src/third_party/icu/source/test/testdata/ssearch.xml

<?xml version="1.0" encoding="UTF-8"?>

<!-- Copyright (c) 2007-2009 IBM Corporation and others. All rights reserved -->

<!-- Test data file for string search  -->
<!DOCTYPE stringsearch-tests [
<!ELEMENT stringsearch-tests (test-case+)>
<!ATTLIST stringsearch-tests debug IDREF #IMPLIED >
<!ELEMENT test-case (pattern, pre?, m?, post?)>
<!ATTLIST test-case 
          id ID #REQUIRED
          locale CDATA "en" 
          strength (PRIMARY | SECONDARY | TERTIARY | QUATERNARY | IDENTICAL) "TERTIARY" 
          norm (ON | OFF) "OFF"
          alternate_handling (NON_IGNORABLE | SHIFTED) "NON_IGNORABLE"
          >

<!ELEMENT pattern (#PCDATA)>
<!ELEMENT pre  (#PCDATA)>
<!ELEMENT m    (#PCDATA)>
<!ELEMENT post (#PCDATA)>
]>

<stringsearch-tests>
  <!-- debug="test11"     (for copying into the above element)  -->
    
    <!-- Very simple match  -->
    <test-case id="test01" >
       <pattern>abc</pattern>
       <pre>xxx</pre><m>abc</m><post>yyy</post>
    </test-case>
    
    <!-- Very simple no-match  -->
    <test-case id="test02" >
       <pattern>abc</pattern>
       <pre>xxx</pre><post>yyy</post>
    </test-case>

    <!-- Match after several near-misses. -->
    <test-case id="test03" >
       <pattern>string</pattern>
       <pre>silly spring stling strxng strilg strinx stri</pre><m>string</m><post> fling</post>
    </test-case>
    
    <test-case id="test04" strength="PRIMARY" >
       <pattern>FUSS</pattern>
       <pre>abc</pre><m>fuss</m><post>sss</post>
    </test-case>
    
    <test-case id="test05" strength="PRIMARY" >
       <pattern>FUSS</pattern>
       <pre>abc</pre><m>fuß</m><post>sss</post>
    </test-case>

  <test-case id="test05.5" strength="PRIMARY" >
    <pattern>fuss</pattern>
    <pre>a </pre>
    <m>fuß</m>
    <post>ball table</post>
  </test-case>

  <test-case id="test06" strength="PRIMARY" >
      <pattern>fuß</pattern>
       <pre>abc</pre><m>fuss</m><post>xyz</post>
    </test-case>
    
    <test-case id="test07" strength="SECONDARY" >
      <pattern>fuß</pattern>
      <pre>abcfussxyz</pre>
    </test-case>
    
    <test-case id="test08" strength="PRIMARY" >
      <pattern>fus</pattern>
      <pre>abcfuß</pre><post>xyz</post>
    </test-case>
    
    <!-- A good match following an initial match that failed because
         of not ending on a character boundary -->
    <test-case id="test09" strength="PRIMARY">
      <pattern>fus</pattern>
      <pre>fuß  </pre><m>fus</m><post>sss</post>
    </test-case>


    <!-- Test cases from usrchdat.c  BREAKITERATOREXACT -->

    <test-case id="test10" strength="TERTIARY">
      <pattern>fox</pattern>
      <m>fox</m><post>y fox</post>
    </test-case>

    <test-case id="test11" strength="PRIMARY" locale="de_DE@collation=phonebook">
      <pattern>toe</pattern>
      <pre>This is a </pre><m>Tö</m><post>ne</post>
    </test-case>
    
    <test-case id="test11a" strength="SECONDARY" locale="de_DE@collation=phonebook">
      <pattern>toe</pattern>
      <pre>This is a </pre><post>Töne</post>
    </test-case>
    
    <test-case id="test12" strength="TERTIARY">
      <pattern>e</pattern>
      <pre>tésting that é doés not match </pre><m>e</m><post></post>
    </test-case>
    
    <test-case id="test13" strength="PRIMARY" locale="fr">
      <pattern>e</pattern>
      <pre></pre><m>É</m><post>É</post>
    </test-case>
    
    <test-case id="test14" strength="PRIMARY" locale="fr">
      <pattern>O</pattern>
      <pre>C</pre><m>O\u0302</m><post>TÉ</post>
    </test-case>


    <!-- Test cases from usrchdat.c  STRENGTH -->


    <test-case id="test15" strength="PRIMARY" locale="en">
      <pattern>fox</pattern>
      <pre>The quick brown </pre><m>fox</m><post> jumps over the lazy foxes</post>
    </test-case>
    
    <test-case id="test16" strength="PRIMARY" locale="fr">
      <pattern>peche</pattern>
      <pre>blackbirds pat </pre><m>p\u00E9ch\u00E9</m><post> </post>
    </test-case>
    
    <test-case id="test17" strength="PRIMARY" locale="fr">
      <pattern>peche</pattern>
      <pre>blackbirds pat </pre><m>p\u00EAche</m><post> </post>
    </test-case>
    
    <test-case id="test18" strength="PRIMARY" locale="fr">
      <pattern>peche</pattern>
      <pre>blackbirds pat </pre><m>p\u00E9che</m><post>r </post>
    </test-case>
    
    <test-case id="test19" strength="PRIMARY" locale="fr">
      <pattern>peche</pattern>
      <pre>blackbirds pat </pre><m>p\u00EAche</m><post>r </post>
    </test-case>
    
    <test-case id="test20" strength="PRIMARY" locale="es">
      <pattern>channel</pattern>
      <pre>A </pre><m>channel</m><post>, </post>
    </test-case>
    
    <test-case id="test21" strength="PRIMARY" locale="es">
      <pattern>channel</pattern>
      <pre>A </pre><m>CHANNEL</m><post>, </post>
    </test-case>
    
    <test-case id="test22" strength="PRIMARY" locale="es">
      <pattern>channel</pattern>
      <pre>A </pre><m>Channel</m><post>s, </post>
    </test-case>
    
    <test-case id="test23" strength="PRIMARY" locale="es">
      <pattern>channel</pattern>
      <pre>A </pre><m>channel</m><post>... </post>
    </test-case>
    
    <test-case id="test24" strength="TERTIARY" locale="en">
      <pattern>A\u0300</pattern>
      <pre>A miss, and then </pre><m>\u00c0</m><post> should match but not A"</post>
    </test-case>
    
    <!-- TODO:  In the original test data, this test matched at IDENTICAL strength.
                Doesn't seem right.  The characters are different.
                -->
    <test-case id="test24a" strength="IDENTICAL" locale="en">
      <pattern>A\u0300</pattern>
      <pre>At IDENTICAL, shoud this match?  </pre><m>\u00c0</m><post></post>
    </test-case>

  <test-case id="test24b" strength="IDENTICAL" alternate_handling="SHIFTED" locale="en">
    <pattern>A\u0300</pattern>
    <pre>At IDENTICAL, shoud this match?  </pre>
    <m>\u00c0</m>
    <post></post>
  </test-case>

  <test-case id="test25" strength="SECONDARY" locale="en">
      <pattern>Ű</pattern>
      <pre>12</pre><m>ű</m><post> Ű</post>
    </test-case>
    
    <test-case id="test26" strength="SECONDARY" locale="en">
      <pattern>A</pattern>
      <pre>12</pre><m>a</m><post>...</post>
    </test-case>


    <!--  Test Cases from usrchdat.c,  VARIABLE -->
    <test-case id="test27" strength="TERTIARY" locale="en">
      <pattern>blackbird</pattern>
      <pre>black-bird </pre><m>blackbird</m><post>...</post>
    </test-case>

    <test-case id="test28" strength="TERTIARY" locale="en">
      <pattern>go</pattern>
      <pre> on</pre>
    </test-case>

    <!-- TODO:  this gives an U_ILLEGAL_ARGUMENT error when opening
                the UStringSearch.  How did the orignal test run? -->
    <!--
    <test-case id="test29" strength="PRIMARY" locale="en">
      <pattern>  </pattern>
      <pre></pre><m></m><post>abc</post>
    </test-case>
    -->

    <test-case id="test30" strength="SECONDARY" locale="en">
      <pattern>abc</pattern>
      <pre>  a bc   ab c    a  bc     ab  c"</pre>
    </test-case>

    <test-case id="test31" strength="SECONDARY" locale="en">
      <pattern>abc</pattern>
      <pre>           ---------------</pre>
    </test-case>


    <!--  Normalization test cases from usrchdat.c  -->
    <test-case id="test32" strength="TERTIARY"  norm="ON">
      <pattern>a\u0325\u0300</pattern>
      <pre></pre><m>a\u0300\u0325</m>
    </test-case>


    <test-case id="test32a" strength="TERTIARY"  norm="OFF">
      <pattern>a\u0325\u0300</pattern>
      <pre>a\u0300\u0325</pre>
    </test-case>


    <!-- COMPOSITEBOUNDARIES from usrchdat.c
         Boundaries are not identical to orignal test data because
         of matching only full combining sequences
    -->
    <test-case id="test40" strength="TERTIARY">
      <pattern>A</pattern>
      <pre>À</pre>   <!-- \u00C0 -->
    </test-case>
    
    <test-case id="test41" strength="TERTIARY">
      <pattern>A</pattern>
      <pre>À</pre><m>A</m><post>C</post>
    </test-case>
    
    <test-case id="test42" strength="TERTIARY">
      <pattern>A\u030A</pattern>
      <pre>À\u01FA</pre>
    </test-case>



    <!-- SUPPLEMENTARYCANONICAL from usrchdat.c  -->
    <test-case id="test50" strength="TERTIARY">
      <pattern>\uD800\uDC00</pattern>
      <pre>abc \uD802\uDC00 \uD800\uDC01 \uD801\uDC00 </pre><m>\uD800\uDC00</m>
      <post>abc abc\uD800\uDC00 \uD800\uD800\uDC00 \uD800\uDC00\uDC00</post>
    </test-case>
    
    <test-case id="test51" strength="TERTIARY">
      <pattern>\\uD834\\uDDB9</pattern>
      <pre>and</pre><m>\\uD834\\uDDB9</m><post>this sentence</post>
    </test-case>

    <test-case id="test52" strength="TERTIARY">
      <pattern> \\uD834\\uDDB9 </pattern>
      <pre>and</pre><m> \\uD834\\uDDB9 </m><post>this sentence</post>
    </test-case>
    
    <test-case id="test53" strength="TERTIARY">
      <pattern>-\\uD834\\uDDB9-</pattern>
      <pre>and</pre><m>-\\uD834\\uDDB9-</m><post>this sentence</post>
    </test-case>
    
    <test-case id="test54" strength="TERTIARY">
      <pattern>,\\uD834\\uDDB9,</pattern>
      <pre>and</pre><m>,\\uD834\\uDDB9,</m><post>this sentence</post>
    </test-case>
    
    <test-case id="test55" strength="TERTIARY">
      <pattern>?\\uD834\\uDDB9?</pattern>
      <pre>and</pre><m>?\\uD834\\uDDB9?</m><post>this sentence</post>
    </test-case>
    

    <!-- Long combining sequences  -->
    <!-- Backwards search fails because patterns ends w/ ignorables
    <test-case id="test60" strength="PRIMARY">
      <pattern>A\u0301\u0301\u0301\u0301</pattern>
      <m>A\u0301\u0301\u0301\u0301\u0301</m>
    </test-case>
    -->

    <test-case id="test61" strength="TERTIARY">
      <pattern>A\u0301\u0301\u0301\u0301</pattern>
          <pre>A\u0301\u0301\u0301\u0301\u0301</pre>
    </test-case>
    
    <test-case id="test62" strength="TERTIARY">
      <pattern>A\u0301\u0301\u0301\u0301</pattern>
            <m>A\u0301\u0301\u0301\u0301</m>
    </test-case>

    <!-- stand-alone combining marks don't match attached marks  -->
    <test-case id="test63" strength="TERTIARY">
      <pattern>\u0301</pattern>
      <pre>A\u0301\u0301\u0301\u0301</pre>
    </test-case>
    
    <test-case id="test64" strength="TERTIARY">
      <pattern>\u0301</pattern>
      <post>\u0301\u0301\u0301\u0301</post>
    </test-case>

  <!-- stand-alone combining mark does match an un-attached combining mark -->
    <test-case id="test65" strength="TERTIARY">
       <pattern>\u0301</pattern>
       <m>\u0301</m><post>A\u0301\u0301</post>
    </test-case>

    <test-case id="test66" strength="TERTIARY">
       <pattern>\u0301</pattern>
       <m>\u0301</m>
    </test-case>
          
    <!-- stand-alone combining marks at end of the target text -->
    <test-case id="test67" strength="TERTIARY">
       <pattern>\u0301</pattern>
       <pre>abcd\r</pre><m>\u0301</m>
    </test-case>

      <!-- attached combining marks at end of the target text, no match -->
    <test-case id="test68" strength="TERTIARY">
       <pattern>\u0301</pattern>
       <pre>abcd\u0301</pre>
    </test-case>



   <!-- no match within expansions at the start -->
    <test-case id="test70" strength="PRIMARY">
      <pattern>Eligature</pattern>
      <pre>Æligature</pre>
    </test-case>

    <test-case id="test71" strength="PRIMARY">
      <pattern>AEligature</pattern>
      <m>Æligature</m>
    </test-case>

    <test-case id="test72" strength="PRIMARY">
        <pattern>AEligature</pattern>
        <m>Æligature</m>
    </test-case>
    
    <!-- unattached combining Tilde will not match a Tilde that is
         part of a composed Ñ  (\u00D1)  -->
    <test-case id="test73" strength="SECONDARY">
        <pattern>\u0303</pattern>  <!-- combining tilde -->
        <pre>Ñ&#x0d;</pre><m>\u0303</m>
    </test-case>
    
    <test-case id="test74" strength="SECONDARY">
        <pattern>\u0303</pattern>  <!-- combining tilde -->
        <pre>Ñ &#x0d;</pre><m>\u0303</m><post>a</post>
    </test-case>

  <test-case id="test75" strength="TERTIARY" locale="fr">
    <pattern>\u00EA</pattern>
    <pre>p</pre><m>\u00EA</m><post>che</post>
  </test-case>

  <test-case id="test76" strength="TERTIARY" locale="fr">
    <pattern>\u00EA</pattern>
    <pre>p</pre><m>e\u0302</m><post>che</post>
  </test-case>

  <test-case id="test77" strength="TERTIARY" locale="fr">
    <pattern>e\u0302</pattern>
    <pre>p</pre><m>\u00EA</m><post>che</post>
  </test-case>

  <!-- Test cases from ticket:5382 -->
  <test-case id="test78" strength="SECONDARY" locale="hu_HU">
    <pattern>\u0170</pattern>
    <m>\u0171</m>
    <post>12</post>
  </test-case>

  <test-case id="test79" strength="SECONDARY" locale="hu_HU">
    <pattern>\u0170</pattern>
    <pre>1</pre>
    <m>\u0171</m>
    <post>2</post>
  </test-case>

  <test-case id="test80" strength="SECONDARY" locale="hu_HU">
    <pattern>\u0170</pattern>
    <pre>12</pre>
    <m>\u0171</m>
  </test-case>
  
  <!-- Test cases from ticket:5959 -->
  <test-case id="test81" strength="SECONDARY">
    <pattern>\u2166</pattern>
    <m>VII</m>
  </test-case>

  <test-case id="test82" strength="SECONDARY">
    <pattern>VII</pattern>
    <m>\u2166</m>
  </test-case>

  <test-case id="test83" strength="IDENTICAL" alternate_handling="SHIFTED" locale="en">
    <pattern>Universal Declaration of Human Rights</pattern>
    <pre>Proclaims this </pre><m>Universal Declaration of Human Rights</m><post> as a common standard of achievement for all peoples and all nations</post>
  </test-case>

  <test-case id="test83b" strength="TERTIARY" alternate_handling="SHIFTED" locale="en">
    <pattern>Universal Declaration of Human Rights</pattern>
    <pre>Proclaims this </pre>
    <m>Universal-Declaration-of-Human-Rights</m>
    <post> as a common standard of achievement for all peoples and all nations</post>
  </test-case>

  <test-case id="test84" strength="TERTIARY" locale="en">
    <pattern>\u05E9\u0591\u05E9</pattern>
    <m>\u05E9\u0592\u05E9</m>
  </test-case>

  <test-case id="test84b" strength="IDENTICAL" locale="en">
    <pattern>\u05E9\u0591\u05E9</pattern>
    <pre>\u05E9\u0592\u05E9</pre>
  </test-case>
</stringsearch-tests>