Change unknown-type literals to type text in SELECT and RETURNING lists.

Previously, we left such literals alone if the query or subquery had
no properties forcing a type decision to be made (such as an ORDER BY or
DISTINCT clause using that output column).  This meant that "unknown" could
be an exposed output column type, which has never been a great idea because
it could result in strange failures later on.  For example, an outer query
that tried to do any operations on an unknown-type subquery output would
generally fail with some weird error like "failed to find conversion
function from unknown to text" or "could not determine which collation to
use for string comparison".  Also, if the case occurred in a CREATE VIEW's
query then the view would have an unknown-type column, causing similar
failures in queries trying to use the view.

To fix, at the tail end of parse analysis of a query, forcibly convert any
remaining "unknown" literals in its SELECT or RETURNING list to type text.
However, provide a switch to suppress that, and use it in the cases of
SELECT inside a set operation or INSERT command.  In those cases we already
had type resolution rules that make use of context information from outside
the subquery proper, and we don't want to change that behavior.

Also, change creation of an unknown-type column in a relation from a
warning to a hard error.  The error should be unreachable now in CREATE
VIEW or CREATE MATVIEW, but it's still possible to explicitly say "unknown"
in CREATE TABLE or CREATE (composite) TYPE.  We want to forbid that because
it's nothing but a foot-gun.

This change creates a pg_upgrade failure case: a matview that contains an
unknown-type column can't be pg_upgraded, because reparsing the matview's
defining query will now decide that the column is of type text, which
doesn't match the cstring-like storage that the old materialized column
would actually have.  Add a checking pass to detect that.  While at it,
we can detect tables or composite types that would fail, essentially
for free.  Those would fail safely anyway later on, but we might as
well fail earlier.

This patch is by me, but it owes something to previous investigations
by Rahila Syed.  Also thanks to Ashutosh Bapat and Michael Paquier for
review.

Discussion: https://postgr.es/m/CAH2L28uwwbL9HUM-WR=hromW1Cvamkn7O-g8fPY2m=_7muJ0oA@mail.gmail.com

doc/src/sgml/ref/create_view.sgml

@@ -251,9 +251,10 @@ CREATE VIEW [ <replaceable>schema</> . ] <replaceable>view_name</> AS WITH RECUR
 <programlisting>
 CREATE VIEW vista AS SELECT 'Hello World';
 </programlisting>
-    is bad form in two ways: the column name defaults to <literal>?column?</>,
-    and the column data type defaults to <type>unknown</>.  If you want a
-    string literal in a view's result, use something like:
+    is bad form because the column name defaults to <literal>?column?</>;
+    also, the column data type defaults to <type>text</>, which might not
+    be what you wanted.  Better style for a string literal in a view's
+    result is something like:
 <programlisting>
 CREATE VIEW vista AS SELECT text 'Hello World' AS hello;
 </programlisting>

doc/src/sgml/typeconv.sgml

@@ -984,7 +984,8 @@ domain's base type for all subsequent steps.
 <para>
 If all inputs are of type <type>unknown</type>, resolve as type
 <type>text</type> (the preferred type of the string category).
-Otherwise, <type>unknown</type> inputs are ignored.
+Otherwise, <type>unknown</type> inputs are ignored for the purposes
+of the remaining rules.
 </para>
 </step>
 
@@ -1076,6 +1077,53 @@ but <type>integer</> can be implicitly cast to <type>real</>, the union
 result type is resolved as <type>real</>.
 </para>
 </example>
+</sect1>
+
+<sect1 id="typeconv-select">
+<title><literal>SELECT</literal> Output Columns</title>
+
+<indexterm zone="typeconv-select">
+ <primary>SELECT</primary>
+ <secondary>determination of result type</secondary>
+</indexterm>
+
+<para>
+The rules given in the preceding sections will result in assignment
+of non-<type>unknown</> data types to all expressions in a SQL query,
+except for unspecified-type literals that appear as simple output
+columns of a <command>SELECT</> command.  For example, in
+
+<screen>
+SELECT 'Hello World';
+</screen>
+
+there is nothing to identify what type the string literal should be
+taken as.  In this situation <productname>PostgreSQL</> will fall back
+to resolving the literal's type as <type>text</>.
+</para>
+
+<para>
+When the <command>SELECT</> is one arm of a <literal>UNION</>
+(or <literal>INTERSECT</> or <literal>EXCEPT</>) construct, or when it
+appears within <command>INSERT ... SELECT</>, this rule is not applied
+since rules given in preceding sections take precedence.  The type of an
+unspecified-type literal can be taken from the other <literal>UNION</> arm
+in the first case, or from the destination column in the second case.
+</para>
+
+<para>
+<literal>RETURNING</> lists are treated the same as <command>SELECT</>
+output lists for this purpose.
+</para>
+
+<note>
+ <para>
+  Prior to <productname>PostgreSQL</> 10, this rule did not exist, and
+  unspecified-type literals in a <command>SELECT</> output list were
+  left as type <type>unknown</>.  That had assorted bad consequences,
+  so it's been changed.
+ </para>
+</note>
 
 </sect1>
 </chapter>