On Fri, Apr 10, 2020 at 11:30:34AM -0400, Tom Lane wrote:
> Bruce Momjian <[email protected]> writes:
> > doc: add examples of creative use of unique expression indexes
> > https://git.postgresql.org/pg/commitdiff/a9760d0f3cb523336b5fdd9d6c5985e39a8588a1
>
> We had a complaint [1] that this dropped an example into the middle of
> two related paragraphs. I agree with that objection, and also notice
> that the extra example broke subsequent references to the "first example"
> and "second example". I'm also unhappy that the other addition that this
> commit made was dropped inside Example 11.3; if we're going to use
> <example> markup at all, each one ought to be a coherent entity.
>
> On top of that, I don't find that either example actually adds anything
> to the discussion, as the same points are being made in the existing
> text. Therefore, I don't think it's worth trying to fix these problems,
> and propose just reverting this patch.
>
> regards, tom lane
>
> [1]
> https://www.postgresql.org/message-id/158648685043.655.3074746555320970574%40wrigleys.postgresql.org
I agree with your analysis. I still want to have some mention that
partial indexes can be used to create single-NULL columns, which might
be required for compatibility with other databases. Attached is an
updated patch which removes the previous commit but adds a mention of
this.
--
Bruce Momjian <[email protected]> https://momjian.us
EnterpriseDB https://enterprisedb.com
+ As you are, so once was I. As I am, so you will be. +
+ Ancient Roman grave inscription +
diff --git a/doc/src/sgml/indices.sgml b/doc/src/sgml/indices.sgml
index 1be209a2fe..3a8f93bac7 100644
--- a/doc/src/sgml/indices.sgml
+++ b/doc/src/sgml/indices.sgml
@@ -705,15 +705,6 @@ CREATE INDEX test1_lower_col1_idx ON test1 (lower(col1));
</programlisting>
</para>
- <para>
- Expression indexes also allow control over the scope of unique indexes.
- For example, this unique index prevents duplicate integer values from
- being stored in a <type>double precision</type>-typed column:
-<programlisting>
-CREATE UNIQUE INDEX test1_uniq_int ON tests ((floor(double_col)))
-</programlisting>
- </para>
-
<para>
If we were to declare this index <literal>UNIQUE</literal>, it would prevent
creation of rows whose <literal>col1</literal> values differ only in case,
@@ -953,17 +944,9 @@ CREATE UNIQUE INDEX tests_success_constraint ON tests (subject, target)
WHERE success;
</programlisting>
This is a particularly efficient approach when there are few
- successful tests and many unsuccessful ones.
- </para>
-
- <para>
- This index allows only one null in the indexed column by using a
- partial index clause to process only null column values, and using
- an expression index clause to index <literal>true</literal> instead
- of <literal>null</literal>:
-<programlisting>
-CREATE UNIQUE INDEX tests_target_one_null ON tests ((target IS NULL)) WHERE target IS NULL;
-</programlisting>
+ successful tests and many unsuccessful ones. Creating a unique
+ index with an <literal>IS NULL</literal> qualification can restrict
+ a column to a single <literal>NULL</literal> value.
</para>
</example>
