mirror of https://github.com/gnosygnu/xowa
gnosygnu
7 years ago
parent
ae9d0fccd3
commit
77de7215ce
@ -0,0 +1,34 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.specials; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.langs.mustaches.*;
|
||||
public class Xofulltext_indexer_doc implements Mustache_doc_itm {
|
||||
private final byte[] wikis_bry;
|
||||
public Xofulltext_indexer_doc
|
||||
( byte[] wikis_bry) {
|
||||
this.wikis_bry = wikis_bry;
|
||||
}
|
||||
public boolean Mustache__write(String key, Mustache_bfr bfr) {
|
||||
if (String_.Eq(key, "wikis"))
|
||||
bfr.Add_bry(wikis_bry);
|
||||
else
|
||||
return false;
|
||||
return true;
|
||||
}
|
||||
public Mustache_doc_itm[] Mustache__subs(String key) {
|
||||
return Mustache_doc_itm_.Ary__empty;
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,50 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.specials; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.xowa.specials.*; import gplx.langs.mustaches.*; import gplx.xowa.wikis.pages.*; import gplx.xowa.wikis.pages.tags.*;
|
||||
import gplx.dbs.*;
|
||||
class Xofulltext_indexer_html extends Xow_special_wtr__base {
|
||||
private final byte[] wikis_bry;
|
||||
public Xofulltext_indexer_html
|
||||
( byte[] wikis_bry) {
|
||||
this.wikis_bry = wikis_bry;
|
||||
}
|
||||
@Override protected Io_url Get_addon_dir(Xoa_app app) {return Addon_dir(app);}
|
||||
@Override protected Io_url Get_mustache_fil(Io_url addon_dir) {return addon_dir.GenSubFil_nest("bin", "xofulltext_indexer.template.html");}
|
||||
@Override protected Mustache_doc_itm Bld_mustache_root(Xoa_app app) {
|
||||
return new Xofulltext_indexer_doc(wikis_bry);
|
||||
}
|
||||
@Override protected void Bld_tags(Xoa_app app, Io_url addon_dir, Xopage_html_data page_data) {
|
||||
Xopg_tag_mgr head_tags = page_data.Head_tags();
|
||||
Xopg_tag_wtr_.Add__xoelem (head_tags, app.Fsys_mgr().Http_root());
|
||||
|
||||
Xopg_tag_wtr_.Add__xocss (head_tags, app.Fsys_mgr().Http_root());
|
||||
Xopg_tag_wtr_.Add__xohelp (head_tags, app.Fsys_mgr().Http_root());
|
||||
Xopg_tag_wtr_.Add__xolog (head_tags, app.Fsys_mgr().Http_root());
|
||||
Xopg_tag_wtr_.Add__xoajax (head_tags, app.Fsys_mgr().Http_root(), app);
|
||||
Xopg_tag_wtr_.Add__jquery (head_tags, app.Fsys_mgr().Http_root());
|
||||
Xopg_tag_wtr_.Add__xonotify (head_tags, app.Fsys_mgr().Http_root());
|
||||
Xopg_alertify_.Add_tags (head_tags, app.Fsys_mgr().Http_root());
|
||||
|
||||
head_tags.Add(Xopg_tag_itm.New_css_file(addon_dir.GenSubFil_nest("bin", "xofulltext_indexer.css")));
|
||||
head_tags.Add(Xopg_tag_itm.New_js_file(addon_dir.GenSubFil_nest("bin", "xofulltext_indexer.js")));
|
||||
|
||||
page_data.Js_enabled_y_();
|
||||
}
|
||||
public static Io_url Addon_dir(Xoa_app app) {
|
||||
return app.Fsys_mgr().Http_root().GenSubDir_nest("bin", "any", "xowa", "addon", "wiki", "fulltext", "indexer");
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,35 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.specials; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.xowa.specials.*; import gplx.core.net.qargs.*;
|
||||
import gplx.xowa.addons.apps.cfgs.*;
|
||||
public class Xofulltext_indexer_special implements Xow_special_page {
|
||||
public void Special__gen(Xow_wiki wiki, Xoa_page page, Xoa_url url, Xoa_ttl ttl) {
|
||||
// get qry if any
|
||||
Gfo_qarg_mgr url_args = new Gfo_qarg_mgr().Init(url.Qargs_ary());
|
||||
byte[] wikis_bry = url_args.Read_bry_or("wikis", Bry_.Empty);
|
||||
|
||||
// get options and create page
|
||||
// Xocfg_mgr cfg_mgr = wiki.App().Cfg();
|
||||
new Xofulltext_indexer_html
|
||||
( wikis_bry
|
||||
).Bld_page_by_mustache(wiki.App(), page, this);
|
||||
}
|
||||
Xofulltext_indexer_special(Xow_special_meta special__meta) {this.special__meta = special__meta;}
|
||||
public Xow_special_meta Special__meta() {return special__meta;} private final Xow_special_meta special__meta;
|
||||
public Xow_special_page Special__clone() {return this;}
|
||||
public static final Xow_special_page Prototype = new Xofulltext_indexer_special(Xow_special_meta.New_xo("XowaSearchBuilder", "Indexer"));
|
||||
}
|
||||
@ -0,0 +1,41 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.svcs; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.langs.jsons.*;
|
||||
import gplx.xowa.htmls.bridges.*;
|
||||
public class Xofulltext_indexer_bridge implements Bridge_cmd_itm {
|
||||
private Xofulltext_indexer_svc svc;
|
||||
public void Init_by_app(Xoa_app app) {
|
||||
this.svc = new Xofulltext_indexer_svc(app);
|
||||
}
|
||||
public String Exec(Json_nde data) {
|
||||
byte proc_id = proc_hash.Get_as_byte_or(data.Get_as_bry_or(Bridge_cmd_mgr.Msg__proc, null), Byte_ascii.Max_7_bit);
|
||||
Json_nde args = data.Get_kv(Bridge_cmd_mgr.Msg__args).Val_as_nde();
|
||||
switch (proc_id) {
|
||||
case Proc__index: svc.Index(args); break;
|
||||
default: throw Err_.new_unhandled_default(proc_id);
|
||||
}
|
||||
return "";
|
||||
}
|
||||
|
||||
private static final byte Proc__index = 0;
|
||||
private static final Hash_adp_bry proc_hash = Hash_adp_bry.cs()
|
||||
.Add_str_byte("index" , Proc__index)
|
||||
;
|
||||
|
||||
public byte[] Key() {return BRIDGE_KEY;} public static final byte[] BRIDGE_KEY = Bry_.new_a7("xowa.wiki.fulltext.indexer");
|
||||
public static final Xofulltext_indexer_bridge Prototype = new Xofulltext_indexer_bridge(); Xofulltext_indexer_bridge() {}
|
||||
}
|
||||
@ -0,0 +1,80 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.svcs; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.core.btries.*;
|
||||
import gplx.langs.jsons.*;
|
||||
import gplx.dbs.*; import gplx.xowa.wikis.data.tbls.*;
|
||||
import gplx.xowa.guis.cbks.*;
|
||||
import gplx.xowa.addons.apps.cfgs.*;
|
||||
import gplx.xowa.addons.wikis.fulltexts.indexers.specials.*;
|
||||
import gplx.xowa.addons.wikis.fulltexts.indexers.bldrs.*;
|
||||
class Xofulltext_indexer_svc implements Gfo_invk {
|
||||
private final Xoa_app app;
|
||||
private final Xog_cbk_trg cbk_trg = Xog_cbk_trg.New(Xofulltext_indexer_special.Prototype.Special__meta().Ttl_bry());
|
||||
public Xofulltext_indexer_svc(Xoa_app app) {
|
||||
this.app = app;
|
||||
}
|
||||
public void Index(Json_nde args) {
|
||||
// create args
|
||||
byte[] wikis_bry = args.Get_as_bry("wikis");
|
||||
Xofulltext_indexer_args indexer_args = new Xofulltext_indexer_args(wikis_bry);
|
||||
|
||||
// launch thread
|
||||
gplx.core.threads.Thread_adp_.Start_by_val("index", Cancelable_.Never, this, Invk__index, indexer_args);
|
||||
}
|
||||
private void Index(Xofulltext_indexer_args args) {
|
||||
// loop wikis
|
||||
byte[][] domain_ary = Bry_split_.Split(args.wikis, Byte_ascii.Pipe);
|
||||
for (byte[] domain : domain_ary) {
|
||||
// get wiki
|
||||
Xow_wiki wiki = app.Wiki_mgri().Get_by_or_make_init_n(domain);
|
||||
if (!Io_mgr.Instance.ExistsDir(wiki.Fsys_mgr().Root_dir())) {
|
||||
app.Gui__cbk_mgr().Send_json(cbk_trg, "xo.fulltext_indexer.status__note__recv", gplx.core.gfobjs.Gfobj_nde.New()
|
||||
.Add_str("note", Datetime_now.Get().XtoStr_fmt_yyyy_MM_dd_HH_mm_ss() + ": wiki does not exist: " + String_.new_u8(domain)));
|
||||
continue;
|
||||
}
|
||||
|
||||
wiki.Init_by_wiki();
|
||||
Io_url search_dir = wiki.Fsys_mgr().Root_dir().GenSubDir_nest("data", "search");
|
||||
if (Io_mgr.Instance.ExistsDir(search_dir)) {
|
||||
app.Gui__cbk_mgr().Send_json(cbk_trg, "xo.fulltext_indexer.status__note__recv", gplx.core.gfobjs.Gfobj_nde.New()
|
||||
.Add_str("note", Datetime_now.Get().XtoStr_fmt_yyyy_MM_dd_HH_mm_ss() + ": search dir already exists; please delete it manually before reindexing; " + String_.new_u8(domain)));
|
||||
continue;
|
||||
}
|
||||
|
||||
app.Gui__cbk_mgr().Send_json(cbk_trg, "xo.fulltext_indexer.status__note__recv", gplx.core.gfobjs.Gfobj_nde.New()
|
||||
.Add_str("note", Datetime_now.Get().XtoStr_fmt_yyyy_MM_dd_HH_mm_ss() + ": wiki index started: " + String_.new_u8(domain)));
|
||||
|
||||
new Xofulltext_indexer_mgr().Exec((Xowe_wiki)wiki, new Xofulltext_indexer_ui(app.Gui__cbk_mgr(), cbk_trg));
|
||||
|
||||
app.Gui__cbk_mgr().Send_json(cbk_trg, "xo.fulltext_indexer.status__note__recv", gplx.core.gfobjs.Gfobj_nde.New()
|
||||
.Add_str("note", Datetime_now.Get().XtoStr_fmt_yyyy_MM_dd_HH_mm_ss() + ": wiki index ended: " + String_.new_u8(domain)));
|
||||
}
|
||||
}
|
||||
|
||||
public Object Invk(GfsCtx ctx, int ikey, String k, GfoMsg m) {
|
||||
if (ctx.Match(k, Invk__index)) this.Index((Xofulltext_indexer_args)m.ReadObj("v"));
|
||||
else return Gfo_invk_.Rv_unhandled;
|
||||
return this;
|
||||
}
|
||||
private static final String Invk__index = "index";
|
||||
}
|
||||
class Xofulltext_indexer_args {
|
||||
public byte[] wikis;
|
||||
public Xofulltext_indexer_args(byte[] wikis) {
|
||||
this.wikis = wikis;
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,30 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.indexers.svcs; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.indexers.*;
|
||||
import gplx.xowa.guis.cbks.*;
|
||||
public class Xofulltext_indexer_ui {
|
||||
private final Xog_cbk_mgr cbk_mgr;
|
||||
private final Xog_cbk_trg cbk_trg;
|
||||
public Xofulltext_indexer_ui(Xog_cbk_mgr cbk_mgr, Xog_cbk_trg cbk_trg) {
|
||||
this.cbk_mgr = cbk_mgr;
|
||||
this.cbk_trg = cbk_trg;
|
||||
}
|
||||
public void Send_prog(String prog) {
|
||||
cbk_mgr.Send_json(cbk_trg, "xo.fulltext_indexer.status__prog__recv", gplx.core.gfobjs.Gfobj_nde.New()
|
||||
.Add_str("prog", prog)
|
||||
);
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,20 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.xowa.addons.wikis.fulltexts.searchers.mgrs; import gplx.*; import gplx.xowa.*; import gplx.xowa.addons.*; import gplx.xowa.addons.wikis.*; import gplx.xowa.addons.wikis.fulltexts.*; import gplx.xowa.addons.wikis.fulltexts.searchers.*;
|
||||
import gplx.xowa.addons.wikis.fulltexts.searchers.mgrs.uis.*;
|
||||
public interface Xofulltext_searcher {
|
||||
void Search(Xofulltext_searcher_ui ui, Xow_wiki wiki, Xofulltext_searcher_args args);
|
||||
}
|
||||
@ -0,0 +1,91 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.gflucene; import gplx.*;
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
|
||||
import org.apache.lucene.analysis.standard.StandardAnalyzer;
|
||||
import org.apache.lucene.document.*;
|
||||
import org.apache.lucene.index.IndexOptions;
|
||||
import org.apache.lucene.index.IndexWriter;
|
||||
import org.apache.lucene.index.IndexWriterConfig;
|
||||
import org.apache.lucene.store.Directory;
|
||||
import org.apache.lucene.store.FSDirectory;
|
||||
public class Gflucene_index_bldr {
|
||||
private final StandardAnalyzer analyzer = new StandardAnalyzer();
|
||||
private final IndexWriterConfig config;
|
||||
private Directory index;
|
||||
private IndexWriter wtr;
|
||||
private FieldType body_fld;
|
||||
|
||||
public Gflucene_index_bldr() {
|
||||
this.config = new IndexWriterConfig(analyzer);
|
||||
}
|
||||
|
||||
public void Init(String index_dir) {
|
||||
|
||||
// create index
|
||||
Path path = Paths.get(index_dir);
|
||||
try {
|
||||
this.index = FSDirectory.open(path);
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to open lucene index", "path", path);
|
||||
}
|
||||
|
||||
// create writer
|
||||
try {
|
||||
wtr = new IndexWriter(index, config);
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to create writer");
|
||||
}
|
||||
|
||||
// create field for body
|
||||
this.body_fld = new FieldType();
|
||||
body_fld.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS);
|
||||
// body_fld.setIndexOptions(IndexOptions.DOCS_AND_FREQS_AND_POSITIONS_AND_OFFSETS);
|
||||
// body_fld.setStored(true);
|
||||
body_fld.setTokenized(true);
|
||||
// body_fld.setStoreTermVectors(true);
|
||||
// body_fld.setStoreTermVectorOffsets(true);
|
||||
}
|
||||
public void Exec(Gflucene_index_data data) {
|
||||
// org.apache.lucene.document.
|
||||
Document doc = new Document();
|
||||
// doc.add(new SortedNumericDocValuesField("page_score", data.score));
|
||||
doc.add(new StoredField("page_score", data.score));
|
||||
doc.add(new StoredField("page_id", data.page_id));
|
||||
doc.add(new TextField("title", data.title, Field.Store.YES));
|
||||
doc.add(new Field("body", data.body, body_fld));
|
||||
try {
|
||||
wtr.addDocument(doc);
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to add document", "title", data.title);
|
||||
}
|
||||
}
|
||||
public void Term() {
|
||||
try {
|
||||
wtr.close();
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to close writer");
|
||||
}
|
||||
try {
|
||||
index.close();
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to close writer");
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,73 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.gflucene; import gplx.*;
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
|
||||
import org.apache.lucene.analysis.standard.StandardAnalyzer;
|
||||
import org.apache.lucene.document.Document;
|
||||
import org.apache.lucene.index.DirectoryReader;
|
||||
import org.apache.lucene.index.IndexReader;
|
||||
import org.apache.lucene.queryparser.classic.QueryParser;
|
||||
import org.apache.lucene.search.IndexSearcher;
|
||||
import org.apache.lucene.search.Query;
|
||||
import org.apache.lucene.search.ScoreDoc;
|
||||
import org.apache.lucene.search.TopDocs;
|
||||
import org.apache.lucene.store.Directory;
|
||||
import org.apache.lucene.store.FSDirectory;
|
||||
public class Gflucene_searcher {
|
||||
private final StandardAnalyzer analyzer = new StandardAnalyzer();
|
||||
private Directory index;
|
||||
|
||||
public Gflucene_searcher() {
|
||||
}
|
||||
|
||||
public void Init(String index_dir) {
|
||||
Path path = Paths.get(index_dir);
|
||||
try {
|
||||
this.index = FSDirectory.open(path);
|
||||
} catch (IOException e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to init searcher", "dir", index_dir);
|
||||
}
|
||||
}
|
||||
public void Exec(List_adp list, Gflucene_searcher_data data) {
|
||||
try {
|
||||
IndexReader reader = DirectoryReader.open(index);
|
||||
IndexSearcher searcher = new IndexSearcher(reader);
|
||||
|
||||
|
||||
Query query = new QueryParser("body", analyzer).parse(data.query);
|
||||
// TopDocs docs = searcher.search(query, reader.maxDoc());
|
||||
TopDocs docs = searcher.search(query, data.match_max);
|
||||
ScoreDoc[] hits = docs.scoreDocs;
|
||||
|
||||
for(int i = 0; i < hits.length; i++) {
|
||||
int docId = hits[i].doc;
|
||||
Document d = searcher.doc(docId);
|
||||
Gflucene_index_data doc = new Gflucene_index_data(Integer.parseInt(d.get("page_id")), Integer.parseInt(d.get("page_score")), d.get("title"), "");
|
||||
doc.lucene_score = hits[i].score;
|
||||
list.Add(doc);
|
||||
}
|
||||
|
||||
reader.close();
|
||||
} catch (Exception e) {
|
||||
throw Err_.new_exc(e, "lucene_index", "failed to exec seearch", "query", data.query);
|
||||
}
|
||||
}
|
||||
public void Term() {
|
||||
}
|
||||
}
|
||||
@ -0,0 +1,24 @@
|
||||
/*
|
||||
XOWA: the XOWA Offline Wiki Application
|
||||
Copyright (C) 2012-2017 gnosygnu@gmail.com
|
||||
|
||||
XOWA is licensed under the terms of the General Public License (GPL) Version 3,
|
||||
or alternatively under the terms of the Apache License Version 2.0.
|
||||
|
||||
You may use XOWA according to either of these licenses as is most appropriate
|
||||
for your project on a case-by-case basis.
|
||||
|
||||
The terms of each license can be found in the source code repository:
|
||||
|
||||
GPLv3 License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-GPLv3.txt
|
||||
Apache License: https://github.com/gnosygnu/xowa/blob/master/LICENSE-APACHE2.txt
|
||||
*/
|
||||
package gplx.gflucene; import gplx.*;
|
||||
public class Gflucene_searcher_data {
|
||||
public String query;
|
||||
public int match_max;
|
||||
public Gflucene_searcher_data(String query, int match_max) {
|
||||
this.query = query;
|
||||
this.match_max = match_max;
|
||||
}
|
||||
}
|
||||
Loading…
Reference in new issue