postgresql/doc/src/sgml/isn.sgml

<sect1 id="isn">
 <title>isn</title>
 
 <indexterm zone="isn">
  <primary>isn</primary>
 </indexterm>

 <para>
  The <literal>isn</literal> module adds data types for the following 
  international-standard namespaces: EAN13, UPC, ISBN (books), ISMN (music), 
  and ISSN (serials). This module is inspired by Garrett A. Wollman's 
  isbn_issn code.
 </para>
 <para>
  This module validates, and automatically adds the correct 
  hyphenations to the numbers. Also, it supports the new ISBN-13 
  numbers to be used starting in January 2007.
 </para>

 <para>
  Premises:
 </para>

 <orderedlist>
  <listitem>
   <para>ISBN13, ISMN13, ISSN13 numbers are all EAN13 numbers</para>
  </listitem>
  <listitem>
   <para>EAN13 numbers aren't always ISBN13, ISMN13 or ISSN13 (some are)</para>
  </listitem>
  <listitem>
   <para>some ISBN13 numbers can be displayed as ISBN</para>
  </listitem>
  <listitem>
   <para>some ISMN13 numbers can be displayed as ISMN</para>
  </listitem>
  <listitem>
   <para>some ISSN13 numbers can be displayed as ISSN</para>
  </listitem>
  <listitem>
   <para>all UPC, ISBN, ISMN and ISSN can be represented as EAN13 numbers</para>
  </listitem>
 </orderedlist>

 <note>
  <para>
   All types are internally represented as 64 bit integers, 
   and internally all are consistently interchangeable.
  </para>
 </note>
 <note>
  <para>
   We have two operator classes (for btree and for hash) so each data type
   can be indexed for faster access.
  </para>
 </note>

 <sect2>
  <title>Data types</title>
  
  <para>
   We have the following data types:
  </para>

  <table>
   <title>Data types</title>
   <tgroup cols="2">
    <thead>
     <row>
      <entry><para>Data type</para></entry>
      <entry><para>Description</para></entry>
     </row>
    </thead>
    <tbody>
     <row>
      <entry><para><literal>EAN13</literal></para></entry>
      <entry>
       <para>
        European Article Numbers. This type will always show the EAN13-display 
        format. Te output function for this is <literal>ean13_out()</literal>
       </para>
      </entry>
     </row>

     <row>
      <entry><para><literal>ISBN13</literal></para></entry>
      <entry>
       <para>
        For International Standard Book Numbers to be displayed in 
          the new EAN13-display format.
       </para>
      </entry>
     </row>

     <row>
      <entry><para><literal>ISMN13</literal></para></entry>
      <entry>
       <para>
        For International Standard Music Numbers to be displayed in
        the new EAN13-display format.
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>ISSN13</literal></para></entry>
      <entry>
       <para>
        For International Standard Serial Numbers to be displayed in the new 
        EAN13-display format.
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>ISBN</literal></para></entry>
      <entry>
       <para>
        For International Standard Book Numbers to be displayed in the current 
        short-display format.
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>ISMN</literal></para></entry>
      <entry>
       <para>
        For International Standard Music Numbers to be displayed in the 
        current short-display format.
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>ISSN</literal></para></entry>
      <entry>
       <para>
        For International Standard Serial Numbers to be displayed in the 
        current short-display format. These types will display the short 
        version of the ISxN (ISxN 10) whenever it's possible, and it will 
        show ISxN 13 when it's impossible to show the short version.  The 
        output function to do this is <literal>isn_out()</literal>
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>UPC</literal></para></entry>
      <entry>
       <para>
        For Universal Product Codes. UPC numbers are a subset of the EAN13 
        numbers (they are basically EAN13 without the first '0' digit.)
        The output function to do this is also <literal>isn_out()</literal>
       </para>
      </entry>
     </row>
    </tbody>
   </tgroup>
  </table>

  <note>
   <para>
    <literal>EAN13</literal>, <literal>ISBN13</literal>, 
    <literal>ISMN13</literal> and <literal>ISSN13</literal> types will always 
    display the long version of the ISxN (EAN13). The output function to do 
    this is <literal>ean13_out()</literal>.
   </para>
   <para>
    The need for these types is just for displaying in different ways the same 
    data: <literal>ISBN13</literal> is actually the same as 
    <literal>ISBN</literal>, <literal>ISMN13=ISMN</literal> and 
    <literal>ISSN13=ISSN</literal>.
   </para>
  </note>
 </sect2>

 <sect2>
  <title>Input functions</title>
 
  <para>
   We have the following input functions:
  </para>

  <table>
   <title>Input functions</title>
   <tgroup cols="2">
    <thead>
     <row>
      <entry>Function</entry>
      <entry>Description</entry>
     </row>
    </thead>
    <tbody>
     <row>
      <entry><para><literal>ean13_in()</literal></para></entry>
      <entry>
       <para>
        To take a string and return an EAN13.
       </para>
      </entry>
     </row>

     <row>
      <entry><para><literal>isbn_in()</literal></para></entry>
      <entry>
       <para>
        To take a string and return valid ISBN or ISBN13 numbers.
       </para>
      </entry>
     </row>

     <row>
      <entry><para><literal>ismn_in()</literal></para></entry>
      <entry>
       <para>
        To take a string and return valid ISMN or ISMN13 numbers.
       </para>
      </entry>
     </row>

     <row>
      <entry><para><literal>issn_in()</literal></para></entry>
      <entry>
       <para>
        To take a string and return valid ISSN or ISSN13 numbers.
       </para>
      </entry>
     </row>
     <row>
      <entry><para><literal>upc_in()</literal></para></entry>
      <entry>
       <para>
        To take a string and return an UPC codes.
       </para>
      </entry>
     </row>
    </tbody>
   </tgroup>
  </table>
 </sect2>

 <sect2>
  <title>Casts</title>

  <para>
   We are able to cast from:
  </para>
  <itemizedlist>
   <listitem>
    <para>
     ISBN13 -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISMN13 -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISSN13 -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISBN -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISMN -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISSN -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     UPC  -&gt; EAN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISBN &lt;-&gt; ISBN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISMN &lt;-&gt; ISMN13
    </para>
   </listitem>
   <listitem>
    <para>
     ISSN &lt;-&gt; ISSN13
    </para>
   </listitem>
  </itemizedlist>
 </sect2>

 <sect2>
  <title>C API</title>
  <para>
   The C API is implemented as:
  </para>
  <programlisting>
   extern Datum isn_out(PG_FUNCTION_ARGS);
   extern Datum ean13_out(PG_FUNCTION_ARGS);
   extern Datum ean13_in(PG_FUNCTION_ARGS);
   extern Datum isbn_in(PG_FUNCTION_ARGS);
   extern Datum ismn_in(PG_FUNCTION_ARGS);
   extern Datum issn_in(PG_FUNCTION_ARGS);
   extern Datum upc_in(PG_FUNCTION_ARGS);
  </programlisting>

  <para>
   On success:
  </para>
  <itemizedlist>
   <listitem>
    <para>
     <literal>isn_out()</literal> takes any of our types and returns a string containing 
  the shortes possible representation of the number.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>ean13_out()</literal> takes any of our types and returns the 
  EAN13 (long) representation of the number.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>ean13_in()</literal> takes a string and return a EAN13. Which, as stated in (2) 
  could or could not be any of our types, but it certainly is an EAN13 
  number. Only if the string is a valid EAN13 number, otherwise it fails.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>isbn_in()</literal> takes a string and return an ISBN/ISBN13. Only if the string
  is really a ISBN/ISBN13, otherwise it fails.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>ismn_in()</literal> takes a string and return an ISMN/ISMN13. Only if the string
  is really a ISMN/ISMN13, otherwise it fails.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>issn_in()</literal> takes a string and return an ISSN/ISSN13. Only if the string
  is really a ISSN/ISSN13, otherwise it fails.
    </para>
   </listitem>
   <listitem>
    <para>
     <literal>upc_in()</literal> takes a string and return an UPC. Only if the string is
  really a UPC, otherwise it fails.
    </para>
   </listitem>
  </itemizedlist>

  <para>
   (on failure, the functions 'ereport' the error)
  </para>
 </sect2>

 <sect2>
  <title>Testing functions</title>
  <table>
   <title>Testing functions</title>
   <tgroup cols="2">
    <thead>
     <row>
      <entry><para>Function</para></entry>
      <entry><para>Description</para></entry>
     </row>
    </thead>
    <tbody>
     <row>
      <entry><para><literal>isn_weak(boolean)</literal></para></entry>
      <entry><para>Sets the weak input mode.</para></entry>
     </row>
     <row>
      <entry><para><literal>isn_weak()</literal></para></entry>
      <entry><para>Gets the current status of the weak mode.</para></entry>
     </row>
     <row>
      <entry><para><literal>make_valid()</literal></para></entry>
      <entry><para>Validates an invalid number (deleting the invalid flag).</para></entry>
     </row>
     <row>
      <entry><para><literal>is_valid()</literal></para></entry>
      <entry><para>Checks for the invalid flag prsence.</para></entry>
     </row>
    </tbody>
   </tgroup>
  </table>

  <para>
   <literal>Weak</literal> mode is used to be able to insert invalid data to
   a table. Invalid as in the check digit being wrong, not missing numbers.
  </para>
  <para>
   Why would you want to use the weak mode? Well, it could be that
   you have a huge collection of ISBN numbers, and that there are so many of
   them that for weird reasons some have the wrong check digit (perhaps the
   numbers where scanned from a printed list and the OCR got the numbers wrong,
   perhaps the numbers were manually captured... who knows.) Anyway, the thing
   is you might want to clean the mess up, but you still want to be able to have
   all the numbers in your database and maybe use an external tool to access
   the invalid numbers in the database so you can verify the information and
   validate it more easily; as selecting all the invalid numbers in the table.
  </para>
  <para>
   When you insert invalid numbers in a table using the weak mode, the number
   will be inserted with the corrected check digit, but it will be flagged 
   with an exclamation mark ('!') at the end (i.e. 0-11-000322-5!)
  </para>
  <para>
   You can also force the insertion of invalid numbers even not in the weak mode,
   appending the '!' character at the end of the number.
  </para>
 </sect2>

 <sect2>
  <title>Examples</title>
  <programlisting>
--Using the types directly:
SELECT isbn('978-0-393-04002-9');
SELECT isbn13('0901690546');
SELECT issn('1436-4522');

--Casting types:
-- note that you can only cast from ean13 to other type when the casted 
-- number would be valid in the realm of the casted type; 
-- thus, the following will NOT work: select isbn(ean13('0220356483481')); 
-- but these will: 
SELECT upc(ean13('0220356483481')); 
SELECT ean13(upc('220356483481')); 

--Create a table with a single column to hold ISBN numbers:
CREATE TABLE test ( id isbn );
INSERT INTO test VALUES('9780393040029');

--Automatically calculating check digits (observe the '?'):
INSERT INTO test VALUES('220500896?');
INSERT INTO test VALUES('978055215372?');

SELECT issn('3251231?');
SELECT ismn('979047213542?');

--Using the weak mode:
SELECT isn_weak(true);
INSERT INTO test VALUES('978-0-11-000533-4');
INSERT INTO test VALUES('9780141219307');
INSERT INTO test VALUES('2-205-00876-X');
SELECT isn_weak(false);
  
SELECT id FROM test WHERE NOT is_valid(id);
UPDATE test SET id=make_valid(id) WHERE id = '2-205-00876-X!';
  
SELECT * FROM test;

SELECT isbn13(id) FROM test;
  </programlisting>
 </sect2>

 <sect2>
  <title>Bibliography</title>
  <para>
   The information to implement this module was collected through
   several sites, including:
  </para>
  <programlisting>
   http://www.isbn-international.org/
   http://www.issn.org/
   http://www.ismn-international.org/
   http://www.wikipedia.org/
  </programlisting>
  <para>
   the prefixes used for hyphenation where also compiled from:
  </para>
  <programlisting>
   http://www.gs1.org/productssolutions/idkeys/support/prefix_list.html
   http://www.isbn-international.org/en/identifiers.html
   http://www.ismn-international.org/ranges.html
  </programlisting>
  <para>
   Care was taken during the creation of the algorithms and they
   were meticulously verified against the suggested algorithms
   in the official ISBN, ISMN, ISSN User Manuals.
  </para>
 </sect2>
 
 <sect2>
  <title>Author</title>
  <para>
   Germ<72>n M<>ndez Bravo (Kronuz), 2004 - 2006
  </para>
 </sect2>
</sect1>