLowerCase().equals("span") && elt.parent().classNames().toString().contains("custom")) {
                    continue;
                }
                String cc = elt.attr("class");
                elt.attr("class", cc + " " + cssclass);
                elt.attr("title", title);
                elt.attr("onclick", "window.location='" + link + "'");
                // A tag may have nested tags in it and is involved to get the text in it.
                elt.attr("data-text", entity);
                elt.attr("data-docId", StringEscapeUtils.escapeHtml(docId));
            }
        } catch (Exception e) {
            Util.print_exception("Some unknown error while highlighting", e, log);
        }
    }
    // The output Jsoup .html() will dump each tag in separate line
    String html = doc.html();
    if (showDebugInfo) {
        String debug_html = html + "<br>";
        debug_html += "<div class='debug' style='display:none'>";
        debug_html += "docId: " + docId;
        debug_html += "<br>-------------------------------------------------<br>";
        for (String str : entitiesWithId.keySet()) debug_html += str + ":" + entitiesWithId.get(str).types + ";;; ";
        debug_html += "<br>-------------------------------------------------<br>";
        String[] opennlp = new String[] { "person", "place", "org" };
        String[] custom = new String[] { "cp", "cl", "co" };
        for (int j = 0; j < opennlp.length; j++) {
            String t1 = opennlp[j];
            String t2 = custom[j];
            Set<String> e1 = new HashSet<>();
            Set<String> e2 = new HashSet<>();
            for (String str : entitiesWithId.keySet()) {
                Set<String> types = entitiesWithId.get(str).types;
                if (types.contains(t1) && !types.contains(t2))
                    e1.add(entitiesWithId.get(str).name);
                else if (types.contains(t2) && !types.contains(t1))
                    e2.add(entitiesWithId.get(str).name);
            }
            debug_html += opennlp[j] + " entities recognised by only opennlp: " + e1;
            debug_html += "<br>";
            debug_html += opennlp[j] + " entities recognised by only custom: " + e2;
            debug_html += "<br><br>";
        }
        debug_html += "-------------------------------------------------<br>";
        lines = contents.split("\\n");
        for (String line : lines) debug_html += line + "<br>";
        debug_html += "</div>";
        debug_html += "<button onclick='$(\".debug\").style(\"display\",\"block\");'>Show Debug Info</button>";
        return debug_html;
    }
    return html;
}

Example 100 with Element

use of org.jsoup.nodes.Element in project ripme by RipMeApp.

the class VkRipper method ripImages.

private void ripImages() throws IOException {
    Map<String, String> photoIDsToURLs = new HashMap<>();
    int offset = 0;
    while (true) {
        logger.info("    Retrieving " + this.url);
        // al=1&offset=80&part=1
        Map<String, String> postData = new HashMap<>();
        postData.put("al", "1");
        postData.put("offset", Integer.toString(offset));
        postData.put("part", "1");
        Document doc = Http.url(this.url).referrer(this.url).ignoreContentType().data(postData).post();
        String body = doc.toString();
        if (!body.contains("<div")) {
            break;
        }
        body = body.substring(body.indexOf("<div"));
        doc = Jsoup.parseBodyFragment(body);
        List<Element> elements = doc.select("a");
        Set<String> photoIDsToGet = new HashSet<>();
        for (Element a : elements) {
            if (!a.attr("onclick").contains("showPhoto('")) {
                logger.error("a: " + a);
                continue;
            }
            String photoID = a.attr("onclick");
            photoID = photoID.substring(photoID.indexOf("showPhoto('") + "showPhoto('".length());
            photoID = photoID.substring(0, photoID.indexOf("'"));
            if (!photoIDsToGet.contains(photoID)) {
                photoIDsToGet.add(photoID);
            }
        }
        for (String photoID : photoIDsToGet) {
            if (!photoIDsToURLs.containsKey(photoID)) {
                try {
                    photoIDsToURLs.putAll(getPhotoIDsToURLs(photoID));
                } catch (IOException e) {
                    logger.error("Exception while retrieving photo id " + photoID, e);
                    continue;
                }
            }
            if (!photoIDsToURLs.containsKey(photoID)) {
                logger.error("Could not find URL for photo ID: " + photoID);
                continue;
            }
            String url = photoIDsToURLs.get(photoID);
            addURLToDownload(new URL(url));
            if (isStopped() || isThisATest()) {
                break;
            }
        }
        if (elements.size() < 40 || isStopped() || isThisATest()) {
            break;
        }
        offset += elements.size();
    }
    waitForThreads();
}