<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
< html xmlns = "http://www.w3.org/1999/xhtml" >
< head >
< title > YaCy '#[clientname]#': Index Sources & Targets< / title >
#%env/templates/metas.template%#
< / head >
< body id = "IndexFederated_p" >
#%env/templates/header.template%#
#%env/templates/submenuIndexControl.template%#
< h2 > Index Sources & Targets< / h2 >
< p >
YaCy supports multiple index storage locations.
As an internal indexing database a deep-embedded multi-core Solr is used and it is possible to attach also a remote Solr.
< / p >
< form id = "config" action = "IndexFederated_p.html" method = "post" enctype = "multipart/form-data" accept-charset = "UTF-8" >
< fieldset >
< legend > Solr Search Index< / legend >
Solr stores the main search index. It is the home of two cores, the default 'collection1' core for documents and the 'webgraph' core for a web structure graph. Detailed information about the used Solr fields can be edited in the < a href = "IndexSchema_p.html" > Schema Editor< / a > .
< dl >
< dt class = "TableCellDark" > Lazy Value Initialization < input type = "checkbox" name = "solr.indexing.lazy" id = "solr.indexing.lazy" # ( solr . indexing . lazy . checked ) # :: checked = "checked" # ( / solr . indexing . lazy . checked ) # / > < / dt >
< dd > If checked, only non-zero values and non-empty strings are written to Solr fields.< / dd >
< dt > < / dt > < dd > < / dd >
< dt > Use deep-embedded local Solr < input type = "checkbox" name = "core.service.fulltext" id = "core_service_fulltext" # ( core . service . fulltext . checked ) # :: checked = "checked" # ( / core . service . fulltext . checked ) # onclick = "if(!document.getElementById('config').core_service_fulltext.checked) {document.getElementById('config').solr_indexing_solrremote.checked = true;}" / > < / dt >
< dd >
This will write the YaCy-embedded Solr index which stored within the YaCy DATA directory.< br / >
The Solr native search interface is accessible at< br / >
< a href = "solr/select?q=*:*&start=0&rows=3&core=collection1" > /solr/select?q=*:*& start=0& rows=3& core=collection1< / a >
for the default search index (core: collection1) and at< br / >
< a href = "solr/select?q=*:*&start=0&rows=3&core=webgraph" > /solr/select?q=*:*& start=0& rows=3& core=webgraph< / a > for the webgraph core.< br / >
If you switch off this index, a remote Solr must be activated.< / dd >
< dt > Use remote Solr server(s) < input type = "checkbox" name = "solr.indexing.solrremote" id = "solr_indexing_solrremote" # ( solr . indexing . solrremote . checked ) # :: checked = "checked" # ( / solr . indexing . solrremote . checked ) # onclick = "if(!document.getElementById('config').solr_indexing_solrremote.checked) {document.getElementById('config').core_service_fulltext.checked = true;}" / > < / dt >
< dd > It's easy to < a href = "http://www.yacy-websearch.net/wiki/index.php/Dev:Solr" target = "_blank" > attach an external Solr to YaCy< / a > .
This external Solr can be used instead the internal Solr. It can also be used additionally to the internal Solr, then both Solr indexes are mirrored.
< / dd >
#(table)#::
< dt class = "TableCellDark" > < / dt >
< dd > Solr Hosts< br / >
< div >
< table class = "sortable" border = "0" >
< tr class = "TableHeader" valign = "bottom" >
< td > < strong > Solr Host Administration Interface< / strong > < br / > < / td >
< td > < strong > Index Size< / strong > < / td >
< / tr >
#{list}#
< tr class = "TableCell#(dark)#Light::Dark::Summary#(/dark)#" >
< td > < a href = "#[url]#" target = "_blank" > #[url]#< / a > < / td >
< td align = "right" > #[size]#< / td >
< / tr >
#{/list}#
< / table >
< / div >
< / dd >
#(/table)#
< dt class = "TableCellDark" > < / dt >
< dd > Solr URL(s)< br / > < textarea rows = "2" cols = "80" name = "solr.indexing.url" id = "solr.indexing.url" / > #[solr.indexing.url]#< / textarea > < br / >
You can set one or more Solr targets here which are accessed as a shard. For several targets, list them using a ',' (comma) as separator.
The set of remote targets are used as shard of a complete index. The host part of the url is used as key for a hash function which selects one of the shards (one of your remote servers).
When a search request is made, all servers are accessed synchronously and the result is combined.< / dd >
< dt class = "TableCellDark" > < / dt >
< dd > Sharding Method< br / >
< select id = "solr.indexing.sharding" name = "solr.indexing.sharding" >
#{solr.indexing.sharding.methods}#
< option value = "#[method]#" # ( selected ) # ::selected = "selected" # ( / selected ) # > #[method]#: #[description]#< / option >
#{/solr.indexing.sharding.methods}#
< / select > < / dd >
< dt class = "TableCellDark" > < / dt >
< dd > < input type = "checkbox" name = "solr.indexing.solrremote.writeenabled" id = "solr_indexing_solrremote_writeenabled" # ( solr . indexing . solrremote . writeenabled . checked ) # :: checked = "checked" # ( / solr . indexing . solrremote . writeenabled . checked ) # / > write-enabled (if unchecked, the remote server(s) will only be used as search peers)< / dd >
< dt > < input type = "submit" name = "setsolr" value = "Set" / > < / dt > < dd > < / dd >
< / dd >
< / dl >
< / fieldset >
< / form >
< form id = "config" action = "IndexFederated_p.html" method = "get" enctype = "multipart/form-data" accept-charset = "UTF-8" >
< fieldset >
< legend >
Web Structure Index
< / legend >
The web structure index is used for host browsing (to discover the internal file/folder structure), ranking (counting the number of references) and file search (there are about fourty times more links from loaded pages as in documents of the main search index).
< dl >
< dt > < input type = "checkbox" name = "core.service.citation.tmp" id = "core_service_citation" # ( core . service . citation . tmp . checked ) # :: checked = "checked" # ( / core . service . citation . tmp . checked ) # / > < / dt >
< dd > use citation reference index (lightweight and fast)< / dd >
< dt > < input type = "checkbox" name = "core.service.webgraph.tmp" id = "core_service_webgraph" # ( core . service . webgraph . tmp . checked ) # :: checked = "checked" # ( / core . service . webgraph . tmp . checked ) # / > < / dt >
< dd > use webgraph search index (rich information in second Solr core)< / dd >
< dt > < input type = "submit" name = "setcitation" value = "Set" / > < / dt > < dd > < / dd >
< / dl >
< / fieldset >
< / form >
< form id = "config" action = "IndexFederated_p.html" method = "get" enctype = "multipart/form-data" accept-charset = "UTF-8" >
< fieldset >
< legend >
Peer-to-Peer Operation
< / legend >
The 'RWI' (Reverse Word Index) is necessary for index transmission in distributed mode. For portal or intranet mode this must be switched off.
< dl >
< dt > < input type = "checkbox" name = "core.service.rwi" id = "core_service_rwi" # ( core . service . rwi . checked ) # :: checked = "checked" # ( / core . service . rwi . checked ) # / > < / dt >
< dd > support peer-to-peer index transmission (DHT RWI index)< / dd >
< dt > < input type = "submit" name = "setrwi" value = "Set" / > < / dt > < dd > < / dd >
< / dl >
< / fieldset >
< / form >
#%env/templates/footer.template%#
< / body >
< / html >
