Textpattern Support Forum

Very good job larry,

I wrote a pluggin dzd_counter_view to do the same but by using a custom field but you idea to use a log table is great!

Catalyst,

Well, i dont know how to help you, but i can tell you my txp setup, i’m using the Permanent Link Mode = /title and I’m Logging the Referers Only, maybe you could send me the output via email, and I’ll try to figure it out.

As always, feedback is appreciated.

I have the same problem. The result I get (as html) is :

<code>
<ul>
<li><a href=“http://www.tilkkusatu.net/”></a> (689)</li>
<li><a href=“http://www.tilkkusatu.net/”></a> (600)</li>
<li><a href=“http://www.tilkkusatu.net/”></a> (426)</li>
<li><a href=“http://www.tilkkusatu.net/”></a> (294)</li>
<li><a href=“http://www.tilkkusatu.net/”></a> (277)</li>
</ul>
</code>

setting are :

• Logging all hits
• permalink : year/month/day/title
• txp version : 4.0.3 (r1188)

I hope you’ll get this plugin working since it would be nice feature to have.

Edit:

Problem must be in these lines (with safe_row function) since $rs and $rs2 arrays seems to be empty.

<pre>
// Get URLs ordered by number of visits $rs = safe_rows($what,‘txp_log’, $where . $rest); // Extract article IDs foreach ($rs as $data) {

$rs2 = safe_rows(‘Title, url_title’, ‘textpattern’, ‘url_title = “’ . substr($data[“page”],1) . ‘”’); </pre>

I managed to get this working somehow when I changed it to output the $data[‘page’] variable instead $rs2… Could someone with better skills take a look at this?

Other thing is that I would like to make it filter everything else than articles. Any suggestions how this line should be changed to do that?

<pre> $where = ‘page <> “/” AND page <> “/index.php” ‘ . $range;</pre>

Last edited by pedro (2006-07-18 09:46:42)

me too …

any ideas at this time ??

This plug-in does not work for clean URLs.. on ly messy…

reason is… the first query pulls the page referred and counts the number of times it is referred.

the second search looks for the page referred dropping the leading slash… trouble is, the title of your article in the textpattern table is never going to include the whole URL!

I tried a few things – using strrpos() to find the last slash in the URL and then searching for matches… the trick with that is that you end up with a handful of popular links that are actually displaying… some examples of hits that would not register (popular thought they might be):

/section
/category
/?q=search+word

if it was possible to lock these out then maybe something could work, but I think the other idea of using an article tag replacement that increments a custom field works a lot better in this case.

Sorry for the bad news!