X-Git-Url: http://git.wpitchoune.net/gitweb/?a=blobdiff_plain;f=war%2Fsrc%2Fmain%2Fjava%2Fpnews%2Fservlet%2FArticleProvider.java;h=1dece09c19970607c6dd36571ad6858f2854bccf;hb=538d228606be7cd7e432a9086cb04552cf2b0958;hp=010317a20a53a6721d6ebaadc98240c2de707a8c;hpb=1737ff8517d83a0f2cf6a2d223d26b116a8eb834;p=pnews.git diff --git a/war/src/main/java/pnews/servlet/ArticleProvider.java b/war/src/main/java/pnews/servlet/ArticleProvider.java index 010317a..1dece09 100644 --- a/war/src/main/java/pnews/servlet/ArticleProvider.java +++ b/war/src/main/java/pnews/servlet/ArticleProvider.java @@ -1,7 +1,6 @@ package pnews.servlet; import java.io.IOException; -import java.io.InputStream; import java.net.MalformedURLException; import java.net.URL; import java.util.ArrayList; @@ -18,13 +17,13 @@ import java.util.logging.Level; import java.util.logging.Logger; import org.jsoup.Jsoup; -import org.xml.sax.InputSource; import com.rometools.rome.feed.synd.SyndEnclosure; import com.rometools.rome.feed.synd.SyndEntry; import com.rometools.rome.feed.synd.SyndFeed; import com.rometools.rome.io.FeedException; import com.rometools.rome.io.SyndFeedInput; +import com.rometools.rome.io.XmlReader; import pnews.Article; import pnews.Category; @@ -37,15 +36,15 @@ public class ArticleProvider { private ArticleProvider() { for (Category cat:Category.values()) - scheduler.scheduleAtFixedRate(new Refresher(cat), 2, 120, TimeUnit.SECONDS); + scheduler.scheduleAtFixedRate(new Refresher(cat), 2, 600, TimeUnit.SECONDS); } private static SyndFeed getSyndFeed(String u) throws IllegalArgumentException, FeedException, MalformedURLException, IOException { - InputStream is = new URL(u).openConnection().getInputStream(); - InputSource source = new InputSource(is); - - return new SyndFeedInput().build(source); + XmlReader r; + r = new XmlReader(new URL(u)); + + return new SyndFeedInput().build(r); } private static Map getFeeds() { @@ -55,32 +54,42 @@ public class ArticleProvider { result.put(Category.TOP, new String[] { + "http://www.europe1.fr/var/export/rss/europe1/actus.xml", "http://www.francetvinfo.fr/titres.rss", - "http://www.france24.com/fr/actualites/rss", - //"https://www.franceinter.fr/rss/a-la-une.xml", "http://www.rfi.fr/general/rss", "http://www.cnews.fr/rss/une", + "http://www.ladepeche.fr/rss/a-la-une.rss", + "https://www.franceinter.fr/rss/a-la-une.xml", + "https://www.francebleu.fr/rss/a-la-une.xml", "http://www.bfmtv.com/rss/info/flux-rss/flux-toutes-les-actualites/" }); result.put(Category.SPORT, - new String[] { "http://www.france24.com/fr/sports/rss" }); + new String[] { "http://www.europe1.fr/var/export/rss/europe1/sport.xml", + "http://www.sportune.fr/feed", + "http://www.france24.com/fr/sports/rss" }); result.put(Category.FRANCE, new String[] { "http://www.france24.com/fr/france/rss", + "http://www.francetvinfo.fr/france.rss", "http://www.rfi.fr/france/rss"}); result.put(Category.EUROPE, new String[] { "http://www.france24.com/fr/europe/rss" }); + + result.put(Category.MONDE, + new String[] { "http://www.europe1.fr/var/export/rss/europe1/international.xml", + "http://www.france24.com/fr/actualites/rss" }); + result.put(Category.ECO, new String[] { "http://www.france24.com/fr/economie/rss", + "http://www.europe1.fr/var/export/rss/europe1/economie.xml", "http://www.rfi.fr/economie/rss" }); result.put(Category.ESSONNE, new String[] { "http://www.tourisme-essonne.com/rss/actus/", - "http://www.ville-palaiseau.fr/rss/actualites.htm" - /*"https://www.essonneinfo.fr/feed/"*/ }); + "http://www.ville-palaiseau.fr/rss/actualites.htm" }); result.put(Category.PEOPLE, new String[] { "http://www.premiere.fr/rss/actu-live", @@ -88,74 +97,120 @@ public class ArticleProvider { }); result.put(Category.TECHNOLOGIE, - new String[] { "http://feeds.feedburner.com/lesnumeriques/news", - "http://www.zdnet.fr/feeds/rss/actualites/"}); + new String[] { "http://www.generation-nt.com/export/rss.xml", + "http://www.europe1.fr/var/export/rss/europe1/sciences.xml", + "http://feeds.feedburner.com/lesnumeriques/news", + "http://www.zdnet.fr/feeds/rss/actualites/", + "http://www.frandroid.com/feed", + "http://www.silicon.fr/feed", + "http://www.fredzone.org/feed", + "http://www.futura-sciences.com/rss/actualites.xml", + "https://www-03.ibm.com/press/fr/fr/rssfeed.wss?keyword=null&maxFeed=&feedType=RSS&topic=all"}); return result; } - private void addArticles(Category cat, SyndFeed feed) { - String thumbnail; - String desc, link, title; + private List
getArticlesForUpdate(Category cat) { + List
result; + + synchronized (articlesByCategory) { + result = articlesByCategory.get(cat); + if (result == null) { + result = new ArrayList<>(); + articlesByCategory.put(cat, result); + } + return result; + } + } + + private boolean exists(String articleLink, List
articles) { + synchronized (articles) { + for (Article a: articles) + if (a.link.equals(articleLink)) + return true; + } + return false; + } + + private static Article toArticle(String link, SyndEntry entry, SyndFeed feed) { + String desc, title, thumbnail, feedTitle, str; Date date; + + feedTitle = feed.getTitle(); + if (feedTitle != null) { + feedTitle = feedTitle.trim(); + } + + thumbnail = null; + for (SyndEnclosure e: entry.getEnclosures()) { + if (e.getType().startsWith("image/")) + thumbnail = e.getUrl(); + break; + } + + if (thumbnail == null && feed.getImage() != null) + thumbnail = feed.getImage().getUrl(); + + + title = entry.getTitle().trim(); + + if (entry.getDescription() != null) { + str = entry.getDescription().getValue(); + desc = Jsoup.parse(str).text(); + } else { + desc = null; + LOG.severe("No description for " + feedTitle + " - " + title); + } + + date = entry.getPublishedDate(); + if (date == null) + date = entry.getUpdatedDate(); + if (date == null) + LOG.severe("The article " + feedTitle + " - " + title + " does not have a date"); + + return new Article(link, title, desc, thumbnail, date, feedTitle); + } + + private void addArticles(Category cat, SyndFeed feed) { + String feedTitle; List
articles; - boolean exist; + Article a; - LOG.info("addArticles" + cat.getId()); + feedTitle = feed.getTitle().trim(); + + LOG.info("addArticles " + cat.getId() + " " + feedTitle + " number of articles: " + feed.getEntries().size()); for (SyndEntry entry: feed.getEntries()) { - thumbnail = null; - for (SyndEnclosure e: entry.getEnclosures()) { - if (e.getType().startsWith("image/")) - thumbnail = e.getUrl(); - break; - } - - if (entry.getDescription() != null) { - desc = Jsoup.parse(entry.getDescription().getValue()).text(); - } else { - desc = null; - LOG.severe("No description for " + feed.getTitle() + " - " + entry.getTitle()); + String link = entry.getLink().trim(); + articles = getArticlesForUpdate(cat); + if (exists(link, articles)) { + LOG.fine("addArticles " + link + " is already present"); + continue ; } - date = entry.getPublishedDate(); - if (date == null) - date = entry.getUpdatedDate(); + a = ArticleStore.singleton.getArticle(link, ()->toArticle(link, entry, feed)); - synchronized(articlesByCategory) { - link = entry.getLink().trim(); - title = entry.getTitle().trim(); - articles = articlesByCategory.get(cat); - exist = false; - if (articles == null) { - articles = new ArrayList<>(); - articlesByCategory.put(cat, articles); - } else { - for (Article a: articles) - if (a.link.equals(link)) { - LOG.finest(link + " already present"); - exist = true; - } - } - - if (!exist) { - LOG.info("add " + cat.getId() + " " + feed.getTitle() + " " + title); - - articles.add(new Article(link, cat, title, desc, thumbnail, date, - feed.getTitle())); + synchronized (articles) { + articles.add(a); - Collections.sort(articles, new Comparator
() { - @Override - public int compare(Article o1, Article o2) { - return o2.publicationDate.compareTo(o1.publicationDate); - } - - }); - } + Collections.sort(articles, new Comparator
() { + @Override + public int compare(Article o1, Article o2) { + if (o1.publicationDate == o2.publicationDate) + return 0; + if (o1.publicationDate == null) + return 1; + if (o2.publicationDate == null) + return -1; + return o2.publicationDate.compareTo(o1.publicationDate); + } + }); } - } + } + + LOG.info("addArticles done " + cat.getId()); } - + private void retrieveArticles(Category cat) throws IllegalArgumentException, MalformedURLException, FeedException, IOException { String[] feeds; @@ -165,17 +220,28 @@ public class ArticleProvider { for (String str: feeds) try { addArticles(cat, getSyndFeed(str)); - } catch (IOException e) { - LOG.log(Level.SEVERE, "retrieveArticles failure " + cat.getId(), e); + } catch (Throwable e) { + LOG.log(Level.SEVERE, + "retrieveArticles failure " + cat.getId() + " " + str, + e); } else LOG.severe("No feed for category " + cat); } + /** + * Returns a copy. + */ public List
getArticles(Category cat) throws IllegalArgumentException, MalformedURLException, FeedException, IOException { + List
articles; + synchronized (articlesByCategory) { - return articlesByCategory.get(cat); + articles = getArticlesForUpdate(cat); + } + + synchronized (articles) { + return new ArrayList<>(articles); } } @@ -187,22 +253,11 @@ public class ArticleProvider { } @Override - public void run() { - List
articles; - + public void run() { LOG.info("refresher "+ category.getId()); try { retrieveArticles(category); - - synchronized (articlesByCategory) { - articles = articlesByCategory.get(category); - if (articles != null && articles.size() > 100) { - articlesByCategory.put(category, - articles.subList(0, 100)); - - } - } } catch (IllegalArgumentException | FeedException | IOException e) { LOG.log(Level.SEVERE, "refresher failure", e); }