selecting entire rows as documents

endatabas · Jan 12, 2024 · c2ffe4f · c2ffe4f
1 parent 98c12a6
commit c2ffe4f
Show file tree

Hide file tree

Showing 5 changed files with 47 additions and 5 deletions.
diff --git a/docs/print.html b/docs/print.html
@@ -821,7 +821,8 @@ <h2 id="object"><a class="header" href="#object">OBJECT</a></h2>
 <h2 id="dynamic-literals"><a class="header" href="#dynamic-literals">Dynamic Literals</a></h2>
 <h3 id="row-literals"><a class="header" href="#row-literals">Row Literals</a></h3>
 <p>It is possible return an entire document (row) as a single literal value.
-The syntax is akin to Postgres <a href="https://www.postgresql.org/docs/current/rowtypes.html"><code>ROW</code> literals</a>.
+The syntax is akin to Postgres
+<a href="https://www.postgresql.org/docs/current/rowtypes.html"><code>ROW</code> literals</a>.
 Unlike <code>table.*</code>, which pads non-existent columns with <code>NULL</code>,
 a row literal returns exactly the schema specified for each individual row.</p>
 <ul>
@@ -830,6 +831,17 @@ <h3 id="row-literals"><a class="header" href="#row-literals">Row Literals</a></h
 <p>Example usage:</p>
 <pre><code class="language-sql">SELECT { products.* } FROM products;
 </code></pre>
+<p>As a shorthand, a table's name may be used in the <code>SELECT</code> clause to return
+entire rows as documents:</p>
+<pre><code class="language-sql">-&gt; SELECT users FROM users;
+-- [{'users': {'email': '[email protected]', 'name': 'Patrick'}},
+--  {'users': {'email': '[email protected]', 'name': 'Preethi'}}]
+</code></pre>
+<p>NOTE: When a table contains a column of the same name, the column takes precedence
+and a single column is returned, as usual:</p>
+<pre><code class="language-sql">-&gt; SELECT status FROM status;
+-- [{'status': 'error'}, {'status': 'ok'}]
+</code></pre>
 <h3 id="spread"><a class="header" href="#spread">Spread</a></h3>
 <p>The Spread Operator (<code>...</code>, sometimes known as &quot;splat&quot;)
 can be used to directly flatten/unnest one collection

diff --git a/docs/searchindex.js b/docs/searchindex.js
diff --git a/docs/searchindex.json b/docs/searchindex.json
diff --git a/docs/sql/data_types.html b/docs/sql/data_types.html
@@ -297,7 +297,8 @@ <h2 id="object"><a class="header" href="#object">OBJECT</a></h2>
 <h2 id="dynamic-literals"><a class="header" href="#dynamic-literals">Dynamic Literals</a></h2>
 <h3 id="row-literals"><a class="header" href="#row-literals">Row Literals</a></h3>
 <p>It is possible return an entire document (row) as a single literal value.
-The syntax is akin to Postgres <a href="https://www.postgresql.org/docs/current/rowtypes.html"><code>ROW</code> literals</a>.
+The syntax is akin to Postgres
+<a href="https://www.postgresql.org/docs/current/rowtypes.html"><code>ROW</code> literals</a>.
 Unlike <code>table.*</code>, which pads non-existent columns with <code>NULL</code>,
 a row literal returns exactly the schema specified for each individual row.</p>
 <ul>
@@ -306,6 +307,17 @@ <h3 id="row-literals"><a class="header" href="#row-literals">Row Literals</a></h
 <p>Example usage:</p>
 <pre><code class="language-sql">SELECT { products.* } FROM products;
 </code></pre>
+<p>As a shorthand, a table's name may be used in the <code>SELECT</code> clause to return
+entire rows as documents:</p>
+<pre><code class="language-sql">-&gt; SELECT users FROM users;
+-- [{'users': {'email': '[email protected]', 'name': 'Patrick'}},
+--  {'users': {'email': '[email protected]', 'name': 'Preethi'}}]
+</code></pre>
+<p>NOTE: When a table contains a column of the same name, the column takes precedence
+and a single column is returned, as usual:</p>
+<pre><code class="language-sql">-&gt; SELECT status FROM status;
+-- [{'status': 'error'}, {'status': 'ok'}]
+</code></pre>
 <h3 id="spread"><a class="header" href="#spread">Spread</a></h3>
 <p>The Spread Operator (<code>...</code>, sometimes known as &quot;splat&quot;)
 can be used to directly flatten/unnest one collection

diff --git a/src/sql/data_types.md b/src/sql/data_types.md
@@ -155,7 +155,8 @@ Object equality is tested by comparing each key-value pair as an array.
 ### Row Literals
 
 It is possible return an entire document (row) as a single literal value.
-The syntax is akin to Postgres [`ROW` literals](https://www.postgresql.org/docs/current/rowtypes.html).
+The syntax is akin to Postgres
+[`ROW` literals](https://www.postgresql.org/docs/current/rowtypes.html).
 Unlike `table.*`, which pads non-existent columns with `NULL`,
 a row literal returns exactly the schema specified for each individual row.
 
@@ -167,6 +168,23 @@ Example usage:
 SELECT { products.* } FROM products;
 ```
 
+As a shorthand, a table's name may be used in the `SELECT` clause to return
+entire rows as documents:
+
+```sql
+-> SELECT users FROM users;
+-- [{'users': {'email': '[email protected]', 'name': 'Patrick'}},
+--  {'users': {'email': '[email protected]', 'name': 'Preethi'}}]
+```
+
+NOTE: When a table contains a column of the same name, the column takes precedence
+and a single column is returned, as usual:
+
+```sql
+-> SELECT status FROM status;
+-- [{'status': 'error'}, {'status': 'ok'}]
+```
+
 ### Spread
 
 The Spread Operator (`...`, sometimes known as "splat")