Question

I have the following code which works fine to format text from an SQL table. It seems a little long winded though.

It will create paragraphs from the line breaks but ignore header and list tags (not wrap those in “p” tags.

Can anyone see an obvious way to condense this?

<?php

function format_html($content)
 {
  $content = str_replace("<h1>\r\n", "<h1>", $content);
  $content = str_replace("</h1>\r\n", "</h1><p>", $content);
  $content = str_replace("<h2>\r\n", "<h2>", $content);
  $content = str_replace("</h2>\r\n", "</h2><p>", $content);
  $content = str_replace("<h3>\r\n", "<h3>", $content);
  $content = str_replace("</h3>\r\n", "</h3><p>", $content);
  $content = str_replace("<h4>\r\n", "<h4>", $content);
  $content = str_replace("</h4>\r\n", "</h4><p>", $content);
  $content = str_replace("<h5>\r\n", "<h5>", $content);
  $content = str_replace("</h5>\r\n", "</h5><p>", $content);
  $content = str_replace("<h6>\r\n", "<h6>", $content);
  $content = str_replace("</h6>\r\n", "</h6><p>", $content);
  $content = str_replace("<ul>\r\n", "<ul>", $content);
  $content = str_replace("</ul>\r\n", "</ul><p>", $content);
  $content = str_replace("<ol>\r\n", "<ol>", $content);
  $content = str_replace("</ol>\r\n", "</ol><p>", $content);
  $content = str_replace("<li>\r\n", "<li>", $content);
  $content = str_replace("</li>\r\n", "</li>", $content);
  $content = "<p>" . str_replace("\r\n", "</p><p>", $content);
  $content = str_replace("<p><h1>", "<h1>", $content);
  $content = str_replace("<p><h2>", "<h2>", $content);
  $content = str_replace("<p><h3>", "<h3>", $content);
  $content = str_replace("<p><h4>", "<h4>", $content);
  $content = str_replace("<p><h5>", "<h5>", $content);
  $content = str_replace("<p><h6>", "<h6>", $content);
  $content = str_replace("<p><ul>", "<ul>", $content);
  $content = str_replace("<p><ol>", "<ol>", $content);
  return $content;
 }

function format_html_end($content)
 {
  $content = str_replace("</h1></p>", "</h1>", $content);
  $content = str_replace("</h2></p>", "</h2>", $content);
  $content = str_replace("</h3></p>", "</h3>", $content);
  $content = str_replace("</h4></p>", "</h4>", $content);
  $content = str_replace("</h5></p>", "</h5>", $content);
  $content = str_replace("</h6></p>", "</h6>", $content);
  $content = str_replace("</ul></p>", "</ul>", $content);
  $content = str_replace("</ol></p>", "</ol>", $content);
  return $content;
 }

?>

<?php
$con = mysql_connect("localhost","username","password");
if (!$con)
  {
  die('Could not connect: ' . mysql_error());
  }

mysql_select_db("db", $con);

$result = mysql_query("SELECT column FROM table WHERE id = '1'");

while($row = mysql_fetch_array($result))
  {
  $content = $row['column'];
  echo format_html_end(format_html("$content</p>"));
  }

mysql_close($con);
?>

The content from the table will look something like this…

<h1>Header</h1>
ertertert
ertertertert
rhdfgh
dfghdfghdfgh
ddfgh
<ul>
<li>fdghdfghd</li>
<li>fghjfghj</li>
</ul>

Answer 1

You can deal with nearly all of that with some regular expressions:

$content = preg_replace("/<(h[1-6]|ul|ol)>\r\n/", "<$1>", $content);
$content = preg_replace("/<\/(h[1-6]|ul|ol)>\r\n/", "</$1><p>", $content);
$content = preg_replace("/<(\/?)li>\r\n/", "<$1li>", $content);
$content = preg_replace("/<p><(h[1-6]|ul|ol)>/", "<$1>", $content);
$content = preg_replace("/<\/(h[1-6]|ul|ol)><\/p>/", "</$1>", $content);

The trick with these is that you can use capturing and backward references when doing the replacement. So for example the first regexp can match h1-h6, ul or ol, and during replacing $1 has the value of whichever of those it matched.

The following line of code I would leave as is, since it doesn’t have anything in common with the other regular expressions, and works fine.

$content = "<p>" . str_replace("\r\n", "</p><p>", $content);