Add a note warning that COPY BINARY is very datatype-specific.

Per a complaint from Gordon Shannon.

doc/src/sgml/ref/copy.sgml

@@ -1,5 +1,5 @@
 <!--
-$PostgreSQL: pgsql/doc/src/sgml/ref/copy.sgml,v 1.86 2009/07/25 00:07:10 adunstan Exp $
+$PostgreSQL: pgsql/doc/src/sgml/ref/copy.sgml,v 1.87 2009/09/05 23:58:01 tgl Exp $
 PostgreSQL documentation
 -->
 
@@ -300,6 +300,10 @@ COPY <replaceable class="parameter">count</replaceable>
     somewhat faster than the normal text mode, but a binary-format
     file is less portable across machine architectures and
     <productname>PostgreSQL</productname> versions.
+    Also, the binary format is very data type specific; for example
+    it will not work to output binary data from a <type>smallint</> column
+    and read it into an <type>integer</> column, even though that would work
+    fine in text format.
    </para>
 
    <para>
@@ -574,7 +578,7 @@ COPY <replaceable class="parameter">count</replaceable>
      In <literal>CSV</> mode, all characters are significant. A quoted value
      surrounded by white space, or any characters other than
      <literal>DELIMITER</>, will include those characters. This can cause
-     errors if  you import data from a system that pads <literal>CSV</> 
+     errors if you import data from a system that pads <literal>CSV</>
      lines with white space out to some fixed width. If such a situation
      arises you might need to preprocess the <literal>CSV</> file to remove
      the trailing white space, before importing the data into