Baoding-FenShuaJiang
/
Green_GoodERP18_FSJ_Standard
关注 1
点赞 0
百科 捐赠
gooderp18绿色标准版
您最多选择25个主题 主题必须以字母或数字开头，可以包含连字符 (-)，并且长度不得超过35个字符
目录树: 0f75e31513
分支列表 标签列表
${ item.name }
创建分支 ${ searchTerm }
从 '0f75e31513'
${ noResults }
Green_GoodERP18_FSJ_Standard/PostgreSQL/doc/postgresql/html/textsearch-debugging.html

253 行
18KB
原始文件 Blame 文件历史 

  1. <?xml version="1.0" encoding="UTF-8" standalone="no"?>

  2. <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>12.8. Testing and Debugging Text Search</title><link rel="stylesheet" type="text/css" href="stylesheet.css" /><link rev="made" href="pgsql-docs@lists.postgresql.org" /><meta name="generator" content="DocBook XSL Stylesheets V1.79.1" /><link rel="prev" href="textsearch-configuration.html" title="12.7. Configuration Example" /><link rel="next" href="textsearch-indexes.html" title="12.9. GIN and GiST Index Types" /></head><body><div xmlns="http://www.w3.org/TR/xhtml1/transitional" class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="5" align="center">12.8. Testing and Debugging Text Search</th></tr><tr><td width="10%" align="left"><a accesskey="p" href="textsearch-configuration.html" title="12.7. Configuration Example">Prev</a> </td><td width="10%" align="left"><a accesskey="u" href="textsearch.html" title="Chapter 12. Full Text Search">Up</a></td><th width="60%" align="center">Chapter 12. Full Text Search</th><td width="10%" align="right"><a accesskey="h" href="index.html" title="PostgreSQL 12.4 Documentation">Home</a></td><td width="10%" align="right"> <a accesskey="n" href="textsearch-indexes.html" title="12.9. GIN and GiST Index Types">Next</a></td></tr></table><hr></hr></div><div class="sect1" id="TEXTSEARCH-DEBUGGING"><div class="titlepage"><div><div><h2 class="title" style="clear: both">12.8. Testing and Debugging Text Search</h2></div></div></div><div class="toc"><dl class="toc"><dt><span class="sect2"><a href="textsearch-debugging.html#TEXTSEARCH-CONFIGURATION-TESTING">12.8.1. Configuration Testing</a></span></dt><dt><span class="sect2"><a href="textsearch-debugging.html#TEXTSEARCH-PARSER-TESTING">12.8.2. Parser Testing</a></span></dt><dt><span class="sect2"><a href="textsearch-debugging.html#TEXTSEARCH-DICTIONARY-TESTING">12.8.3. Dictionary Testing</a></span></dt></dl></div><p>

  3.    The behavior of a custom text search configuration can easily become

  4.    confusing.  The functions described

  5.    in this section are useful for testing text search objects.  You can

  6.    test a complete configuration, or test parsers and dictionaries separately.

  7.   </p><div class="sect2" id="TEXTSEARCH-CONFIGURATION-TESTING"><div class="titlepage"><div><div><h3 class="title">12.8.1. Configuration Testing</h3></div></div></div><p>

  8.    The function <code class="function">ts_debug</code> allows easy testing of a

  9.    text search configuration.

  10.   </p><a id="id-1.5.11.11.3.3" class="indexterm"></a><pre class="synopsis">

  11. ts_debug([<span class="optional"> <em class="replaceable"><code>config</code></em> <code class="type">regconfig</code>, </span>] <em class="replaceable"><code>document</code></em> <code class="type">text</code>,

  12.          OUT <em class="replaceable"><code>alias</code></em> <code class="type">text</code>,

  13.          OUT <em class="replaceable"><code>description</code></em> <code class="type">text</code>,

  14.          OUT <em class="replaceable"><code>token</code></em> <code class="type">text</code>,

  15.          OUT <em class="replaceable"><code>dictionaries</code></em> <code class="type">regdictionary[]</code>,

  16.          OUT <em class="replaceable"><code>dictionary</code></em> <code class="type">regdictionary</code>,

  17.          OUT <em class="replaceable"><code>lexemes</code></em> <code class="type">text[]</code>)

  18.          returns setof record

  19. </pre><p>

  20.    <code class="function">ts_debug</code> displays information about every token of

  21.    <em class="replaceable"><code>document</code></em> as produced by the

  22.    parser and processed by the configured dictionaries.  It uses the

  23.    configuration specified by <em class="replaceable"><code>config</code></em>,

  24.    or <code class="varname">default_text_search_config</code> if that argument is

  25.    omitted.

  26.   </p><p>

  27.    <code class="function">ts_debug</code> returns one row for each token identified in the text

  28.    by the parser.  The columns returned are

  29. 

  30.     </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: bullet; "><li class="listitem" style="list-style-type: disc"><p>

  31.        <em class="replaceable"><code>alias</code></em> <code class="type">text</code> — short name of the token type

  32.       </p></li><li class="listitem" style="list-style-type: disc"><p>

  33.        <em class="replaceable"><code>description</code></em> <code class="type">text</code> — description of the

  34.        token type

  35.       </p></li><li class="listitem" style="list-style-type: disc"><p>

  36.        <em class="replaceable"><code>token</code></em> <code class="type">text</code> — text of the token

  37.       </p></li><li class="listitem" style="list-style-type: disc"><p>

  38.        <em class="replaceable"><code>dictionaries</code></em> <code class="type">regdictionary[]</code> — the

  39.        dictionaries selected by the configuration for this token type

  40.       </p></li><li class="listitem" style="list-style-type: disc"><p>

  41.        <em class="replaceable"><code>dictionary</code></em> <code class="type">regdictionary</code> — the dictionary

  42.        that recognized the token, or <code class="literal">NULL</code> if none did

  43.       </p></li><li class="listitem" style="list-style-type: disc"><p>

  44.        <em class="replaceable"><code>lexemes</code></em> <code class="type">text[]</code> — the lexeme(s) produced

  45.        by the dictionary that recognized the token, or <code class="literal">NULL</code> if

  46.        none did; an empty array (<code class="literal">{}</code>) means it was recognized as a

  47.        stop word

  48.       </p></li></ul></div><p>

  49.   </p><p>

  50.    Here is a simple example:

  51. 

  52. </p><pre class="screen">

  53. SELECT * FROM ts_debug('english','a fat  cat sat on a mat - it ate a fat rats');

  54.    alias   |   description   | token |  dictionaries  |  dictionary  | lexemes 

  55. -----------+-----------------+-------+----------------+--------------+---------

  56.  asciiword | Word, all ASCII | a     | {english_stem} | english_stem | {}

  57.  blank     | Space symbols   |       | {}             |              | 

  58.  asciiword | Word, all ASCII | fat   | {english_stem} | english_stem | {fat}

  59.  blank     | Space symbols   |       | {}             |              | 

  60.  asciiword | Word, all ASCII | cat   | {english_stem} | english_stem | {cat}

  61.  blank     | Space symbols   |       | {}             |              | 

  62.  asciiword | Word, all ASCII | sat   | {english_stem} | english_stem | {sat}

  63.  blank     | Space symbols   |       | {}             |              | 

  64.  asciiword | Word, all ASCII | on    | {english_stem} | english_stem | {}

  65.  blank     | Space symbols   |       | {}             |              | 

  66.  asciiword | Word, all ASCII | a     | {english_stem} | english_stem | {}

  67.  blank     | Space symbols   |       | {}             |              | 

  68.  asciiword | Word, all ASCII | mat   | {english_stem} | english_stem | {mat}

  69.  blank     | Space symbols   |       | {}             |              | 

  70.  blank     | Space symbols   | -     | {}             |              | 

  71.  asciiword | Word, all ASCII | it    | {english_stem} | english_stem | {}

  72.  blank     | Space symbols   |       | {}             |              | 

  73.  asciiword | Word, all ASCII | ate   | {english_stem} | english_stem | {ate}

  74.  blank     | Space symbols   |       | {}             |              | 

  75.  asciiword | Word, all ASCII | a     | {english_stem} | english_stem | {}

  76.  blank     | Space symbols   |       | {}             |              | 

  77.  asciiword | Word, all ASCII | fat   | {english_stem} | english_stem | {fat}

  78.  blank     | Space symbols   |       | {}             |              | 

  79.  asciiword | Word, all ASCII | rats  | {english_stem} | english_stem | {rat}

  80. </pre><p>

  81.   </p><p>

  82.    For a more extensive demonstration, we

  83.    first create a <code class="literal">public.english</code> configuration and

  84.    Ispell dictionary for the English language:

  85.   </p><pre class="programlisting">

  86. CREATE TEXT SEARCH CONFIGURATION public.english ( COPY = pg_catalog.english );

  87. 

  88. CREATE TEXT SEARCH DICTIONARY english_ispell (

  89.     TEMPLATE = ispell,

  90.     DictFile = english,

  91.     AffFile = english,

  92.     StopWords = english

  93. );

  94. 

  95. ALTER TEXT SEARCH CONFIGURATION public.english

  96.    ALTER MAPPING FOR asciiword WITH english_ispell, english_stem;

  97. </pre><pre class="screen">

  98. SELECT * FROM ts_debug('public.english','The Brightest supernovaes');

  99.    alias   |   description   |    token    |         dictionaries          |   dictionary   |   lexemes   

  100. -----------+-----------------+-------------+-------------------------------+----------------+-------------

  101.  asciiword | Word, all ASCII | The         | {english_ispell,english_stem} | english_ispell | {}

  102.  blank     | Space symbols   |             | {}                            |                | 

  103.  asciiword | Word, all ASCII | Brightest   | {english_ispell,english_stem} | english_ispell | {bright}

  104.  blank     | Space symbols   |             | {}                            |                | 

  105.  asciiword | Word, all ASCII | supernovaes | {english_ispell,english_stem} | english_stem   | {supernova}

  106. </pre><p>

  107.    In this example, the word <code class="literal">Brightest</code> was recognized by the

  108.    parser as an <code class="literal">ASCII word</code> (alias <code class="literal">asciiword</code>).

  109.    For this token type the dictionary list is

  110.    <code class="literal">english_ispell</code> and

  111.    <code class="literal">english_stem</code>. The word was recognized by

  112.    <code class="literal">english_ispell</code>, which reduced it to the noun

  113.    <code class="literal">bright</code>. The word <code class="literal">supernovaes</code> is

  114.    unknown to the <code class="literal">english_ispell</code> dictionary so it

  115.    was passed to the next dictionary, and, fortunately, was recognized (in

  116.    fact, <code class="literal">english_stem</code> is a Snowball dictionary which

  117.    recognizes everything; that is why it was placed at the end of the

  118.    dictionary list).

  119.   </p><p>

  120.    The word <code class="literal">The</code> was recognized by the

  121.    <code class="literal">english_ispell</code> dictionary as a stop word (<a class="xref" href="textsearch-dictionaries.html#TEXTSEARCH-STOPWORDS" title="12.6.1. Stop Words">Section 12.6.1</a>) and will not be indexed.

  122.    The spaces are discarded too, since the configuration provides no

  123.    dictionaries at all for them.

  124.   </p><p>

  125.    You can reduce the width of the output by explicitly specifying which columns

  126.    you want to see:

  127. 

  128. </p><pre class="screen">

  129. SELECT alias, token, dictionary, lexemes

  130. FROM ts_debug('public.english','The Brightest supernovaes');

  131.    alias   |    token    |   dictionary   |   lexemes   

  132. -----------+-------------+----------------+-------------

  133.  asciiword | The         | english_ispell | {}

  134.  blank     |             |                | 

  135.  asciiword | Brightest   | english_ispell | {bright}

  136.  blank     |             |                | 

  137.  asciiword | supernovaes | english_stem   | {supernova}

  138. </pre><p>

  139.   </p></div><div class="sect2" id="TEXTSEARCH-PARSER-TESTING"><div class="titlepage"><div><div><h3 class="title">12.8.2. Parser Testing</h3></div></div></div><p>

  140.    The following functions allow direct testing of a text search parser.

  141.   </p><a id="id-1.5.11.11.4.3" class="indexterm"></a><pre class="synopsis">

  142. ts_parse(<em class="replaceable"><code>parser_name</code></em> <code class="type">text</code>, <em class="replaceable"><code>document</code></em> <code class="type">text</code>,

  143.          OUT <em class="replaceable"><code>tokid</code></em> <code class="type">integer</code>, OUT <em class="replaceable"><code>token</code></em> <code class="type">text</code>) returns <code class="type">setof record</code>

  144. ts_parse(<em class="replaceable"><code>parser_oid</code></em> <code class="type">oid</code>, <em class="replaceable"><code>document</code></em> <code class="type">text</code>,

  145.          OUT <em class="replaceable"><code>tokid</code></em> <code class="type">integer</code>, OUT <em class="replaceable"><code>token</code></em> <code class="type">text</code>) returns <code class="type">setof record</code>

  146. </pre><p>

  147.    <code class="function">ts_parse</code> parses the given <em class="replaceable"><code>document</code></em>

  148.    and returns a series of records, one for each token produced by

  149.    parsing. Each record includes a <code class="varname">tokid</code> showing the

  150.    assigned token type and a <code class="varname">token</code> which is the text of the

  151.    token.  For example:

  152. 

  153. </p><pre class="screen">

  154. SELECT * FROM ts_parse('default', '123 - a number');

  155.  tokid | token

  156. -------+--------

  157.     22 | 123

  158.     12 |

  159.     12 | -

  160.      1 | a

  161.     12 |

  162.      1 | number

  163. </pre><p>

  164.   </p><a id="id-1.5.11.11.4.6" class="indexterm"></a><pre class="synopsis">

  165. ts_token_type(<em class="replaceable"><code>parser_name</code></em> <code class="type">text</code>, OUT <em class="replaceable"><code>tokid</code></em> <code class="type">integer</code>,

  166.               OUT <em class="replaceable"><code>alias</code></em> <code class="type">text</code>, OUT <em class="replaceable"><code>description</code></em> <code class="type">text</code>) returns <code class="type">setof record</code>

  167. ts_token_type(<em class="replaceable"><code>parser_oid</code></em> <code class="type">oid</code>, OUT <em class="replaceable"><code>tokid</code></em> <code class="type">integer</code>,

  168.               OUT <em class="replaceable"><code>alias</code></em> <code class="type">text</code>, OUT <em class="replaceable"><code>description</code></em> <code class="type">text</code>) returns <code class="type">setof record</code>

  169. </pre><p>

  170.    <code class="function">ts_token_type</code> returns a table which describes each type of

  171.    token the specified parser can recognize.  For each token type, the table

  172.    gives the integer <code class="varname">tokid</code> that the parser uses to label a

  173.    token of that type, the <code class="varname">alias</code> that names the token type

  174.    in configuration commands, and a short <code class="varname">description</code>.  For

  175.    example:

  176. 

  177. </p><pre class="screen">

  178. SELECT * FROM ts_token_type('default');

  179.  tokid |      alias      |               description                

  180. -------+-----------------+------------------------------------------

  181.      1 | asciiword       | Word, all ASCII

  182.      2 | word            | Word, all letters

  183.      3 | numword         | Word, letters and digits

  184.      4 | email           | Email address

  185.      5 | url             | URL

  186.      6 | host            | Host

  187.      7 | sfloat          | Scientific notation

  188.      8 | version         | Version number

  189.      9 | hword_numpart   | Hyphenated word part, letters and digits

  190.     10 | hword_part      | Hyphenated word part, all letters

  191.     11 | hword_asciipart | Hyphenated word part, all ASCII

  192.     12 | blank           | Space symbols

  193.     13 | tag             | XML tag

  194.     14 | protocol        | Protocol head

  195.     15 | numhword        | Hyphenated word, letters and digits

  196.     16 | asciihword      | Hyphenated word, all ASCII

  197.     17 | hword           | Hyphenated word, all letters

  198.     18 | url_path        | URL path

  199.     19 | file            | File or path name

  200.     20 | float           | Decimal notation

  201.     21 | int             | Signed integer

  202.     22 | uint            | Unsigned integer

  203.     23 | entity          | XML entity

  204. </pre><p>

  205.    </p></div><div class="sect2" id="TEXTSEARCH-DICTIONARY-TESTING"><div class="titlepage"><div><div><h3 class="title">12.8.3. Dictionary Testing</h3></div></div></div><p>

  206.     The <code class="function">ts_lexize</code> function facilitates dictionary testing.

  207.    </p><a id="id-1.5.11.11.5.3" class="indexterm"></a><pre class="synopsis">

  208. ts_lexize(<em class="replaceable"><code>dict</code></em> <code class="type">regdictionary</code>, <em class="replaceable"><code>token</code></em> <code class="type">text</code>) returns <code class="type">text[]</code>

  209. </pre><p>

  210.     <code class="function">ts_lexize</code> returns an array of lexemes if the input

  211.     <em class="replaceable"><code>token</code></em> is known to the dictionary,

  212.     or an empty array if the token

  213.     is known to the dictionary but it is a stop word, or

  214.     <code class="literal">NULL</code> if it is an unknown word.

  215.    </p><p>

  216.     Examples:

  217. 

  218. </p><pre class="screen">

  219. SELECT ts_lexize('english_stem', 'stars');

  220.  ts_lexize

  221. -----------

  222.  {star}

  223. 

  224. SELECT ts_lexize('english_stem', 'a');

  225.  ts_lexize

  226. -----------

  227.  {}

  228. </pre><p>

  229.    </p><div class="note"><h3 class="title">Note</h3><p>

  230.      The <code class="function">ts_lexize</code> function expects a single

  231.      <span class="emphasis"><em>token</em></span>, not text. Here is a case

  232.      where this can be confusing:

  233. 

  234. </p><pre class="screen">

  235. SELECT ts_lexize('thesaurus_astro','supernovae stars') is null;

  236.  ?column?

  237. ----------

  238.  t

  239. </pre><p>

  240. 

  241.      The thesaurus dictionary <code class="literal">thesaurus_astro</code> does know the

  242.      phrase <code class="literal">supernovae stars</code>, but <code class="function">ts_lexize</code>

  243.      fails since it does not parse the input text but treats it as a single

  244.      token. Use <code class="function">plainto_tsquery</code> or <code class="function">to_tsvector</code> to

  245.      test thesaurus dictionaries, for example:

  246. 

  247. </p><pre class="screen">

  248. SELECT plainto_tsquery('supernovae stars');

  249.  plainto_tsquery

  250. -----------------

  251.  'sn'

  252. </pre><p>

  253.     </p></div></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="textsearch-configuration.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="textsearch.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="textsearch-indexes.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">12.7. Configuration Example </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> 12.9. GIN and GiST Index Types</td></tr></table></div></body></html>
上海开阖软件有限公司 沪ICP备12045867号-1