readme_usearch.html

来自「Debian中文参考手册,系统介绍了Debian系统」· HTML 代码 · 共 581 行 · 第 1/2 页

<br><a NAME="999733"></a><!--TOC=h2-"999733"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">4.2.2 Query
Performance Tuning</font></font></h4>
<!--/TOC=h2--><a NAME="999734"></a>

<p CLASS="BP">This section contains suggestions on how to improve the response
time of the Ultra Search query.
<br><a NAME="999735"></a><!--TOC=h3-"999735"-->
<h5 CLASS="H3">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">4.2.2.1
Tune the DB_CACHE_SIZE Parameter</font></font></h5>
<!--/TOC=h3--><a NAME="999736"></a>

<p CLASS="BP">The database buffer cache keeps frequently accessed data
read from datafiles, and efficient usage of the buffer cache can improve
Ultra Search query performance. The cache size is controlled by the <tt>DB_CACHE_SIZE</tt>
initialization parameter.
<br><a NAME="999737"></a>

<p CLASS="BP">For more information on how to tune this parameter, see <i>Oracle9i
Database Tuning Performance Guide and Reference</i>.
<br><a NAME="999738"></a><!--TOC=h3-"999738"-->
<h5 CLASS="H3">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">4.2.2.2
Optimize the Index</font></font></h5>
<!--/TOC=h3--><a NAME="999739"></a>

<p CLASS="BP">Optimize the Ultra Search index after the crawler has made
substantial updates. This can be done by scheduling index optimization
on a regular basis. Make sure index optimization is scheduled during off-peak
hours, because query performance will be significantly degraded during index optimization.
<br><a NAME="999740"></a>

<p CLASS="BP">For information on index optimization schedules, see the
Ultra Search online documentation about the Schedules Page (<tt>$ORACLE_HOME/ultrasearch/doc/help/a_schedules</tt>.<tt>htm</tt>).
<br><a NAME="999741"></a><!--TOC=h3-"999741"-->
<h5 CLASS="H3">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">4.2.2.3
Optimize the Index Based on Tokens</font></font></h5>
<!--/TOC=h3--><a NAME="999742"></a>

<p CLASS="BP">Optimize the Ultra Search index by basing it on frequently
searched tokens. Queries can be logged by turning on query statistics collection
in the Administration tool. The frequently searched tokens then can be
passed to <tt>CTX_DDL</tt>.<tt>OPTIMIZE_INDEX</tt> in token mode. The Ultra
Search index name is <tt>WK$DOC_PATH_IDX</tt>.
<br><a NAME="999743"></a>

<p CLASS="BP">For more information on <tt>OPTIMIZE_INDEX</tt>, see <i>Oracle
Text Reference</i>.
<br><a NAME="999744"></a><!--TOC=h3-"999744"-->
<h5 CLASS="H3">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">4.2.2.4
Simplify Query Expansion</font></font></h5>
<!--/TOC=h3--><a NAME="999745"></a>

<p CLASS="BP">The search response time is directly influenced by the Text
query string used. Although Ultra Search provides a default mechanism to
expand user input into a Text query, simpler expansions can greatly reduce
search time.
<br><a NAME="999746"></a>

<p CLASS="BP">For information on customizing query expansion, see the Ultra
Search online documentation about Customizing the Query Syntax Expansion
(<tt>$ORACLE_HOME/qsyntax</tt>.<tt>htm</tt>) and the Javadoc for the <tt>oracle</tt>.<tt>ultrasearch</tt>.<tt>query</tt>.<tt>Query</tt>
interface.
<br><a NAME="999468"></a><!--TOC=h1-"999468"-->
<h2 CLASS="H1">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5 Known
Bugs</font></font></h2>
<!--/TOC=h1--><a NAME="997685"></a><!--TOC=h2-"997685"-->
<h3 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.1 Query
Bugs</font></font></h3>

<h3>
<!--/TOC=h2--><a NAME="997686"></a></h3>

<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.1.1 Bug
2114417 - Unwanted/Invalid Syntax Displayed For Query String</font></font></h4>
<!--/TOC=h2--><a NAME="997705"></a>

<p CLASS="BP">When query statistics collection is enabled, the query statistics
pages (Daily summary of query statistics, Top 50 queries, Top 50 ineffective
queries, Top 50 failed queries) may show text query strings like '(((WKA2X
&amp;({abc}))WITHIN S2))*2,({abc}))'. This behavior is due to 9.2.0 Java
API expanding the user's query ('abc' in this case) before it is sent to
the database.
<br><a NAME="999251"></a>

<p CLASS="BP">In most cases, the user's query can be deciphered from the
text query.
<br><a NAME="997708"></a><!--TOC=h2-"997708"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.1.2 Bug
2097381 - Portal Users Should Embed Ultra Search Portlets Hosted On Same
OC4J Instance</font></font></h4>
<!--/TOC=h2--><a NAME="997734"></a>

<p CLASS="BP">Portal users should embed Ultra Search portlets that are
hosted on the same OC4J instance as the Oracle9iAS Portal server. For example,
if the Oracle Portal OC4J instance is installed on host A / port 7777,
then the Ultra Search provider must also be hosted as a Web application
on host A / port 7777.
<p><a NAME="997736"></a>

<p CLASS="BP">It is possible that the Ultra Search provider running on
host A / port 777 could be registered with a second Oracle Portal instance
running on a different host / port combination. In such cases, when the
Ultra Search portlet is embedded within portal pages, the pop-up list-of-values
will not work correctly. This is because of a security bug inherent in
Javascript.
<p><a NAME="762987"></a><!--/TOC=h1--><a NAME="997685"></a><!--TOC=h2-"997685"-->
<h3 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.2 Admin
Bugs</font></font></h3>

<h3>
<!--/TOC=h2--><a NAME="997686"></a></h3>

<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.2.1 Bug
2313955 - Admin UI Display Incorrect Message for Database Connect String
Format</font></font></h4>
The message display on "crawler settings" page is incorrect about the format
of the jdbc connect string. The correct message should be -> The format
of the connect string can be a tnsname entry or [hostname]:[port]:[SID].
<h3 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.3 Crawler
Bugs</font></font></h3>

<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.3.1 Bug
2166510 - NLS: Crawler Shutdown Unexpected If Log Directory Name Contains
MBCS</font></font></h4>
<!--/TOC=h2--><a NAME="997692"></a>

<p CLASS="BP">The crawler is unable to handle log directory paths with multibyte
characters.
<br><a NAME="999246"></a>

<p CLASS="BP">Avoid specifying log directories that have Chinese, Japanese,
or Korean characters.
<br><a NAME="997694"></a><!--TOC=h2-"997694"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.3.2 Bug
2166662 - NLS: Can't Find Files With MBCS Name</font></font></h4>
<!--/TOC=h2--><a NAME="997695"></a>

<p CLASS="BP">File data source crawling cannot crawl directories or files
with multibyte characters; for example, Chinese or Japanese.
<br><a NAME="999245"></a>

<p CLASS="BP">Avoid naming the file in Chinese, Japanese, or Korean or
putting them under such directory.
<br><a NAME="997697"></a><!--TOC=h2-"997697"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.3.3 Bug
2186745 - File Data Source Can Not Crawl Directory Or File Name With HTML
Reserved Symbol</font></font></h4>
<!--/TOC=h2--><a NAME="997698"></a>

<p CLASS="BP">The crawler is unable to pick up files or directories whose
name contains HTML reserved symbols, like '&lt;' or '>', when doing file
data source crawling.
<br><a NAME="997699"></a>

<p CLASS="BP">Rename the file or directory that is using such symbol.
<br><a NAME="997700"></a><!--TOC=h2-"997700"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.3.4 Bug
2265100 - Unable To Stop Crawling When Crawler Is Enqueuing URL From A
Crawler Agent</font></font></h4>
<!--/TOC=h2--><a NAME="997701"></a>

<p CLASS="BP">Stop crawling does not stop the crawler, even though the
schedule status shows that the crawler has stopped. This happens when the
crawler agent is used and the crawler is in the process of enqueuing URLs
fetched from the agent. The crawler stops only after enqueuing is finished.
<br><a NAME="999250"></a>

<p CLASS="BP">Currently, there is no way to stop the enqueuing other than
manually killing the crawler process.
<br><a NAME="997704"></a><!--TOC=h2-"997704"-->
<h3 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">
5.4 Oracle Portal Dependencies (Known bugs in 9.0.2) </font></font></h3>

<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">5.4.1 Bug 2304242 -Customer database silent mode install hanged at Ultra Search
phase</font></font></h4>
<!--/TOC=h2--><a NAME="997720"></a>
<p>For database installation using Oracle Portal Configuration Assistant
(OPCA), Ultra Search only supports silent mode when Ultra Search is not installed 
in the target database. For interactive mode, or if Ultra Search has already been installed, the "DISPLAY" environment variable must be set
correctly.

<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">
5.4.2 Bug 2244234 - Translations Of Items Have Same Display URL In Portal XML</font></font></h4>
<!--/TOC=h2--><a NAME="997720"></a>

<p CLASS="BP">In Portal, the URLs for translations of an item have the
same display URL as that of the base language item. Portal users can view
different translations because when users log in to Portal, the language
is established as part of the browser session. However, this language negotiation
process only works with browsers operated by human users. Therefore, the
Ultra Search crawler receives the same display URL for the translated items.
This violates the stated requirement that all display URLs presented to
Ultra Search be unique. The implication is that Ultra Search cannot crawl
translations of an item.
<br><a NAME="999255"></a>

<p CLASS="BP">As in bug 2218987, with multiple translations, only one of
the translation items or the base item itself is indexed by the crawler.
The rest are rejected by the crawler because of the duplicate display URLs.
<br><a NAME="997726"></a><!--TOC=h2-"997726"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">
5.4.3 Bug 2244239 - Some Attributes In Portal XML Are Not Encased In Translations
Section</font></font></h4>
<!--/TOC=h2--><a NAME="997727"></a>

<p CLASS="BP">If there are translations for an item or page, then some
attributes of that item/page cannot be correctly transmitted to the Ultra
Search crawler. As a result, attribute queries may not work correctly for
translated items.
<br><a NAME="997729"></a><!--TOC=h2-"997729"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">
5.4.4 Bug 2244254 - Portal XML Should Not Reveal Item Types Of BASETYPE=NONE</font></font></h4>
<!--/TOC=h2--><a NAME="997730"></a>

<p CLASS="BP">Ultra Search crawler can process specific Portal item types.
However, Portal item type of "none" does not have display URLs. As a result,
they are not revealed to Ultra Search. Because anything that does not have
a display URL cannot be represented in the search application search results
list in such a way that the user can click on it to view the item.

<p>
Observed effects:
<br>
(1) Search results are unaffected because Ultra Search gracefully ignores such items and logs an error message in the crawler log file. 
<br>
(2) However, if there are many such items in the Portal page group being crawled, then the crawler log file will contain many error
messages. 

<p><a NAME="999468"></a><!--TOC=h1-"999468"-->
<h2 CLASS="H1">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">6 Documentation
Errata</font></font></h2>
<a NAME="997729"></a><!--TOC=h2-"997729"-->
<h3 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">6.1 Oracle
Ultra Search Online Documentation</font></font></h3>
<!--/TOC=h2--><a NAME="997730"></a>
<p>The following are known issues with the Oracle Ultra Search Online Documentation.
<p><a NAME="997729"></a><!--TOC=h2-"997729"-->
<h4 CLASS="H2">
<font face="Arial, Helvetica, sans-serif"><font color="#330099">6.1.1 Method
compileForCount is not documented in Ultra Search Java Query API section</font></font></h4>
<!--/TOC=h2--><a NAME="997730"></a>
<p>The Oracle Ultra Search Online Documentation includes a javadoc section
for Ultra Search Java Query API. There is
<br>one method which is not documented in oracle.ultrasearch.query.Query
interface.
<p>The method signature is public java.util.String compileForCount(); The
function of this method is similar to the compile() method. It compiles
the Query object into some equivalent Text query string. But the string
returned by compileForCount() is passed to 'CTX_QUERY.COUNT_HITS', not
'CTX_QUERY.CONTAINS'. compileForCount() is needed because the Text query
for COUNT_HITS can be much simpler than CONTAINS (because scoring is not
a concern), and a simpler Text query can improve performance.
<p>Classes implementing the Query interface must implement both the compile()
and compileForCount() methods. A trivial,
<br>albeit slow, implementation of compileForCount() can return the same
string as compile().
<br>&nbsp;
<br>&nbsp;

<p CLASS="CopyrightBody">Copyright &copy; 2002 Oracle Corporation.
<br><a NAME="756053"></a>

<p CLASS="CopyrightBody">All Rights Reserved.</div>

</body>
</html>