read-csv-func.html

<!DOCTYPE html>
<html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">


<title>READ_CSV Function - SQL Notebook</title>
<link rel="stylesheet" href="sqlnotebook.css">
</head>
<body>
<header>
<table class="nav">
<tr>
<td>
<a href="index.html"><img src="art/SqlNotebookIcon.png" alt="SQL Notebook (logo)" style="width: 58px; height: 58px; float: left; margin-right: 20px;"></a>
</td>
<td>
<a href="index.html" id="title">SQL Notebook</a><br>
<nav>
<ul class="nav">
<li><a href="https://github.com/brianluft/sqlnotebook/releases">Download</a></li>
<li><a href="doc.html"><span id="header-doc-long">Documentation</span><span id="header-doc-short">Docs</span></a></li>
<li><a href="https://github.com/brianluft/sqlnotebook">GitHub</a></li>
</ul>
</nav>
</td>
</tr>
</table>
<hr style="margin-top: 15px; margin-bottom: 15px;">
</header>
<article><div id="article">
<h1><code>READ_CSV</code> Function</h1>
  <p>Reads a CSV file line-by-line and returns it in a table.</p>
  <p>This table-valued function is used in the <code>FROM</code> clause of a <code>SELECT</code> statement and can
  participate in joins as if it were a table. However, it cannot be used in a <code>CREATE TRIGGER</code> statement.
  Internally, the <code>READ_CSV</code> function is translated into an <code>IMPORT CSV</code> statement that runs
  prior to the statement that contains the <code>READ_CSV</code> call.</p>
  <h2>Syntax</h2><code>READ_CSV(<i>file-path</i>,&nbsp;</code><code>[<span style=
  "font-style: italic;">has-header-row</span>]</code><code>, [<span style=
  "font-style: italic;">skip-lines</span>],</code> <code>[<i>file-encoding</i>])</code><br>
  <h2>Parameters</h2>
  <ul class="args">
    <li><i>file-path</i>: text<br>
    The absolute path to the CSV file to read.</li>
    <li>
      <i>has-header-row</i>: integer (optional, 0-1, default: 1)<br>
      Indicates whether the CSV file begins with a column header line. If the file contains a column header but not on
      the first line of the file, use the <span style="font-style: italic;">skip-lines</span> parameter to indicate how
      many lines to skip before the column header appears.
      <ul class="enum">
        <li>0 = No column header. The generic column names <code>column1</code>, <code>column2</code>, etc. will be
        used.</li>
        <li>1 = A column header row exists.</li>
      </ul>
    </li>
    <li><i>skip-lines</i>: integer (optional, non-negative, default: 0)<br>
    Indicates how many initial lines should be skipped in the input file. This is used if the column header (or the
    data if there is no column header) does not appear on the first line of the file.</li>
    <li>
      <i>file-encoding</i>: integer (optional, 0-65535, default: 0)<br>
      Indicates the text encoding to use when reading the file. Specify 0 to detect the encoding automatically. Any
      nonzero integer is treated as a Windows code page number. See <a moz-do-not-send="true" href=
      "character-encodings-in-csv-and-text-files.html">Character Encodings in CSV and Text Files</a> for a list of
      these code page numbers.<br>
    </li>
  </ul>
  <h2>Return Value</h2>A table with&nbsp;columns defined by the input file.<br>
  <h2>Example</h2>
  <pre><i>-- Returns a table containing the first 50 rows in</i><i><br></i><i>-- "file.csv", with column names taken from the first line.</i><br>SELECT * FROM READ_CSV('C:\file.csv') LIMIT 50;<br><br><i>-- Returns a table containing the contents of "ShiftJIS.csv",</i><i><br></i><i>-- which is read using the Japanese Shift-JIS encoding (code</i><i><br></i><i>-- page 932).</i><br>SELECT * FROM READ_CSV('C:\ShiftJIS.csv', 1, 0, 932);<br><br></pre>
  <ul class="examples"></ul>
</div></article>
<footer><div id="footer">
<hr>
© 2016-2025 <a href="https://github.com/electroly">Brian Luft</a>
</div></footer>
</body>
</html>