Very often I have heard people suggesting (and I have done it myself too a few times) to keep things separate: PHP code here, HTML there, external CSS, external JS and so on and so on.
Aside from the obvious readibility and maintenance advantages of doing this are there other strong advantages (e.g. in terms of server load or page processing time) in doing it?
As a trivial example, say we want to implement a table containing some products we read from a DB.
The output we want would be something like
<div class="description">This table lists all our products</div>
<table class="products">
<tr>
<th>Name</th>
<th>Available</th>
<th>Price</th>
</tr>
<tr>
<td>Prod 1</td>
<td>Yes</td>
<td>$100</td>
</tr>
...
...
</table>
<div class="info">Some generic info on the products here</div>
So here we have some static output (the 2 div
elements and the table header) and some dynamic output (the actual table content).
We could leave all the static things out of PHP tags and try to keep PHP only where needed
<div class="description">This table lists all our products</div>
<table class="products">
<tr>
<th>Name</th>
<th>Available</th>
<th>Price</th>
</tr>
<?
for ($p=0; $p<count($products); $p++)
{
echo '<tr>';
echo '<td>'.$products[$p]["name"].'</td>';
echo '<td>'.$products[$p]["availability"].'</td>';
echo '<td>'.$products[$p]["price"].'</td>';
echo '</tr>';
}
?>
</table>
<div>.....</div>
On the other hand we could embed everything in PHP
<?
echo '<div class="description">This table lists all our products</div>';
echo '<table class="products"><tr><th>Name</th>'.
'<th>Available</th><th>Price</th></tr>';
for ($p=0; $p<count($products); $p++)
{
echo '<tr>';
echo '<td>'.$products[$p]["name"].'</td>';
echo '<td>'.$products[$p]["availability"].'</td>';
echo '<td>'.$products[$p]["price"].'</td>';
echo '</tr>';
}
echo '</table>';
echo '<div>.....</div>';
What are the reasons to choose one over the other?