public static List<RssUrlBean> getRssUrlBeanListFromPage(int rssCompo_id, String url) {
List<RssUrlBean> rubList = new ArrayList<RssUrlBean>();
if (url.equals("")) return rubList;
try {
URL feedUrl = new URL(url);
// SyndFeedInput:从远程读到xml结构的内容转成SyndFeedImpl实例
SyndFeedInput input = new SyndFeedInput();
// rome按SyndFeed类型生成rss和atom的实例,
// SyndFeed是rss和atom实现类SyndFeedImpl的接口
SyndFeed syndFeed = input.build(new XmlReader(feedUrl));
List<SyndEntry> entryList = syndFeed.getEntries();
for (SyndEntry entry : entryList) {
RssUrlBean rub = new RssUrlBean();
rub.setRssCompo_id(rssCompo_id);
rub.setTitle(entry.getTitle());
rub.setLink(entry.getUri());
rub.setPublishedDate(CommonUtil.getStandardDate(entry.getPublishedDate().toLocaleString()));
rub.setDescription(entry.getDescription().getValue());
if (entry.getUpdatedDate() != null)
rub.setUpdatedDate(CommonUtil.getStandardDate(entry.getUpdatedDate().toLocaleString()));
rub.setAuthors(entry.getAuthor());
rubList.add(rub);
}
} catch (Exception ex) {
ex.printStackTrace();
}
return rubList;
}
// @Transactional
public int importData(
final SyndFeed feed,
Set<KiWiUriResource> types,
Set<ContentItem> tags,
final User user,
final Collection<ContentItem> output) {
log.info(
"importing entries from #0 feed '#1' found at '#2'",
feed.getFeedType(),
feed.getTitle(),
feed.getUri());
if (types == null) {
types = new HashSet<KiWiUriResource>();
}
if (tags == null) {
tags = new HashSet<ContentItem>();
}
final Set<ContentItem> my_tags = tags;
final Set<KiWiUriResource> my_types = types;
// a hack for importing facebook activity streams: if the type is kiwi:FacebookPost,
// turn facebook activity stream mode on; in this mode, we will skip all entries where
// the remote author name and local user name are not identical
boolean facebookImport = false;
String t_facebookPost = Constants.NS_KIWI_CORE + "FacebookPost";
for (KiWiUriResource r : types) {
if (r.getUri().equals(t_facebookPost)) {
facebookImport = true;
break;
}
}
for (final SyndEntry entry : (List<SyndEntry>) feed.getEntries()) {
// facebook hack ... (see above)
if (facebookImport
&& !entry.getAuthor().equalsIgnoreCase(user.getFirstName() + " " + user.getLastName())) {
log.info("Facebook import: skipping friend post with title", entry.getTitle());
continue;
}
new RunAsOperation() {
@Override
public void execute() {
importEntry(feed, entry, my_types, my_tags, user, output);
}
}.addRole("admin").run();
}
// entityManager.flush();
log.info("#0 content items have been imported from RSS/Atom feed", feed.getEntries().size());
return feed.getEntries().size();
}
private Article mapArticle(SyndEntry syndEntry) {
StringBuilder sb = new StringBuilder();
for (Object obj : syndEntry.getContents()) {
if (!(obj instanceof SyndContent)) {
continue;
}
SyndContent syndContent = (SyndContent) obj;
sb.append(syndContent.getValue());
}
return Article.builder()
.num(0)
.title(syndEntry.getTitle())
.content(sb.toString())
.description(syndEntry.getDescription().getValue())
.author(syndEntry.getAuthor())
.image("")
.writtenDate(syndEntry.getPublishedDate())
.build();
}
// @Transactional
// @TransactionAttribute(TransactionAttributeType.REQUIRES_NEW)
public void importEntry(
final SyndFeed feed,
final SyndEntry entry,
final Set<KiWiUriResource> types,
final Set<ContentItem> tags,
User user,
final Collection<ContentItem> output) {
if (user == null && entry.getAuthor() != null && !"".equals(entry.getAuthor())) {
if (userService.userExists(entry.getAuthor())) {
user = userService.getUserByLogin(entry.getAuthor());
} else {
// user = userService.createUser(entry.getAuthor());
/* In my opinion, it is not ok to create a user entity
* without asking the person if he/she wants to be
* created and persisted in the KiWi dataset.
* Thus I'm changing the user to 'anonymous',
* if he/she is'nt registered with the same nick that
* is given in the rss entry.
*/
user = userService.getUserByLogin("anonymous");
kiwiEntityManager.persist(user);
}
}
log.debug("feed entry: #0 (#1)", entry.getTitle(), entry.getUri());
// create a new content item and copy all data from the feed entry
ContentItem item;
if (entry.getLink() != null) {
item = contentItemService.createExternContentItem(entry.getLink());
} else if (entry.getUri() != null) {
try {
// try parsing URI; if it is not valid,
URI uri = new URI(entry.getUri());
item = contentItemService.createExternContentItem(entry.getUri());
} catch (URISyntaxException e) {
item = contentItemService.createExternContentItem(feed.getLink() + "#" + entry.getUri());
}
} else {
item = contentItemService.createContentItem();
}
contentItemService.updateTitle(item, entry.getTitle());
if (feed.getLanguage() != null) item.setLanguage(new Locale(feed.getLanguage()));
if (entry.getPublishedDate() != null) {
item.setCreated(entry.getPublishedDate());
item.setModified(entry.getPublishedDate());
}
if (entry.getUpdatedDate() != null) {
if (entry.getPublishedDate() == null) {
item.setCreated(entry.getUpdatedDate());
}
item.setModified(entry.getUpdatedDate());
}
item.setAuthor(user);
// read feed content and set it as item's text content
List<SyndContent> contents = entry.getContents();
if (contents.size() == 1) {
log.debug("using RSS content section provided by item");
contentItemService.updateTextContentItem(item, "<p>" + contents.get(0).getValue() + "</p>");
} else if (contents.size() > 1) {
log.warn("feed entry contained more than one content section");
contentItemService.updateTextContentItem(item, "<p>" + contents.get(0).getValue() + "</p>");
} else if (contents.size() == 0) {
if (entry.getDescription() != null && entry.getDescription().getValue() != null) {
log.debug("using RSS description as no content section was available");
contentItemService.updateTextContentItem(
item, "<p>" + entry.getDescription().getValue() + "</p>");
}
}
// save before tagging
contentItemService.saveContentItem(item);
// read feed categories and use them as tags
for (SyndCategory cat : (List<SyndCategory>) entry.getCategories()) {
ContentItem _cat;
if (!taggingService.hasTag(item, cat.getName())) {
if (cat.getTaxonomyUri() != null) {
_cat = contentItemService.getContentItemByUri(cat.getTaxonomyUri());
if (_cat == null) {
_cat = contentItemService.createExternContentItem(cat.getTaxonomyUri());
contentItemService.updateTitle(_cat, cat.getName());
_cat.setAuthor(user);
contentItemService.saveContentItem(_cat);
}
taggingService.createTagging(cat.getName(), item, _cat, user);
} else {
_cat = contentItemService.getContentItemByTitle(cat.getName());
if (_cat == null) {
_cat = contentItemService.createContentItem();
contentItemService.updateTitle(_cat, cat.getName());
_cat.setAuthor(user);
contentItemService.saveContentItem(_cat);
}
taggingService.createTagging(cat.getName(), item, _cat, user);
}
}
}
// scan for Twitter-style hash tags in title (e.g. #kiwiknows, see KIWI-622)
Matcher m_hashtag = p_hashtag.matcher(entry.getTitle());
while (m_hashtag.find()) {
String tag_label = m_hashtag.group(1);
if (!taggingService.hasTag(item, tag_label)) {
ContentItem tag = contentItemService.getContentItemByTitle(tag_label);
if (tag == null) {
tag = contentItemService.createContentItem();
contentItemService.updateTitle(tag, tag_label);
tag.setAuthor(user);
contentItemService.saveContentItem(tag);
}
taggingService.createTagging(tag_label, item, tag, user);
}
}
// check for geo information
GeoRSSModule geoRSSModule = GeoRSSUtils.getGeoRSS(entry);
if (geoRSSModule != null && geoRSSModule.getPosition() != null) {
POI poi = kiwiEntityManager.createFacade(item, POI.class);
poi.setLatitude(geoRSSModule.getPosition().getLatitude());
poi.setLongitude(geoRSSModule.getPosition().getLongitude());
kiwiEntityManager.persist(poi);
}
// check for media information
MediaEntryModule mediaModule = (MediaEntryModule) entry.getModule(MediaModule.URI);
if (mediaModule != null) {
MediaContent[] media = mediaModule.getMediaContents();
if (media.length > 0) {
MediaContent m = media[0];
if (m.getReference() instanceof UrlReference) {
URL url = ((UrlReference) m.getReference()).getUrl();
String type = m.getType();
String name = url.getFile();
if (name.lastIndexOf("/") > 0) {
name = name.substring(name.lastIndexOf("/") + 1);
}
log.debug("importing media data from URL #0", url.toString());
try {
InputStream is = url.openStream();
ByteArrayOutputStream bout = new ByteArrayOutputStream();
int c;
while ((c = is.read()) != -1) {
bout.write(c);
}
byte[] data = bout.toByteArray();
contentItemService.updateMediaContentItem(item, data, type, name);
is.close();
bout.close();
} catch (IOException ex) {
log.error("error importing media content from RSS stream");
}
} else {
log.info("RSS importer can only import media with URL references");
}
} else {
log.warn("media module found without content");
}
Category[] cats = mediaModule.getMetadata().getCategories();
for (Category cat : cats) {
ContentItem _cat;
String label = cat.getLabel() != null ? cat.getLabel() : cat.getValue();
if (!taggingService.hasTag(item, label)) {
if (cat.getScheme() != null) {
_cat = contentItemService.getContentItemByUri(cat.getScheme() + cat.getValue());
if (_cat == null) {
_cat = contentItemService.createExternContentItem(cat.getScheme() + cat.getValue());
contentItemService.updateTitle(_cat, label);
_cat.setAuthor(user);
contentItemService.saveContentItem(_cat);
}
taggingService.createTagging(label, item, _cat, user);
} else {
_cat = contentItemService.getContentItemByTitle(label);
if (_cat == null) {
_cat = contentItemService.createContentItem();
contentItemService.updateTitle(_cat, label);
_cat.setAuthor(user);
contentItemService.saveContentItem(_cat);
}
taggingService.createTagging(label, item, _cat, user);
}
}
}
}
// add parameter categories as tags
for (ContentItem tag : tags) {
if (!taggingService.hasTag(item, tag.getTitle())) {
taggingService.createTagging(tag.getTitle(), item, tag, user);
}
}
// add parameter types as types
for (KiWiUriResource type : types) {
item.addType(type);
}
// add kiwi:FeedPost type
item.addType(tripleStore.createUriResource(Constants.NS_KIWI_CORE + "FeedPost"));
/* the flush is necessary, because CIs or tags will
* otherwise be created multiple times when they
* appear more than once in one RSS feed */
entityManager.flush();
log.debug("imported content item '#0' with URI '#1'", item.getTitle(), item.getResource());
}
/**
* Generate an ebook from an RSS DOM Document.
*
* @param url The URL from where the Document was fetched (used only to set the author metadata)
* @param doc The DOM Document of the feed.
* @return An ebook.
* @throws IllegalArgumentException
* @throws FeedException
* @throws IOException
*/
private static Book createBookFromFeed(URL url, Document doc, List<Keyword> keywords)
throws IllegalArgumentException, FeedException, IOException {
Book book = new Book();
// start parsing our feed and have the above onItem methods called
SyndFeedInput input = new SyndFeedInput();
SyndFeed feed = input.build(doc);
System.out.println(feed);
// Set the title
book.getMetadata().addTitle(feed.getTitle());
// Add an Author
String author = feed.getAuthor();
if (author == null || "".equals(author.trim())) {
author = url.getHost();
}
book.getMetadata().addAuthor(new Author(author));
if (feed.getPublishedDate() != null) {
book.getMetadata().addDate(new nl.siegmann.epublib.domain.Date(feed.getPublishedDate()));
}
if (feed.getDescription() != null) {
book.getMetadata().addDescription(feed.getDescription());
}
if (feed.getCopyright() != null) {
book.getMetadata().getRights().add(feed.getCopyright());
}
// Set cover image - This has never worked.
// if (feed.getImage() != null) {
// System.out.println("There is an image for the feed");
// Promise<HttpResponse> futureImgResponse =
// WS.url(feed.getImage().getUrl()).getAsync();
// HttpResponse imgResponse = await(futureImgResponse);
// System.out.println("Content-type: " + imgResponse.getContentType());
// if (imgResponse.getContentType().startsWith("image/")) {
// String extension =
// imgResponse.getContentType().substring("image/".length());
// InputStream imageStream = imgResponse.getStream();
// book.getMetadata().setCoverImage(new Resource(imageStream, "cover." +
// extension));
// System.out.println("Using default cover");
// imageStream =
// VirtualFile.fromRelativePath("assets/cover.png").inputstream();
// if (imageStream != null) {
// System.out.println("Using default cover");
// book.getMetadata().setCoverImage(new Resource(imageStream,
// "cover.png"));
// } else {
// System.out.println("Could not load default cover");
// }
// }
// }
int entryNumber = 0;
List<SyndEntry> entries = feed.getEntries();
for (SyndEntry entry : entries) {
if (matchesKeyword(entry, keywords)) {
StringBuilder title = new StringBuilder(100);
if (entry.getTitle() != null) {
title.append(entry.getTitle());
}
if (entry.getAuthor() != null) {
title.append(" - ").append(entry.getAuthor());
}
StringBuilder content = new StringBuilder();
// Add title inside text
content.append("<h2>").append(title).append("</h2>");
if (entry.getDescription() != null) {
SyndContent syndContent = (SyndContent) entry.getDescription();
if (!syndContent.getType().contains("html")) {
content.append("<pre>\n");
}
content.append(syndContent.getValue());
if (!syndContent.getType().contains("html")) {
content.append("\n</pre>");
}
content.append("<hr/>");
}
if (entry.getContents().size() > 0) {
SyndContent syndContent = (SyndContent) entry.getContents().get(0);
if (!syndContent.getType().contains("html")) {
content.append("<pre>\n");
}
content.append(syndContent.getValue());
if (!syndContent.getType().contains("html")) {
content.append("\n</pre>");
}
}
String strContent = clean(content.toString());
// Add Chapter
try {
entryNumber++;
book.addSection(
title.toString(),
new Resource(new StringReader(strContent), "entry" + entryNumber + ".xhtml"));
} catch (IOException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
}
}
return book;
}
public String getAuthor() {
return entry.getAuthor();
}
// build a SubscriptionEntry from Rome SyndEntry and SyndFeed
private SubscriptionEntry buildEntry(SyndEntry romeEntry) {
// if we don't have a permalink then we can't continue
if (romeEntry.getLink() == null) {
return null;
}
SubscriptionEntry newEntry = new SubscriptionEntry();
newEntry.setTitle(romeEntry.getTitle());
newEntry.setPermalink(romeEntry.getLink());
// Play some games to get the author
DCModule entrydc = (DCModule) romeEntry.getModule(DCModule.URI);
if (romeEntry.getAuthor() != null) {
newEntry.setAuthor(romeEntry.getAuthor());
} else {
newEntry.setAuthor(entrydc.getCreator()); // use <dc:creator>
}
// Play some games to get the updated date
if (romeEntry.getUpdatedDate() != null) {
newEntry.setUpdateTime(new Timestamp(romeEntry.getUpdatedDate().getTime()));
}
// TODO: should we set a default update time here?
// And more games getting publish date
if (romeEntry.getPublishedDate() != null) {
newEntry.setPubTime(new Timestamp(romeEntry.getPublishedDate().getTime())); // use <pubDate>
} else if (entrydc != null && entrydc.getDate() != null) {
newEntry.setPubTime(new Timestamp(entrydc.getDate().getTime())); // use <dc:date>
} else {
newEntry.setPubTime(newEntry.getUpdateTime());
}
// get content and unescape if it is 'text/plain'
if (romeEntry.getContents().size() > 0) {
SyndContent content = (SyndContent) romeEntry.getContents().get(0);
if (content != null && content.getType().equals("text/plain")) {
newEntry.setText(StringEscapeUtils.unescapeHtml(content.getValue()));
} else if (content != null) {
newEntry.setText(content.getValue());
}
}
// no content, try summary
if (newEntry.getText() == null || newEntry.getText().trim().length() == 0) {
if (romeEntry.getDescription() != null) {
newEntry.setText(romeEntry.getDescription().getValue());
}
}
// copy categories
if (romeEntry.getCategories().size() > 0) {
List list = new ArrayList();
Iterator cats = romeEntry.getCategories().iterator();
while (cats.hasNext()) {
SyndCategory cat = (SyndCategory) cats.next();
list.add(cat.getName());
}
newEntry.setCategoriesString(list);
}
return newEntry;
}
