Changes

This [[Module Library|XQuery Module]] extends the [http://www.w3.org/TR/xpath[Full-full-text-10 W3C Full Text Recommendation] with some useful functions] features of BaseX: The index can be directly accessed, full-text results can be marked with additional elements, or the relevant parts can be extracted. Moreover, the score value, which is generated by the <code>{{Code|contains text</code> }} expression, can be explicitly requested from items. All functions are introduced with the <code>ft:</code> prefix, which is linked to the statically declared <code>http://basex.org/modules/ft</code> namespace.

=Conventions= All functions and errors in this module are assigned to the <code><nowiki>http://basex.org/modules/ft</nowiki></code> namespace, which is statically bound to the {{Code|ft}} prefix.<br/> =Database Functions=

{{Mark|Updated with Version 7.2.2:}} {{Mono|$options}} parameter added, second argument generalized. {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|<code><bpre>ft:search</b>( $db as item(), $terms as item()*) as text()*</code><br/><b>ftxs:search</b>($db as item()string, $terms as item()*, $options as itemmap(*)? := map { }) as text()*</codepre>|-| valign='"top' "| '''Summary'''|Returns all text nodes from the full-text index of the database <code>[[Database Module#Database Argument{{Code|$db]]</code> }} that contain the specified {{MonoCode|$terms}}.<br/>The options used for tokenizing the input and building the full-text will also be applied to the search terms. As an example, if the index terms have been stemmed, the search string will be stemmed as well.The {{MonoCode|$options}} argument can be used to overwrite the default control full-text optionsprocessing. It can be specified as* {{Mono|element(full-text-options)}}: &lt;full-text-The following options/&gt;</code> must be used as root element, and the parameters are specified as child nodes, with supported (the element name representing the key and the text node representing the value:<br /><code>&lt;full-text-options&gt;<br/>&nbsp;&nbsp;&lt;key1&gt;value1&lt;/key1&gt;<br/>&nbsp;&nbsp;...<br/>&lt;/full-text-options&gt;</code>* introduction on [[Map Module|map structureFull-Text]]: all parameters can be directly represented as key/value pairs:<br /><code>map { "key1" := "value1", ... </code>}<br/>This variant is more compact, and it facilitates the binding of arbitrary data types. Note that the map structures are not part of processing gives you equivalent expressions in the XQuery language yet, as the standardization is still in progress.The following keys are supportedFull-Text notation):* {{MonoCode|mode}}: determines determine the search mode (also called [http://www.w3.org/TR/xpath-full-text-10/#ftwords AnyAllOption])how tokens are searched. Allowed values are {{MonoCode|any}}, {{MonoCode|any word}}, {{MonoCode|all}}, {{MonoCode|all words}}, and {{MonoCode|phrase}}. {{MonoCode|any}} is the default search mode.* {{MonoCode|fuzzywildcards}}: turns fuzzy turn wildcard querying on or off. Allowed values are an empty string or {{MonoCode|true}}, or and {{MonoCode|false}}. By default, fuzzy wildcard querying is turned off.* {{MonoCode|wildcardsfuzzy}}: turns wildcard turn fuzzy querying on or off. Allowed values are an empty string or {{MonoCode|true}}, or and {{MonoCode|false}}. By default, wildcard fuzzy querying is turned off.* {{Code|errors}}: control the maximum number of tolerated errors for fuzzy querying. By default, {{Code|0}} is assigned (see [[Full-Text#Fuzzy_Querying|Fuzzy Querying]] for more details).* {{Code|ordered}}: indicate if all tokens must occur in the order in which they are specified. Allowed values are {{Code|true}} and {{Code|false}}. The default is {{Code|false}}.* {{Code|content}}: specify that the matched tokens need to occur at the beginning or end of a searched string, or need to cover the entire string. Allowed values are {{Code|start}}, {{Code|end}}, and {{Code|entire}}. By default, the option is turned off.* {{Code| valign=scope}}: define the scope in which tokens must be located. The option has following sub options:** {{Code|same}}: can be set to {{Code|true}} or {{Code|false}}. It specifies if tokens need to occur in the same or different units.** {{Code|unit}}: can be {{Code|sentence}} or {{Code|paragraph}}. It specifies the unit for finding tokens.* {{Code|window}}: set up a window in which all tokens must be located. By default, the option is turned off. It has following sub options:** {{Code|size}}: specify the size of the window in terms of 'top' | units''.** {{Code|unit}}: can be {{Code|sentences}}, {{Code|sentences}} or {{Code|paragraphs}}. The default is {{Code|words}}.* {{Code|distance}}: specify the distance in which tokens must occur. By default, the option is turned off. It has following sub options:** {{Code|min}}: specify the minimum distance in terms of 'Errors'units''. The default is {{Code|0}}.** {{Code|max}}: specify the maximum distance in terms of '''[[XQuery Errors#BaseX Errors|BASX0001]]'units'' . The default is raised if the full-text index is not available{{Code|∞}}.** {{Code|unit}}: can be {{Code|words}}, {{Code|sentences}} or if the selected option {{Code|paragraphs}}. The default is not supported by the existing index{{Code|words}}.<br/>|- valign="top"| '''[[XQuery Errors#BaseX Errors|BASX0002]]''' is raised if a referenced node is |{{Error|db:get|Database Module#Errors}} The addressed database does not exist or could not stored in a database (i.e., references a main-memory XML fragment)be opened.<br/>'''[[XQuery Errors{{Error|db:no-index|Database Module#BaseX Errors|BASX0021]]''' is raised if }} the specified full-text option index is unknownnot available.<br/>'''[[XQuery Errors{{Error|options|#BaseX Errors|BASX0022]]''' is raised if both }} the fuzzy and wildcard querying has been selectedoption cannot be both specified.|-| valign='"top' "| '''Examples'''

* <code>{{Code|ft:search("DB", "QUERY")</code> returns }}: Return all text nodes of the database {{MonoCode|DB}} that contain the term {{MonoCode|QUERY}}.* Return all text nodes of the database {{Code|DB}} that contain the numbers {{Code|2010}} and {{Code|2020}}:<br/><code>ft:search("DB", ("2010",2011"2020"), map { 'mode':='all' })</code><br/>returns all * Return text nodes of that contain the database terms {{MonoCode|DBA}} that contain the numbers and {{MonoCode|B|2010}} and in a distance of at most 5 words:<pre lang='xquery'>ft:search("db", ("A", "B"), map { "mode": "all words", "distance": map {Mono|20111 "max": "5", "unit": "words" }}.)</pre>* The last example iterates Iterate over five three databases and returns return all elements containing terms similar to {{MonoCode|Hello World}} in the text nodes:<pre classlang="brush:'xquery"'>

return ft:search($dbname, $terms, map { 'fuzzy': $optionsfuzzy })/..

==ft:marktokens== {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|<code><bpre>ft:mark</b>tokens( $nodes db as node()*) as node()*</code><br /><code><b>ftxs:mark</b>($nodes as node()*string, $tag prefix as xs:string := ()) as nodeelement(value)*</codepre>|-| valign='"top' "| '''Summary'''|Puts a marker element around the resulting <code>$nodes</code> of a Returns all full-text tokens stored in the index request.<br />The default tag name of the marker element is <code>mark</code>. An alternative tag name can be chosen via the optional <code>database {{Code|$tag</code> argumentdb}}, along with their numbers of occurrences.<br />Note that If {{Code|$prefix}} is specified, the XML node returned nodes will be refined to be transformed must be an internal "database" nodethe strings starting with that prefix. The <code>transform</code> expression can prefix will be used tokenized according to apply the method to a mainfull-memory fragment (see example)text used for creating the index.|-| valign='"top' "| '''Errors'''|'''[[XQuery Errors{{Error|db:get|Database Module#BaseX Errors|BASX0002]]''' is raised if a referenced node is }} The addressed database does not exist or could not stored in a database (i.e., references a main-memory XML fragment)be opened.<br />'''[[XQuery Errors{{Error|db:no-index|Database Module#Functions Errors|FOCA0002]]''' }} the full-text index is raised if <code>$name</code> is no valid QNamenot available.|-| valign='"top' "| '''Examples'''|* The following query returns <code>&lt;XML&gt;&lt;mark&gt;hello&lt;/mark&gt; world&lt;/XML&gt;</code>Returns the number of occurrences for a single, if one text node of the database <code>DB</code> has the value "hello world"specific index entry:<pre classlang="brush:'xquery"'>let $term := ft:marktokenize(db$term)return number(ft:opentokens('DBdb', $term)//*[text() contains text 'hello'. = $term]/@count)

=General Functions= ==ft:extractcontains== {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|<code><bpre>ft:extract</b>contains( $nodes input as nodeitem()*) as node()*</code><br /><code><b>ft:extract</b>(, $nodes terms as nodeitem()*, $tag options as xs:string) as nodemap(*)*</code><br /><code><b>ft? :extract</b>($nodes as node(= map { })*, $tag as xs:string, $length as xs:integer) as node()*boolean</codepre>|-| valign='"top' "| '''Summary'''|Extracts and returns relevant parts of full-text results. It puts a marker element around Checks if the resulting <code>specified {{Code|$nodes</code> of a full-text index request and chops irrelevant sections of input}} items contain the resultspecified {{Code|$terms}}.<br />The default tag name of function does the same as the marker element is <code>mark</code>. An alternative tag name [[Full-Text]] expression {{Code|contains text}}, but options can be chosen via specified more dynamically. The {{Code|$options}} are the optional <code>$tag</code> argumentsame as for {{Function||ft:search}}, and the following ones exist:* {{Code|case}}: determines how character case is processed. Allowed values are {{Code|insensitive}}, {{Code|sensitive}}, {{Code|upper}} and {{Code|lower}}.<br />The By default length of the returned text , search is <code>150</code> case-insensitive.* {{Code|diacritics}}: determines how diacritical charactersare processed. Allowed values are {{Code|insensitive}} and {{Code|sensitive}}. An alternative length can be specified via the optional <code>$length</code> argumentBy default, search is diacritical insensitive.* {{Code|stemming}}: determines is tokens are stemmed. Note that the effective text length may differ from the specified text due to formatting Allowed values are {{Code|true}} and readibility issues{{Code|false}}. By default, stemming is turned off.* {{Code|-language}}: determines the language. This option is relevant for stemming tokens. All language codes are supported. The default language is {{Code|en}}.| - valign='"top' "| '''Errors'''|'''[[XQuery Errors#BaseX Errors{{Error|options|BASX0002]]''' is raised if a referenced node is not stored in a database (i.e., references a main-memory XML fragment).<br />'''[[XQuery Errors#Functions Errors|FOCA0002]]''' is raised if <code>$name</code> is no valid QName}} specified options are conflicting.|-| valign='"top' "| '''Examples'''

* The following query may return <code>&lt;XML&gt;...&lt;b&gt;hello&lt;/b&gt;...&lt;XML&gt;</code> Checks if a text node of {{Code|jack}} or {{Code|john}} occurs in the database input string {{Code|John Doe}}:<codepre lang='xquery'>DBft:contains("John Doe", ("jack", "john"), map { "mode": "any" })</codepre> contains * Calls the string "hello world"function with stemming turned on and off:<pre classlang="brush:'xquery"'>ft:extract(db:opentrue('DB')//*[text, false() ) ! ft:contains text ("Häuser", "Haus", map { 'hellostemming']: ., 'blanguage', 1:'de' })

 {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|<code><bpre>ft:count</b>( $nodes as node()*) as xs:integer</codepre>|-| valign='"top' "| '''Summary'''

|-| valign='"top' | '''Errors'''|'''[[XQuery Errors#BaseX Errors|BASX0002]]''' is raised if a referenced node is not stored in a database (i.e., references a main-memory XML fragment).|-"| valign='top' | '''Examples'''

* <code>{{Code|ft:count(//*[text() contains text 'QUERY'])</code> }} returns the <code>{{Code|xs:integer</code> }} value <code>{{Code|2</code> }} if a document contains two occurrences of the string "QUERY".

 {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|<code><bpre>ft:score</b>( $item as item()*) as xs:double*</codepre>|-| valign='"top' "| '''Summary'''|Returns the score values (0.0 - 1.0) that have been attached to the specified items. <code>{{Code|0</code> }} is returned a value if no score was attached.|-valign="top"| '''Examples'''| * {{Code|ft:score('a' contains text 'a')}} returns the {{Code|xs:double}} value {{Code|1}}.|} ==ft:tokenize== {| width='100%'|- valign="top"| width='120' | '''Signature'''|<pre>ft:tokenize( $string as xs:string?, $options as map(*)? := map { }) as xs:string*</pre>|- valign="top"| '''Summary''' |Tokenizes the given {{Code|$string}}, using the current default full-text options or the {{Code|$options}} specified as second argument, and returns a sequence with the tokenized string. The following options are available:* {{Code|case}}: determines how character case is processed. Allowed values are {{Code|insensitive}}, {{Code|sensitive}}, {{Code|upper}} and {{Code|lower}}. By default, search is case insensitive.* {{Code|diacritics}}: determines how diacritical characters are processed. Allowed values are {{Code|insensitive}} and {{Code|sensitive}}. By default, search is diacritical insensitive.* {{Code|stemming}}: determines is tokens are stemmed. Allowed values are {{Code|true}} and {{Code|false}}. By default, stemming is turned off.* {{Code|language}}: determines the language. This option is relevant for stemming tokens. All language codes are supported. The default language is {{Code|en}}.The {{Code|$options}} argument can be used to control full-text processing.|- valign="top"| '''Examples'''|* <code>ft:tokenize("No Doubt")</code> returns the two strings {{Code|no}} and {{Code|doubt}}.* <code>ft:tokenize("École", map { 'diacritics': 'sensitive' })</code> returns the string {{Code|école}}.* <code>declare ft-option using stemming; ft:tokenize("GIFTS")</code> returns a single string {{Code|gift}}.|} ==ft:normalize== {| width='100%'|- valign="top"| width='120' | '''Signature'''|<pre>ft:normalize( $string as xs:string?, $options as map(*)? := map { }) as xs:string</pre>|- valign="top"| '''Summary'''|Normalizes the given {{Code|$string}}, using the current default full-text options or the {{Code|$options}} specified as second argument. The function accepts the same arguments as {{Function||ft:tokenize}}; special characters and separators will be preserved.|- valign="top"| '''Examples'''

* <code>ft:scorenormalize("Häuser am Meer", map { 'acase' contains text : 'asensitive'})</code> returns the string {{Code|Hauser am Meer}}.|} ==ft:thesaurus== {| width='100%'|- valign="top"| width='120' | '''Signature'''|<codepre>ft:thesaurus( $node as node(), $term as xs:string, $options as map(*)? := map { }) as xs:doublestring*</pre>|- valign="top"| '''Summary'''|Looks up a {{Code|$term}} in a [[Full-Text#Thesaurus|Thesaurus Structure]] supplied by {{Code|$node}}. The following {{Code|$options}} exist:* {{Code|relationship}}: determines the relationship between terms* {{Code|levels}}: determines the maximum number of levels to traverse|- valign="top"| '''Examples'''| Returns {{Code|happy}} and {{Code|lucky}}:<pre lang='xquery'>ft:thesaurus( <thesaurus> <entry> <term>happy</term> <synonym> <term>lucky</term> <relationship>RT</relationship> </codesynonym> value <code/entry>1 </codethesaurus>, 'happy')</pre>.

=Highlighting Functions= ==ft:tokensmark== {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|{{Mono|<bpre>ft:tokens</b>mark( $db nodes as itemnode()) as element(value)*}}<br/>{{Mono|<b>ft:tokens</b>($db as item(), $prefix name as xs:string := ()) as elementnode(value)*}}</pre>|-| valign='"top' "| '''Summary'''|Returns all Puts a marker element around the resulting {{Code|$nodes}} of a full-text tokens stored in the index of the database <code>[[Database Module#Database Argument|$db]]request.<br/code>, along with their numbers The default name of occurrencesthe marker element is {{Code|mark}}. An alternative name can be chosen via the optional {{MonoCode|$dbname}} may either argument.<br/>Please note that:* The full-text expression that computes the token positions must be an specified as argument of the <code>xsft:stringmark()</code>function, denoting as all position information is lost in subsequent processing steps. You may need to specify more than one full-text expression if you want to use the database namefunction in a FLWOR expression, or as shown in Example 2.* The supplied node must be a node stored [[Database Module#Database Node|Database Node]]. As shown in the database.<br/>If Example 3, {{Code|update}} or {{MonoCode|$prefixtransform}} is specified, the returned nodes will can be refined utilized to the strings starting with that prefix. The prefix will be tokenized according convert a fragment to the full-text used for creating the indexrequired internal representation.|-| valign='"top' "| '''ErrorsExamples'''|'''[[XQuery Errors#BaseX Errors|BASX0001]]Example 1''' is raised : The following query returns {{Code|&lt;XML&gt;&lt;mark&gt;hello&lt;/mark&gt; world&lt;/XML&gt;}}, if one text node of the database {{Code|DB}} has the full-value "hello world":<pre lang='xquery'>ft:mark(db:get('DB')//*[text() contains text index is not available.'hello'])<br/pre>'''Example 2'''[[XQuery Errors#BaseX Errors|BASX0002]]: The following expression loops through the first ten full-text results and marks the results in a second expression:<pre lang='xquery'>let $start := 1let $end := 10let $term := 'welcome' is raised if let $test := function($node) {$node/text() contains text {Mono|$dbterm }} references a node that is not stored for $ft in a database (idb:get('DB')//*[$test(.e)])[position() = $start to $end]return ft:mark($ft[$test(., references a main-memory XML fragment).])<br/pre>'''Example 3''': The following expression returns <code>&lt;xml>hello &lt;b&gt;word&lt;/b&gt;&lt;/xml&gt;</code>:<pre lang='xquery'>copy $p := <xml>hello world</xml>modify ()return ft:mark($p[[XQuery Errors#BaseX Errors|BASX0003]text() contains text 'word'], 'b'' is raised if the addressed database cannot be opened.)</pre>

==ft:tokenizeextract== {|width='100%'|-| valign='"top' "| width='90120' | '''SignaturesSignature'''|{{Mono|<bpre>ft:tokenize</b>extract( $nodes as node()*, $input name as xs:string := () , $length as xs:stringinteger := ()) as node()*}}</pre>|-| valign='"top' "| '''Summary'''|Tokenizes Extracts and returns relevant parts of full-text results. It puts a marker element around the given resulting {{MonoCode|$inputnodes}} of a full-text index request and chops irrelevant sections of the result.<br/>The default element name of the marker element is {{Code|mark}} string, using . An alternative element name can be chosen via the current optional {{Code|$name}} argument.<br/>The default full-length of the returned text is {{Code|150}} characters. An alternative length can be specified via the optional {{Code|$length}} argument. Note that the effective text length may differ from the specified text optionsdue to formatting and readibility issues.<br/>For more details on this function, please have a look at {{Function||ft:mark}}.|-| valign='"top' "| '''Examples'''

* <code>ft:tokenize("No Doubt")</code> returns the two strings The following query may return {{MonoCode|no&lt;XML&gt;...&lt;b&gt;hello&lt;/b&gt;...&lt;XML&gt;}} and if a text node of the database {{MonoCode|doubtDB}}.contains the string "hello world":* <codepre lang='xquery'>declare ft-option using stemming; ft:tokenizeextract(db:get('DB')//*[text("GIFTS")contains text 'hello'], 'b', 1)</codepre> returns a single string |} =Errors= {| class="wikitable" width="100%"! width="110"|Code|Description|- valign="top"|{{MonoCode|giftoptions}}|Both wildcards and fuzzy search have been specified as search options.

===; Version 9.6* Added: {{Function||ft:thesaurus}}* Updated: {{Function||ft:search}}, {{Function||ft:contains}}: new {{Code|errors}} option. ; Version 9.1* Updated: {{Function||ft:tokenize}} and {{Function||ft:normalize}} can be called with empty sequence. ;Version 9.0* Updated: error codes updated; errors now use the module namespace ;Version 8.0* Added: {{Function||ft:contains}}, {{Function||ft:normalize}}* Updated: Options added to {{Function||ft:tokenize}} ;Version 7.1===8* Added: {{Function||ft:contains}}* Updated: Options added to {{Function||ft:search}} ;Version 7.7* Updated: the functions no longer accept [[Database Module#Database Nodes|Database Nodes]] as reference. Instead, the name of a database must now be specified.

;Version 7.2* AddedUpdated: [[#ft:tokens{{Function||ft:tokens]]search}} (second argument generalized, [[#ft:tokenize|ft:tokenize]]third parameter added)

[[Category;Version 7.1* Added:XQuery]]{{Function||ft:tokens}}, {{Function||ft:tokenize}}