Marown/yacy_search_server
1
0
Fork 0
mirror of https://github.com/yacy/yacy_search_server.git synced 2024-09-19 00:01:41 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity

Properly support multiple Boost Queries. (Previous code was broken because it concatenated multiple Boost Queries together rather than passing Solr an array.)

Browse Source
This commit is contained in:
JeremyRand 2016-05-17 21:58:26 -05:00
parent 58824dfa6c
commit 433217b33e
5 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
defaults/yacy.init
View File

@ -1036,7 +1036,7 @@ search.ranking.rwi.profile =
search.ranking.solr.collection.boostname.tmpa.0=Default Profile search.ranking.solr.collection.boostname.tmpa.0=Default Profile
search.ranking.solr.collection.boostfields.tmpa.0=url_paths_sxt^3.0,synonyms_sxt^0.5,title^5.0,text_t^1.0,host_s^6.0,h1_txt^5.0,url_file_name_tokens_t^4.0,h2_txt^3.0,keywords^2.0,description_txt^1.5,author^1.0 search.ranking.solr.collection.boostfields.tmpa.0=url_paths_sxt^3.0,synonyms_sxt^0.5,title^5.0,text_t^1.0,host_s^6.0,h1_txt^5.0,url_file_name_tokens_t^4.0,h2_txt^3.0,keywords^2.0,description_txt^1.5,author^1.0
search.ranking.solr.collection.filterquery.tmpa.0= search.ranking.solr.collection.filterquery.tmpa.0=
search.ranking.solr.collection.boostquery.tmpa.0=crawldepth_i:0^0.8 crawldepth_i:1^0.4 search.ranking.solr.collection.boostquery.tmpa.0=crawldepth_i:0^0.8\ncrawldepth_i:1^0.4
search.ranking.solr.collection.boostfunction.tmpb.0= search.ranking.solr.collection.boostfunction.tmpb.0=
search.ranking.solr.collection.boostname.tmpa.1=Date Profile: sort by date in descending order for a '/date' usage search.ranking.solr.collection.boostname.tmpa.1=Date Profile: sort by date in descending order for a '/date' usage
search.ranking.solr.collection.boostfields.tmpa.1=url_paths_sxt^0.1,title^0.1,text_t^0.1 search.ranking.solr.collection.boostfields.tmpa.1=url_paths_sxt^0.1,title^0.1,text_t^0.1
@ -1051,7 +1051,7 @@ search.ranking.solr.collection.boostfunction.tmpb.2=
search.ranking.solr.collection.boostname.tmpa.3=_unused3 search.ranking.solr.collection.boostname.tmpa.3=_unused3
search.ranking.solr.collection.boostfields.tmpa.3=text_t^1.0 search.ranking.solr.collection.boostfields.tmpa.3=text_t^1.0
search.ranking.solr.collection.filterquery.tmpa.3= search.ranking.solr.collection.filterquery.tmpa.3=
search.ranking.solr.collection.boostquery.tmpa.3=crawldepth_i:0^0.8 crawldepth_i:1^0.4 search.ranking.solr.collection.boostquery.tmpa.3=crawldepth_i:0^0.8\ncrawldepth_i:1^0.4
search.ranking.solr.collection.boostfunction.tmpb.3= search.ranking.solr.collection.boostfunction.tmpb.3=
# the following values are used to identify duplicate content # the following values are used to identify duplicate content

2
htroot/RankingSolr_p.html
View File

@ -50,7 +50,7 @@
<dl> <dl>
<dt style="width:260px;margin:0;padding:0;height:1.8em;"><label for="bq" id="bq_label">bq=</label></dt> <dt style="width:260px;margin:0;padding:0;height:1.8em;"><label for="bq" id="bq_label">bq=</label></dt>
<dd style="width:360px;margin:0;padding:0;float:left;display:inline;" id="bq_dd"> <dd style="width:360px;margin:0;padding:0;float:left;display:inline;" id="bq_dd">
<textarea name="bq" id="bq" align="left" cols="96" rows="1"/>#[bq]#</textarea> <textarea name="bq" id="bq" align="left" cols="96" rows="5"/>#[bq]#</textarea>
</dd> </dd>
<dt style="width:260px;margin:0;padding:0;height:1.8em;"></dt> <dt style="width:260px;margin:0;padding:0;height:1.8em;"></dt>
<dd style="width:360px;margin:0;padding:0;height:1.8em;float:left;display:inline;"> <dd style="width:360px;margin:0;padding:0;height:1.8em;float:left;display:inline;">

2
htroot/RankingSolr_p.java
View File

@ -97,7 +97,7 @@ public class RankingSolr_p {
} }
} }
if (post != null && post.containsKey("ResetBQ")) { if (post != null && post.containsKey("ResetBQ")) {
String bq = "crawldepth_i:0^0.8 crawldepth_i:1^0.4"; String bq = "crawldepth_i:0^0.8\ncrawldepth_i:1^0.4";
if (bq != null) { if (bq != null) {
sb.setConfig(SwitchboardConstants.SEARCH_RANKING_SOLR_COLLECTION_BOOSTQUERY_ + profileNr, bq); sb.setConfig(SwitchboardConstants.SEARCH_RANKING_SOLR_COLLECTION_BOOSTQUERY_ + profileNr, bq);
sb.index.fulltext().getDefaultConfiguration().getRanking(profileNr).setBoostQuery(bq); sb.index.fulltext().getDefaultConfiguration().getRanking(profileNr).setBoostQuery(bq);

1
source/net/yacy/search/Switchboard.java
View File

@ -537,6 +537,7 @@ public final class Switchboard extends serverSwitch {
bf.equals("scale(cr_host_norm_i,1,20)")) bf = ""; bf.equals("scale(cr_host_norm_i,1,20)")) bf = "";
if (bf.equals("recip(rord(last_modified),1,1000,1000))")) bf = "recip(ms(NOW,last_modified),3.16e-11,1,1)"; // that was an outdated date boost that did not work well if (bf.equals("recip(rord(last_modified),1,1000,1000))")) bf = "recip(ms(NOW,last_modified),3.16e-11,1,1)"; // that was an outdated date boost that did not work well
if (i == 0 && bq.equals("fuzzy_signature_unique_b:true^100000.0")) bq = "crawldepth_i:0^0.8 crawldepth_i:1^0.4"; if (i == 0 && bq.equals("fuzzy_signature_unique_b:true^100000.0")) bq = "crawldepth_i:0^0.8 crawldepth_i:1^0.4";
if (bq.equals("crawldepth_i:0^0.8 crawldepth_i:1^0.4")) bq = "crawldepth_i:0^0.8\ncrawldepth_i:1^0.4"; // Fix issue with multiple Boost Queries
if (boosts.equals("url_paths_sxt^1000.0,synonyms_sxt^1.0,title^10000.0,text_t^2.0,h1_txt^1000.0,h2_txt^100.0,host_organization_s^100000.0")) boosts = "url_paths_sxt^3.0,synonyms_sxt^0.5,title^5.0,text_t^1.0,host_s^6.0,h1_txt^5.0,url_file_name_tokens_t^4.0,h2_txt^2.0"; if (boosts.equals("url_paths_sxt^1000.0,synonyms_sxt^1.0,title^10000.0,text_t^2.0,h1_txt^1000.0,h2_txt^100.0,host_organization_s^100000.0")) boosts = "url_paths_sxt^3.0,synonyms_sxt^0.5,title^5.0,text_t^1.0,host_s^6.0,h1_txt^5.0,url_file_name_tokens_t^4.0,h2_txt^2.0";
r.setName(name); r.setName(name);
r.updateBoosts(boosts); r.updateBoosts(boosts);

4
source/net/yacy/search/query/QueryParams.java
View File

@ -386,7 +386,7 @@ public final class QueryParams {
if (!qf.isEmpty()) params.setParam(DisMaxParams.QF, qf); if (!qf.isEmpty()) params.setParam(DisMaxParams.QF, qf);
if (this.queryGoal.getIncludeSize() > 1) { if (this.queryGoal.getIncludeSize() > 1) {
// add boost on combined words // add boost on combined words
if (bq.length() > 0) bq += " "; if (bq.length() > 0) bq += "\n";
bq += CollectionSchema.text_t.getSolrFieldName() + ":\"" + this.queryGoal.getIncludeString() + "\"^10"; bq += CollectionSchema.text_t.getSolrFieldName() + ":\"" + this.queryGoal.getIncludeString() + "\"^10";
} }
if (fq.length() > 0) { if (fq.length() > 0) {
@ -396,7 +396,7 @@ public final class QueryParams {
newfq.add(fq); newfq.add(fq);
params.setFilterQueries(newfq.toArray(new String[newfq.size()])); params.setFilterQueries(newfq.toArray(new String[newfq.size()]));
} }
if (bq.length() > 0) params.setParam(DisMaxParams.BQ, bq); if (bq.length() > 0) params.setParam(DisMaxParams.BQ, bq.split("[\\r\\n]+")); // split on any sequence consisting of CR and/or LF
if (bf.length() > 0) params.setParam("boost", bf); // a boost function extension, see http://wiki.apache.org/solr/ExtendedDisMax#bf_.28Boost_Function.2C_additive.29 if (bf.length() > 0) params.setParam("boost", bf); // a boost function extension, see http://wiki.apache.org/solr/ExtendedDisMax#bf_.28Boost_Function.2C_additive.29
// prepare result // prepare result