% layout 'main', title => 'KorAP: Query Languages';

%# https://letsencrypt.org/howitworks/

%# Store the id of an active section in the session, so the system is able to directly scroll to the relevant section
%# This should be stored when clicking on a specific query
%# but the remembered section contains the id - not the query

<h2>Query Languages</h2>

<section id="intro">
  <h3>Frontend Features</h3>
  <p>This frontend differs to the <%= doc_ext_link_to 'official frontend', 'http://korap.ids-mannheim.de/app/', target => '_blank' %> by providing a serialization view, an integrated tutorial, a comparison view for morphological annotations, and an autocompletion for closed annotations (type in <%= doc_link_to 'foundry prefixes', 'data', 'annotation' %> like <code>cnx/</code>).</p>
</section>

<section id="examples">
  <h3>Example Queries</h3>

  %# Tests:
  %# [is => 'json_pointer', 'result']
  %# [ok => 'json_pointer']
  %# [isnt => 'json_pointer', 'result']
  %# [not_ok => 'json_pointer']
  
  <p><strong>Poliqarp</strong>: Find all occurrences of the lemma &quot;baum&quot; as annotated by the <%= doc_link_to 'default foundry', 'data', 'annotation' %>.</p>
  %= doc_query poliqarp => '[base=Baum]', 'tests' => [[is => '/query', 'tokens:tt/l:Baum'],[is => '/request/query/wrap/layer', 'lemma'],[is => '/request/query/wrap/foundry', 'tt'], [ok => '/matches/10']]

  <p><strong>Poliqarp</strong>: Find all sequences of adjectives as annotated by Treetagger, that are repeated 3 to 5 times in a row.</p>
  %= doc_query poliqarp => '[tt/p=ADJA]{3,5}', 'tests' => [[is => '/query', 'spanRepetition(tokens:tt/p:ADJA{3,5})'], [is => '/request/query/operation', 'operation:repetition'],[is => '/request/query/operands/0/wrap/foundry', 'tt'], [ok => '/matches/5']]

  <p><strong>Cosmas-II</strong>: Find all occurrences of the words &quot;der&quot; and &quot;Baum&quot;, in case they are in a maximum distance of 5 tokens. The order is not relevant.</p>
  %= doc_query cosmas2 => 'der /w5 Baum', 'tests' => [[is => '/query', 'shrink(129: spanDistance({129: tokens:s:der}, {129: tokens:s:Baum}, [(w[0:5], notOrdered, notExcluded)]))'], [is => '/request/query/operation', 'operation:focus'], [is => '/request/query/@type', 'korap:reference'],[is => '/request/query/operands/0/operands/1/operation', 'operation:class'], [is => '/itemsPerPage', 25], [ok => '/matches/20'], [is => '/matches/4/corpusID', 'WPD'], [is => '/matches/12/corpusID', 'WPD']]

  <p><strong>Cosmas-II</strong>: Find all sequences of a word starting with a &quot;d&quot; (using a wildcard) followed by an adjective as annotated in the mate foundry, followed by the word &quot;Baum&quot; (ignore the case), that is in a sentence element annotated by the <%= doc_link_to 'default foundry', 'data', 'annotation' %>.</p>
  <p><em>Be aware</em>: Minor incompatibilities with implemented languages may be announced with warnings.</p>
  %= doc_query cosmas2 => 'd* MORPH(mate/p=ADJA) $Baum #IN #ELEM(s)', 'tests' => [[ok => '/matches/3'], [is => '/query', 'shrink(130: {131: spanContain({129: <tokens:s />}, {130: spanNext(spanNext(SpanMultiTermQueryWrapper(tokens:s:d*), tokens:mate/p:ADJA), tokens:i:baum)})})'], [is => '/request/query/@type', 'korap:reference'], [is => '/request/query/operation', 'operation:focus'], [is => '/request/query/operands/0/operands/0/operation', 'operation:position'], [is => '/request/query/operands/0/operands/0/operands/1/operands/0/operation', 'operation:sequence'], [is => '/request/query/operands/0/operands/0/operands/1/operands/0/operands/0/wrap/type', 'type:wildcard'], [is => '/request/query/operands/0/operands/0/operands/1/operands/0/operands/1/wrap/key', 'ADJA'], [is => '/request/query/operands/0/operands/0/operands/1/operands/0/operands/1/wrap/foundry', 'mate'], [ok => '/request/query/operands/0/operands/0/operands/1/operands/0/operands/2/wrap/caseInsensitive'], [ok => '/matches/5']]

  <p><strong>Poliqarp+</strong>: Find all nominal phrases as annotated using Connexor, that contain an adverb as annotated by OpenNLP, that is annotated as something starting with an &quot;A&quot; using regular expressions in Treetagger.</p>
  %= doc_query poliqarp => 'contains(<cnx/c=np>,{[opennlp/p=ADV & tt/p="A.*"]})', cutoff => 1, 'tests' => [[is => '/query', 'spanContain(<tokens:cnx/c:np />, {1: spanSegment(tokens:opennlp/p:ADV, SpanMultiTermQueryWrapper(tokens:/tt/p:A.*/))})'], [is => '/request/query/operation', 'operation:position'], [is => '/request/query/frames/0', 'frames:contains'], [is => '/request/query/operands/0/foundry', 'cnx'], [is => '/request/query/operands/0/layer', 'c'], [is => '/request/query/operands/0/foundry', 'cnx'], [is => '/request/query/operands/0/key', 'np'], [is => '/request/query/operands/1/operands/0/wrap/operands/0/foundry', 'opennlp'], [is => '/request/query/operands/1/operands/0/wrap/operands/0/layer', 'p'],  [is => '/request/query/operands/1/operands/0/wrap/operands/1/foundry', 'tt'],  [is => '/request/query/operands/1/operands/0/wrap/operands/1/type', 'type:regex'], [is => '/request/query/operands/1/operands/0/wrap/operands/1/key', 'A.*'], [ok => '/matches/5']]

  <p><strong>Poliqarp+</strong>: Find all sentences as annotated by the base foundry that start with a sequence of one token in present tense as annotated by Connexor and the lemma &quot;die&quot; annotated by the <%= doc_link_to 'default foundry', 'data', 'annotation' %>. Highlight both terms of the sequence.</p>
  %= doc_query poliqarp => 'startswith(<s>, {1:[cnx/m=PRES]}{2:[base=die]})', cutoff => 1, tests => [[is => '/query', 'spanStartsWith(<tokens:s />, spanNext({1: tokens:cnx/m:PRES}, {2: tokens:tt/l:die}))'], [is => '/request/meta/startPage', 1], [is => '/request/query/operation', 'operation:position'], [is => '/request/query/operands/0/@type','korap:span'], [is => '/request/query/operands/1/operands/0/operation', 'operation:class'], [is => '/request/query/operands/1/operands/1/operation', 'operation:class'], [is => '/request/query/operands/1/operands/1/operands/0/wrap/foundry', 'tt'], [ok => '/matches/4']]

  <p><strong>Poliqarp+</strong>: Find all sequences of an article, followed by three to four adjectives and a noun as annotated by the Treetagger foundry, that finish a sentence. Highlight all parts of the sequence.</p>
  %= doc_query poliqarp => 'focus(3:endswith(<s>,{3:[tt/p=ART]{1:{2:[tt/p=ADJA]{3,4}}[tt/p=NN]}}))', cutoff => 1, 'tests' => [[is => '/query', 'shrink(3: spanEndsWith(<tokens:s />, {3: spanNext(tokens:tt/p:ART, {1: spanNext({2: spanRepetition(tokens:tt/p:ADJA{3,4})}, tokens:tt/p:NN)})}))'], [is => '/request/query/operation', 'operation:focus'], [is => '/request/query/operands/0/frames/0', 'frames:endswith'], [ok => '/matches/3'], [is => '/matches/4/corpusID', 'WPD']]

  <p><strong>Annis</strong>: Find all occurrences of the sequence of two tokens annotated as adverbs by the <%= doc_link_to 'default foundry', 'data', 'annotation' %>.</p>
  %= doc_query annis => 'pos="ADV" & pos="ADV" & #1 . #2', 'tests' => [[is => '/query', 'spanNext(tokens:tt/p:ADV, tokens:tt/p:ADV)'], [is => '/request/query/operands/0/wrap/foundry', 'tt'], [is => '/request/query/operands/1/wrap/foundry', 'tt'], [ok => '/matches/5'], [ok => '/matches/15'], [is => '/matches/15/corpusID', 'WPD']]

  <p><strong>CQL</strong>: Find all occurrences of the sequence &quot;der alte Mann&quot;.</p>
  %= doc_query cql => '"der alte Mann"', 'tests' => [[is => '/query', 'spanNext(spanNext(tokens:s:der, tokens:s:alte), tokens:s:Mann)'], [is => '/request/query/operation', 'operation:sequence'],[is => '/request/query/operands/0/wrap/key', 'der'],[is => '/request/query/operands/1/wrap/key', 'alte'],[is => '/request/query/operands/2/wrap/key', 'Mann'], [ok => '/matches/5'], [ok => '/matches/5']]
	
</section>