| List: | Commits | « Previous MessageNext Message » | |
| From: | stefan | Date: | August 22 2007 4:14pm |
| Subject: | svn commit - mysqldoc@docsrva: r7504 - in trunk: refman-4.1 refman-5.0 refman-5.1 refman-5.2 | ||
| View as plain text | |||
Author: shinz Date: 2007-08-22 16:14:47 +0200 (Wed, 22 Aug 2007) New Revision: 7504 Log: Add example for the effect that ?\195?\132=A and similar equalities have on comparisons or searches Modified: trunk/refman-4.1/charset.xml trunk/refman-5.0/charset.xml trunk/refman-5.1/charset.xml trunk/refman-5.2/charset.xml Modified: trunk/refman-4.1/charset.xml =================================================================== --- trunk/refman-4.1/charset.xml 2007-08-22 04:44:11 UTC (rev 7503) +++ trunk/refman-4.1/charset.xml 2007-08-22 14:14:47 UTC (rev 7504) Changed blocks: 5, Lines Added: 87, Lines Deleted: 2; 4110 bytes @@ -1953,9 +1953,14 @@ <section id="charset-collation-effect"> - <title>An Example of the Effect of Collation</title> + <title>Examples of the Effect of Collation</title> <para> + <emphasis role="bold">Example 1: Sorting German + Umlauts</emphasis> + </para> + + <para> Suppose that column <literal>X</literal> in table <literal>T</literal> has these <literal>latin1</literal> column values: @@ -2052,6 +2057,75 @@ </itemizedlist> + <para> + <emphasis role="bold">Example 2: Searching for German + Umlauts</emphasis> + </para> + + <para> + Suppose you have three tables that differ only by the character + set and collation used: + </para> + +<programlisting> +mysql> <userinput>CREATE TABLE german1 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin1_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE german2 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin2_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE germanutf8 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=utf8 COLLATE=utf8_general_ci;</userinput> +</programlisting> + + <para> + Each table contains two records: + </para> + +<programlisting> +mysql> <userinput>INSERT INTO german1 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO german2 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO germanutf8 VALUES ('Bar'), ('Bär');</userinput> +</programlisting> + + <para> + Two of the above collations have an <literal>A = Ä</literal> + equality, and one has no such equality + (<literal>latin2_general_ci</literal>). For that reason, you'll + get these results in comparisons: + </para> + +<programlisting> +mysql> <userinput>SELECT * FROM german1 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +mysql> <userinput>SELECT * FROM german2 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bär | ++------+ +mysql> <userinput>SELECT * FROM germanutf8 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +</programlisting> + + <para> + This is not a bug but rather a consequence of the sorting that + <literal>latin1_german1_ci</literal> or + <literal>utf8_general_ci</literal> do (which is according to the + German DIN 5007 standard). + </para> + </section> </section> @@ -3494,7 +3568,9 @@ <para> For example, the following equalities hold in both <literal>utf8_general_ci</literal> and - <literal>utf8_unicode_ci</literal>: + <literal>utf8_unicode_ci</literal> (for the effect this has in + comparisons or when doing searches, see + <xref linkend="charset-collation-effect"/>): </para> <programlisting> @@ -3781,6 +3857,9 @@ Normung</foreignphrase> (the German equivalent of ANSI). DIN-1 is called the <quote>dictionary collation</quote> and DIN-2 is called the <quote>phone book collation.</quote> + For an example which effect this has in comparisons or when + doing searches, see + <xref linkend="charset-collation-effect"/>. </para> <itemizedlist> @@ -3822,6 +3901,12 @@ </itemizedlist> <para> + For an example which effect this has in comparisons or when + doing searches, see + <xref linkend="charset-collation-effect"/>. + </para> + + <para> In the <literal>latin1_spanish_ci</literal> collation, ‘<literal>ñ</literal>’ (n-tilde) is a separate letter between ‘<literal>n</literal>’ and Modified: trunk/refman-5.0/charset.xml =================================================================== --- trunk/refman-5.0/charset.xml 2007-08-22 04:44:11 UTC (rev 7503) +++ trunk/refman-5.0/charset.xml 2007-08-22 14:14:47 UTC (rev 7504) Changed blocks: 4, Lines Added: 81, Lines Deleted: 2; 3643 bytes @@ -1885,9 +1885,14 @@ <section id="charset-collation-effect"> - <title>An Example of the Effect of Collation</title> + <title>Examples of the Effect of Collation</title> <para> + <emphasis role="bold">Example 1: Sorting German + Umlauts</emphasis> + </para> + + <para> Suppose that column <literal>X</literal> in table <literal>T</literal> has these <literal>latin1</literal> column values: @@ -1984,6 +1989,75 @@ </itemizedlist> + <para> + <emphasis role="bold">Example 2: Searching for German + Umlauts</emphasis> + </para> + + <para> + Suppose you have three tables that differ only by the character + set and collation used: + </para> + +<programlisting> +mysql> <userinput>CREATE TABLE german1 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin1_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE german2 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin2_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE germanutf8 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=utf8 COLLATE=utf8_general_ci;</userinput> +</programlisting> + + <para> + Each table contains two records: + </para> + +<programlisting> +mysql> <userinput>INSERT INTO german1 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO german2 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO germanutf8 VALUES ('Bar'), ('Bär');</userinput> +</programlisting> + + <para> + Two of the above collations have an <literal>A = Ä</literal> + equality, and one has no such equality + (<literal>latin2_general_ci</literal>). For that reason, you'll + get these results in comparisons: + </para> + +<programlisting> +mysql> <userinput>SELECT * FROM german1 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +mysql> <userinput>SELECT * FROM german2 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bär | ++------+ +mysql> <userinput>SELECT * FROM germanutf8 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +</programlisting> + + <para> + This is not a bug but rather a consequence of the sorting that + <literal>latin1_german1_ci</literal> or + <literal>utf8_general_ci</literal> do (which is according to the + German DIN 5007 standard). + </para> + </section> </section> @@ -3388,7 +3462,9 @@ <para> For example, the following equalities hold in both <literal>utf8_general_ci</literal> and - <literal>utf8_unicode_ci</literal>: + <literal>utf8_unicode_ci</literal> (for the effect this has in + comparisons or when doing searches, see + <xref linkend="charset-collation-effect"/>): </para> <programlisting> @@ -3675,6 +3751,9 @@ Normung</foreignphrase> (the German equivalent of ANSI). DIN-1 is called the <quote>dictionary collation</quote> and DIN-2 is called the <quote>phone book collation.</quote> + For an example which effect this has in comparisons or when + doing searches, see + <xref linkend="charset-collation-effect"/>. </para> <itemizedlist> Modified: trunk/refman-5.1/charset.xml =================================================================== --- trunk/refman-5.1/charset.xml 2007-08-22 04:44:11 UTC (rev 7503) +++ trunk/refman-5.1/charset.xml 2007-08-22 14:14:47 UTC (rev 7504) Changed blocks: 4, Lines Added: 81, Lines Deleted: 2; 3643 bytes @@ -1882,9 +1882,14 @@ <section id="charset-collation-effect"> - <title>An Example of the Effect of Collation</title> + <title>Examples of the Effect of Collation</title> <para> + <emphasis role="bold">Example 1: Sorting German + Umlauts</emphasis> + </para> + + <para> Suppose that column <literal>X</literal> in table <literal>T</literal> has these <literal>latin1</literal> column values: @@ -1981,6 +1986,75 @@ </itemizedlist> + <para> + <emphasis role="bold">Example 2: Searching for German + Umlauts</emphasis> + </para> + + <para> + Suppose you have three tables that differ only by the character + set and collation used: + </para> + +<programlisting> +mysql> <userinput>CREATE TABLE german1 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin1_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE german2 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin2_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE germanutf8 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=utf8 COLLATE=utf8_general_ci;</userinput> +</programlisting> + + <para> + Each table contains two records: + </para> + +<programlisting> +mysql> <userinput>INSERT INTO german1 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO german2 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO germanutf8 VALUES ('Bar'), ('Bär');</userinput> +</programlisting> + + <para> + Two of the above collations have an <literal>A = Ä</literal> + equality, and one has no such equality + (<literal>latin2_general_ci</literal>). For that reason, you'll + get these results in comparisons: + </para> + +<programlisting> +mysql> <userinput>SELECT * FROM german1 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +mysql> <userinput>SELECT * FROM german2 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bär | ++------+ +mysql> <userinput>SELECT * FROM germanutf8 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +</programlisting> + + <para> + This is not a bug but rather a consequence of the sorting that + <literal>latin1_german1_ci</literal> or + <literal>utf8_general_ci</literal> do (which is according to the + German DIN 5007 standard). + </para> + </section> </section> @@ -3385,7 +3459,9 @@ <para> For example, the following equalities hold in both <literal>utf8_general_ci</literal> and - <literal>utf8_unicode_ci</literal>: + <literal>utf8_unicode_ci</literal> (for the effect this has in + comparisons or when doing searches, see + <xref linkend="charset-collation-effect"/>): </para> <programlisting> @@ -3672,6 +3748,9 @@ Normung</foreignphrase> (the German equivalent of ANSI). DIN-1 is called the <quote>dictionary collation</quote> and DIN-2 is called the <quote>phone book collation.</quote> + For an example which effect this has in comparisons or when + doing searches, see + <xref linkend="charset-collation-effect"/>. </para> <itemizedlist> Modified: trunk/refman-5.2/charset.xml =================================================================== --- trunk/refman-5.2/charset.xml 2007-08-22 04:44:11 UTC (rev 7503) +++ trunk/refman-5.2/charset.xml 2007-08-22 14:14:47 UTC (rev 7504) Changed blocks: 4, Lines Added: 81, Lines Deleted: 2; 3643 bytes @@ -1882,9 +1882,14 @@ <section id="charset-collation-effect"> - <title>An Example of the Effect of Collation</title> + <title>Examples of the Effect of Collation</title> <para> + <emphasis role="bold">Example 1: Sorting German + Umlauts</emphasis> + </para> + + <para> Suppose that column <literal>X</literal> in table <literal>T</literal> has these <literal>latin1</literal> column values: @@ -1981,6 +1986,75 @@ </itemizedlist> + <para> + <emphasis role="bold">Example 2: Searching for German + Umlauts</emphasis> + </para> + + <para> + Suppose you have three tables that differ only by the character + set and collation used: + </para> + +<programlisting> +mysql> <userinput>CREATE TABLE german1 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin1_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE german2 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=latin1 COLLATE=latin2_german1_ci;</userinput> +mysql> <userinput>CREATE TABLE germanutf8 (</userinput> + -> <userinput> c CHAR(10)</userinput> + -> <userinput>) CHARSET=utf8 COLLATE=utf8_general_ci;</userinput> +</programlisting> + + <para> + Each table contains two records: + </para> + +<programlisting> +mysql> <userinput>INSERT INTO german1 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO german2 VALUES ('Bar'), ('Bär');</userinput> +mysql> <userinput>INSERT INTO germanutf8 VALUES ('Bar'), ('Bär');</userinput> +</programlisting> + + <para> + Two of the above collations have an <literal>A = Ä</literal> + equality, and one has no such equality + (<literal>latin2_general_ci</literal>). For that reason, you'll + get these results in comparisons: + </para> + +<programlisting> +mysql> <userinput>SELECT * FROM german1 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +mysql> <userinput>SELECT * FROM german2 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bär | ++------+ +mysql> <userinput>SELECT * FROM germanutf8 WHERE c = 'Bär';</userinput> ++------+ +| c | ++------+ +| Bar | +| Bär | ++------+ +</programlisting> + + <para> + This is not a bug but rather a consequence of the sorting that + <literal>latin1_german1_ci</literal> or + <literal>utf8_general_ci</literal> do (which is according to the + German DIN 5007 standard). + </para> + </section> </section> @@ -3385,7 +3459,9 @@ <para> For example, the following equalities hold in both <literal>utf8_general_ci</literal> and - <literal>utf8_unicode_ci</literal>: + <literal>utf8_unicode_ci</literal> (for the effect this has in + comparisons or when doing searches, see + <xref linkend="charset-collation-effect"/>): </para> <programlisting> @@ -3672,6 +3748,9 @@ Normung</foreignphrase> (the German equivalent of ANSI). DIN-1 is called the <quote>dictionary collation</quote> and DIN-2 is called the <quote>phone book collation.</quote> + For an example which effect this has in comparisons or when + doing searches, see + <xref linkend="charset-collation-effect"/>. </para> <itemizedlist>
| Thread | ||
|---|---|---|
| • svn commit - mysqldoc@docsrva: r7504 - in trunk: refman-4.1 refman-5.0 refman-5.1 refman-5.2 | stefan | 22 Aug |
© 1995-2008 MySQL AB, 2008-2009 Sun Microsystems, Inc.
Page generated in 0.475 sec. using MySQL 5.1.14-beta-log