Two kinds of paging operations for Lucene

Lucene-based paging is available in two ways:

lucene3.5 before paging is provided in the way of re-query (all records each query, and then take some of the records, this way with the most), Lucene official explanation: Because we are fast enough. memory is prone to memory overflow when processing massive amounts of data.

lucene3.5 later provide a searchafter, this is in the large amount of data use (billion-level data volume), relatively slow, like Google search images, click More, and then out of a batch. The way to do this is to keep the data in the cache. And then go fetch it.
The following is the query part of the code:

/** * This is to query all the data first, then go to paging data     * Note that this way of handling massive amounts of data, easy memory Overflow * @param query * @param pageindex--page * @param pagesize--how much data per page */public void Searchpage (String query,int pageindex,int pageSize) {try {Directory dir = fileindexutils.getdirectory (); in Dexsearcher searcher = Getsearcher (dir); Queryparser parser = new Queryparser (version.lucene_35, "content", new StandardAnalyzer (version.lucene_35)); Query q = parser.parse (query); Topdocs TDS = Searcher.search (q, 500);//note here put 500 data in memory. scoredoc[] SDS = tds.scoredocs;int start = (pageIndex-1) *pagesize;int end = pageindex*pagesize;for (int i=start;i<end; i++) {Document doc = Searcher.doc (sds[i].doc); System.out.println (sds[i].doc+ ":" +doc.get ("path") + "--" +doc.get ("filename");} Searcher.close ();} catch (Org.apache.lucene.queryParser.ParseException e) {e.printstacktrace ();} catch (IOException e) { E.printstacktrace ();}}

Finally, let's take a look at paging using Searcherafter, the code is as follows (This method is not supported before lucene3.5 ):

/** * gets last Scoredocs * @param pageIndex * @param pageSize * @param query * @param searcher * @return * @thro based on page number and paging size WS IOException */private scoredoc getlastscoredoc (int pageindex,int pagesize,query Query,indexsearcher searcher) throws IOException {if (pageindex==1) return null;//If the first page returns an empty int num = pagesize* (pageIndex-1);//Gets the last number of previous page Topdocs TDS = Searcher.search (query, num); return tds.scoredocs[num-1];} public void Searchpagebyafter (String query,int pageindex,int pageSize) {try {Directory dir = fileindexutils.getdirectory (); Indexsearcher searcher = Getsearcher (dir); Queryparser parser = new Queryparser (version.lucene_35, "content", new StandardAnalyzer (version.lucene_35)); Query q = parser.parse (query);//Gets the last element of the previous page Scoredoc LASTSD = Getlastscoredoc (PageIndex, PageSize, Q, searcher);// Go through the last element to search for the next page of elements topdocs tds = Searcher.searchafter (Lastsd,q, pageSize); for (Scoredoc Sd:tds.scoreDocs) {Document doc = Searcher.doc (Sd.doc); System.out.println (sd.doc+ ":" +doc.get ("path") + "-" +doc.get ("FileName ")); Searcher.close ();} catch (Org.apache.lucene.queryParser.ParseException e) {e.printstacktrace ();} catch (IOException e) { E.printstacktrace ();}}

Two kinds of paging operations for Lucene