--- android/TrainInfoService/src/dk/thoerup/traininfoservice/banedk/DepartureFetcher.java 2010/06/10 22:26:09 829 +++ android/TrainInfoService/src/dk/thoerup/traininfoservice/banedk/DepartureFetcher.java 2011/04/19 15:07:23 1305 @@ -1,119 +1,180 @@ package dk.thoerup.traininfoservice.banedk; -import java.util.ArrayList; + +import java.net.URL; +import java.net.URLEncoder; import java.util.Collections; -import java.util.List; import java.util.Map; import java.util.logging.Logger; -import com.gargoylesoftware.htmlunit.BrowserVersion; -import com.gargoylesoftware.htmlunit.WebClient; -import com.gargoylesoftware.htmlunit.html.DomNodeList; -import com.gargoylesoftware.htmlunit.html.HtmlElement; -import com.gargoylesoftware.htmlunit.html.HtmlPage; - +import org.jsoup.nodes.Document; +import org.jsoup.nodes.Element; +import org.jsoup.select.Elements; + +import dk.thoerup.android.traininfo.common.DepartureBean; +import dk.thoerup.android.traininfo.common.DepartureEntry; +import dk.thoerup.android.traininfo.common.StationBean.StationEntry; import dk.thoerup.circuitbreaker.CircuitBreaker; import dk.thoerup.circuitbreaker.CircuitBreakerManager; -import dk.thoerup.traininfoservice.StationBean; -import dk.thoerup.traininfoservice.StationDAO; import dk.thoerup.traininfoservice.Statistics; +import dk.thoerup.traininfoservice.TraininfoSettings; +import dk.thoerup.traininfoservice.db.StationDAO; public class DepartureFetcher { + enum TrainType{ + STOG, + REGIONAL + } + + enum FetchTrainType { + STOG, + REGIONAL, + BOTH + } + Logger logger = Logger.getLogger(DepartureFetcher.class.getName()); - Map> cache; + Map cache; StationDAO stationDao = new StationDAO(); - private boolean useTempSite; + + private TraininfoSettings settings; - public DepartureFetcher(boolean tempSite, int cacheTimeout) { - useTempSite = tempSite; - cache = new TimeoutMap>(cacheTimeout); + public DepartureFetcher(TraininfoSettings settings) { + this.settings = settings; + cache = new TimeoutMap( settings.getCacheTimeout() ); } - public List cachedLookupDepartures(int stationID, boolean arrival) throws Exception { - final String key = "" + stationID + ":" + arrival; + public DepartureBean cachedLookupDepartures(int stationID, boolean arrival, FetchTrainType type) throws Exception { - List list = cache.get(key); + final String key = "" + stationID + ":" + arrival + ":" + type.toString(); + + DepartureBean departureBean = cache.get(key); - if (list == null) { - list = lookupDepartures(stationID,arrival); - cache.put(key, list); + if (departureBean == null) { + departureBean = lookupDepartures(stationID, arrival, type); + cache.put(key, departureBean); } else { Statistics.getInstance().incrementDepartureCacheHits(); logger.info("Departure: Cache hit " + key); //remove before production } - return list; + return departureBean; } - public List lookupDepartures(int stationID, boolean arrival) throws Exception { - List departureList = new ArrayList(); + public DepartureBean lookupDepartures(int stationID, boolean arrival, FetchTrainType type) throws Exception { + + DepartureBean departureBean = new DepartureBean(); - StationBean station = stationDao.getById(stationID); + StationEntry station = stationDao.getById(stationID); - if (station.getRegional() != null) { - List list = lookupDepartures(station.getRegional(), "FJRN", arrival); - departureList.addAll(list); + departureBean.stationName = station.getName(); + + if (station.getRegional() != null && (type == FetchTrainType.REGIONAL||type == FetchTrainType.BOTH) ) { + DepartureBean tempBean = lookupDepartures(station.getRegional(), TrainType.REGIONAL, arrival); + departureBean.entries.addAll( tempBean.entries ); + departureBean.notifications.addAll(tempBean.notifications); } - if (station.getStrain() != null) { - List list = lookupDepartures(station.getStrain(), "S2", arrival); - departureList.addAll(list); + if (station.getStrain() != null && (type == FetchTrainType.STOG||type == FetchTrainType.BOTH)) { + DepartureBean tempBean = lookupDepartures(station.getStrain(), TrainType.STOG, arrival); + departureBean.entries.addAll( tempBean.entries ); + departureBean.notifications.addAll(tempBean.notifications); } - Collections.sort( departureList ); + if (departureBean.entries.size() == 0) { + logger.info("No departures found for station " + stationID); + } + + if (type == FetchTrainType.BOTH) { //if we have both S-tog and regional order by departure/arrival time + Collections.sort( departureBean.entries ); + } - return departureList; + return departureBean; } - public List lookupDepartures(String stationcode, String type, boolean arrival) throws Exception { - if (useTempSite == false) { - return lookupDeparturesNormalSite(stationcode, type, arrival); - } else { - return lookupDeparturesFromTemporarySite(stationcode, type); + public DepartureBean lookupDepartures(String stationcode, TrainType type, boolean arrival) throws Exception { + if ( settings.getUseAzureSite() == true) { + return lookupDeparturesAzureSite(stationcode, type, arrival); + } else { + return lookupDeparturesWwwSite(stationcode, type, arrival); } } - public List lookupDeparturesNormalSite(String stationcode, String type, boolean arrival) throws Exception { - - List departureList = new ArrayList(); + private String getTypeStringAzure(TrainType type) { + switch (type) { + case STOG: + return "S-Tog"; + case REGIONAL: + return "Fjerntog"; + default: + return ""; //Can not happen + } + } + + private String getTypeStringWww(TrainType type) { + switch (type) { + case STOG: + return "S2"; + case REGIONAL: + return "FJRN"; + default: + return ""; //Can not happen + } + } + + public DepartureBean lookupDeparturesAzureSite(String stationcode, TrainType type, boolean arrival) throws Exception { - final WebClient webClient = new WebClient( BrowserVersion.FIREFOX_3 ); - webClient.setTimeout(2500); - webClient.setJavaScriptEnabled(false); - - String uri = "http://www.bane.dk/visStation.asp?ArtikelID=4275&W=" + type + "&S=" + stationcode; - BanedkInvocation wrapper = new BanedkInvocation(webClient, uri); - CircuitBreaker breaker = CircuitBreakerManager.getManager().getCircuitBreaker("banedk"); + DepartureBean departureBean = new DepartureBean(); + - HtmlPage page = (HtmlPage) breaker.invoke(wrapper); + String typeString = getTypeStringAzure(type); + String arrivalDeparture = (arrival==false) ? "Afgang" : "Ankomst"; - String tableName = arrival == false ? "afgangtabel" : "ankomsttabel"; + stationcode = URLEncoder.encode(stationcode,"ISO-8859-1"); + + String uri = "http://trafikinfo.bane.dk/Trafikinformation/AfgangAnkomst/" + arrivalDeparture + "/" + stationcode + "/" + typeString + "/UdvidetVisning"; - System.out.println(uri); - System.out.println(tableName); + logger.fine("URI: " + uri); + JsoupInvocation wrapper = new JsoupInvocation( new URL(uri), settings.getReplyTimeout() ); + CircuitBreaker breaker = CircuitBreakerManager.getManager().getCircuitBreaker("banedk"); + + Document page = (Document) breaker.invoke(wrapper); - HtmlElement table = page.getElementById(tableName); + String tableName = arrival == false ? "afgangtabel" : "ankomsttabel"; + Element table = page.getElementById(tableName); if (table != null) { - DomNodeList tableRows = table.getElementsByTagName("tr"); + Elements tableRows = table.getElementsByTag("tr"); + + //boolean tidsstregExists = (table.getElementsByAttributeValue("class", "Tidsstreg").size() > 0); + //boolean passedTidsstreg = false; - for (HtmlElement currentRow : tableRows) { - String rowClass = currentRow.getAttribute("class"); + for (Element currentRow : tableRows) { + String rowClass = currentRow.attr("class"); + /* + if (tidsstregExists == true && passedTidsstreg == false) { + if (currentRow.getElementsByAttributeValue("class", "Tidsstreg").size() > 0) { + passedTidsstreg = true; + } else { + continue; + } + }*/ + if (rowClass != null && rowClass.toLowerCase().contains("station") ) { - DomNodeList fields = currentRow.getElementsByTagName("td"); + + Elements fields = currentRow.getElementsByTag("td"); - DepartureBean departure = new DepartureBean(); + DepartureEntry departure = new DepartureEntry(); - String time = fields.get(0).asText(); + String time = fields.get(0).text(); if (time.equals("")) time = "0:00"; //Bane.dk bug work-around departure.setTime(time); @@ -121,111 +182,157 @@ int updated = extractUpdated( fields.get(1) ); departure.setUpdated(updated); - String trainNumber = fields.get(2).asText(); - if (type.equalsIgnoreCase("S2")) //If it is S-train we need to extract the trainNumber - trainNumber = trainNumber + " " + extractTrainNumber(fields.get(2)); + String trainNumber = fields.get(2).text(); + if (type == TrainType.STOG) //If it is S-train we need to extract the trainNumber + trainNumber = trainNumber + " " + extractTrainNumberAzure(fields.get(2)); departure.setTrainNumber(trainNumber); - String destination = fields.get(3).asText(); + String destination = fields.get(3).text(); departure.setDestination(destination); - String origin = fields.get(4).asText(); + String origin = fields.get(4).text(); departure.setOrigin(origin); - String location = fields.get(5).asText(); + String location = fields.get(5).text(); departure.setLocation(location); - String status = fields.get(6).asText(); + String status = fields.get(6).text().trim(); departure.setStatus(status); String note = extractNote( fields.get(7) ); departure.setNote(note); - departure.setType(type); + departure.setType(typeString); - departureList.add(departure); + departureBean.entries.add( departure ); } } } else { logger.warning("No departures found for station=" + stationcode + ", type=" + type); } - webClient.closeAllWindows(); - return departureList; + Element notifDiv = page.getElementById("station_planlagte_text"); + if (notifDiv != null) { + + Elements tables = notifDiv.getElementsByTag("table"); + for (Element tab : tables) { + + Elements anchors = tab.getElementsByTag("a"); + if (anchors.size() == 2) { + departureBean.notifications.add( anchors.get(1).text() ); + } + } + + } + + + return departureBean; } - public List lookupDeparturesFromTemporarySite(String stationcode, String type) throws Exception { + + + public static String cleanText(String input) { + //apparently JSoup translates   characters on www.bane.dk to 0xA0 + return input.replace((char) 0xA0, (char)0x20).trim(); + } + + public DepartureBean lookupDeparturesWwwSite(String stationcode, TrainType trainType, boolean arrival) throws Exception { + + DepartureBean departureBean = new DepartureBean(); + + String type = getTypeStringWww(trainType); - List departureList = new ArrayList(); + stationcode = URLEncoder.encode(stationcode, "ISO-8859-1"); - final WebClient webClient = new WebClient(BrowserVersion.FIREFOX_3); - webClient.setTimeout(2500); - webClient.setJavaScriptEnabled(false); + + String uri = "http://www.bane.dk/visStation.asp?ArtikelID=4275&W=" + type + "&S=" + stationcode; + logger.fine("URI:" + uri); - String uri = "http://bane.dk/lite/station.asp?w=" + type + "&s=" + stationcode; - - BanedkInvocation wrapper = new BanedkInvocation(webClient, uri); + JsoupInvocation wrapper = new JsoupInvocation( new URL(uri), settings.getReplyTimeout() ); CircuitBreaker breaker = CircuitBreakerManager.getManager().getCircuitBreaker("banedk"); - HtmlPage page = (HtmlPage) breaker.invoke(wrapper); + Element page = (Element) breaker.invoke(wrapper); + + String tableName = arrival == false ? "afgangtabel" : "ankomsttabel"; + Element table = page.getElementById(tableName); - HtmlElement table = page.getElementById("traf_afgang"); + - if (table != null) { - DomNodeList tableRows = table.getElementsByTagName("tr"); + if (table != null) { + Elements tableRows = table.getElementsByTag("tr"); - boolean isFirst = true; + //boolean passedTidsstreg = false; + //boolean tidsstregExists = (table.getElementsByAttributeValue("class", "Tidsstreg").size() > 0); - for (HtmlElement currentRow : tableRows) { - if (isFirst == true) { //skip table headers - isFirst = false; - continue; - } + for (Element currentRow : tableRows) { + String rowClass = currentRow.attr("class"); + /* + if (tidsstregExists == true && passedTidsstreg == false) { + if (currentRow.getElementsByAttributeValue("class", "Tidsstreg").size() > 0) { + passedTidsstreg = true; + } else { + continue; + } + }*/ - DomNodeList fields = currentRow.getElementsByTagName("td"); - - DepartureBean departure = new DepartureBean(); - - String time = fields.get(0).asText().trim(); - - if (time.equals("")) - time = "0:00"; //Bane.dk bug work-around - departure.setTime(time); - - - String trainNumber = fields.get(1).asText(); - departure.setTrainNumber(trainNumber); - - String destination = fields.get(2).asText(); - departure.setDestination(destination); - - String origin = fields.get(3).asText(); - departure.setOrigin(origin); - - String status = fields.get(4).asText(); - departure.setStatus(status); - - String note = fields.get(5).asText(); - departure.setNote(note); + + if (rowClass != null && rowClass.toLowerCase().contains("station") ) { + Elements fields = currentRow.getElementsByTag("td"); + + DepartureEntry departure = new DepartureEntry(); + - departureList.add(departure); + + String time = cleanText( fields.get(0).getAllElements().get(2).text() ); + if (time.equals("")) + time = "0:00"; //Bane.dk bug work-around + departure.setTime(time); + + int updated = extractUpdated( fields.get(1) ); + departure.setUpdated(updated); + + String trainNumber = cleanText( fields.get(2).text() ); + if (type.equalsIgnoreCase("S2")) //If it is S-train we need to extract the trainNumber + trainNumber = trainNumber + " " + extractTrainNumberWww(fields.get(2)); + departure.setTrainNumber(trainNumber); + + String destination = cleanText( fields.get(3).text() ); + departure.setDestination(destination); + + String origin = cleanText( fields.get(4).text() ); + departure.setOrigin(origin); + + String location = cleanText( fields.get(5).text() ); + departure.setLocation(location); + + String status = cleanText( fields.get(6).text() ); + departure.setStatus(status); + + String note = cleanText( extractNote( fields.get(7) ) ); + departure.setNote(note); + + departure.setType(type); + + departureBean.entries.add(departure); + + + } } } else { logger.warning("No departures found for station=" + stationcode + ", type=" + type); } - webClient.closeAllWindows(); - return departureList; + return departureBean; } - + - private int extractUpdated(HtmlElement updatedTd) { //extract the digit (in this case: 4) from "media/trafikinfo/opdater4.gif" + private int extractUpdated(Element updatedTd) { //extract the digit (in this case: 4) from "media/trafikinfo/opdater4.gif" int updated = -1; - DomNodeList updatedImgs = updatedTd.getElementsByTagName("img"); - String updatedStr = updatedImgs.get(0).getAttribute("src"); + Elements updatedImgs = updatedTd.getElementsByTag("img"); + String updatedStr = updatedImgs.get(0).attr("src"); if (updatedStr != null) { for (int i=0; i elems = noteTd.getElementsByAttribute("span", "class", "bemtype"); + Elements elems = noteTd.getElementsByClass("bemtype"); if (elems.size() > 0 && note.charAt(note.length()-1) == 'i') note = note.substring(0,note.length() -1 ); - return note; + return note.trim(); } - private String extractTrainNumber(HtmlElement trainTd) { + private String extractTrainNumberAzure(Element trainTd) { + Element anchorElement = trainTd.getElementsByTag("a").get(0); + String href = anchorElement.attr("href"); + + int pos = href.lastIndexOf('/'); + String number = href.substring(pos+1); + + return number; + } + + private String extractTrainNumberWww(Element trainTd) { String number = ""; - HtmlElement anchorElement = trainTd.getElementsByTagName("a").get(0); - String href = anchorElement.getAttribute("href"); + Element anchorElement = trainTd.getElementsByTag("a").get(0); + String href = anchorElement.attr("href"); String argstring = href.substring( href.indexOf('?') + 1); String args[] = argstring.split("&"); @@ -266,6 +384,7 @@ return number; } + //test /*