Simple SEO URL (OLD version) [support thread]

[skaimauve]

If you re-read my original post you will see that I said that you can rebuild the full cPath from the category ID.

Right. My point anyway was that "you have to rebuild the full cPath" in addition to "you can rebuild the full cPath" which some mods forget to do.

And I forgot, nice code suggestion!

[CJPinder]

No, if I ever decide to do that, I wont use[FONT=monospace] zen_get_parent_categories because it will require mysql query. I will store the category tree in file which can be retrieved later.

Example:
17_19_21 will be stored in file t21
[/FONT]

Exactly, the code is there to show how the cPath can be rebuilt. Once it has been then cache it in a file or db table. Depending on the server setup, db caching may provide better performance then file caching, but maybe not.

Regards,
Christian.

[CJPinder]

Right. My point anyway was that "you have to rebuild the full cPath" in addition to "you can rebuild the full cPath" which some mods forget to do.

The best option for product URLs is to not include the cPath at all except on linked products which require special consideration for SEO anyway.

Regards,
Christian.

[yellow1912]

Some people prefer having the category name in the link, and even want to have more than 1 name. Now I dont know which one should I listen to, lol

[CJPinder]

Some people prefer having the category name in the link, and even want to have more than 1 name. Now I dont know which one should I listen to, lol

I meant remove the cPath except on linked products to help control duplicate content issues (and keep the URLs short). As for keyword stuffing in the URLs, well that is purely a personal preference and you will never please everyone

Regards,
Christian.

[skaimauve]

The best option for product URLs is to not include the cPath at all except on linked products which require special consideration for SEO anyway.

Exactly.

For my own education, what is a "linked product"?

[yellow1912]

A product that is in more than 1 categories. For instance, I have a book which is in both science and fiction categories.

[skaimauve]

Here is a patch for revision @123 to get the following URL style:

http://localhost/site/c2s6-mens-clothing/p4-shirt/product_info.html

diff includes/init_includes/init_ssu.php /www/site/includes/init_includes/init_ssu.php
31,33c31,32
<         $this->cache_folder = DIR_FS_SQL_CACHE."/ssu/";
<         $this->extension = SSU_FILE_EXTENSION;
<         $this->extension = trim($this->extension);
---
>         $this->cache_folder = DIR_FS_SQL_CACHE.'/ssu/';
>         $this->extension = '.'.trim(SSU_FILE_EXTENSION);
36c35,37
<         $this->mr = trim(($this->catalog_dir=='/' && empty($this->extension)) ? getenv('REQUEST_URI') : preg_replace(array($this->catalog_dir,'/\.'.$this->extension.'/'),'',getenv('REQUEST_URI'), 1),'/ ');
---
>         $this->mr = getenv('REQUEST_URI');
>         if(substr($this->mr,0,strlen($this->catalog_dir)) == $this->catalog_dir)
>             $this->mr = substr($this->mr,strlen($this->catalog_dir));
63,74c64,80
<         // first element is always main_page
<         // even if this is a "static" page, assigning first element to
<         $_GET['main_page'] = $this->mr_parts[0];
<         for($i= 1; $i < $this->mr_parts_count; $i++){
<             // TODO: a more efficient way to do this?
<             if(strstr($this->mr_parts[$i],'c'.SSU_ID_DELIMITER)){
<                 $cPath = explode(SSU_ID_DELIMITER, $this->mr_parts[$i]);
<                 $_GET['cPath'] = $cPath[count($cPath)-1];
<             }
<             elseif(strstr($this->mr_parts[$i],'p'.SSU_ID_DELIMITER)){
<                 $products_id = explode(SSU_ID_DELIMITER, $this->mr_parts[$i]);
<                 $_GET['products_id'] = $products_id[count($products_id)-1];
---
>         if($debug)
>         print_r($this->mr_parts);
>         // element with extension is always main_page 
>         for($i= 0; $i < $this->mr_parts_count; $i++){
>             $part = $this->mr_parts[$i];
>             // TODO: an more efficient way to do this?
>             $part_break = strpos($part,SSU_ID_DELIMITER);
>             if($part_break>0){
>               // Use cache to validate structure
>                 $file_name = substr($part,0,$part_break);
>                 if(file_get_contents($this->cache_folder.$file_name) != substr($part,$part_break)){
>                     if ($file_name[0] == 'c') 
>                         $_GET['cPath'] = str_replace('s','_',substr($file_name,1));                     
>                     elseif ($file_name[0] == 'p') 
>                         $_GET['products_id'] = substr($file_name,1);
>                 continue;
>                 }
75a82,84
>             if(substr($part,strlen($part)-strlen($this->extension )) == $this->extension ){
>                 $_GET['main_page'] = substr($part,0,strlen($part)-strlen($this->extension ));
>             }                    
77c86
<                 $_GET[$this->mr_parts[$i]] = $this->mr_parts[($i + 1)];
---
>                 $_GET[$part] = $this->mr_parts[($i + 1)];
81c90
<                 $_GET[$this->mr_parts[$i]] = '';
---
>                 $_GET[$part] = '';
111,113c120
<         $link = HTTP_SERVER;
<         if ($connection == 'SSL' && ENABLE_SSL == 'true')
<         $link = HTTPS_SERVER ;
---
>         $link = ($connection == 'SSL' && ENABLE_SSL == 'true') ? HTTPS_SERVER : HTTP_SERVER;
116,122c123,124
<         if ($use_dir_ws_catalog) {
<             if ($connection == 'SSL' && ENABLE_SSL == 'true') {
<                 $link .= DIR_WS_HTTPS_CATALOG;
<             } else {
<                 $link .= DIR_WS_CATALOG;
<             }
<         }
---
>         if ($use_dir_ws_catalog)
>       $link .= ($connection == 'SSL' && ENABLE_SSL == 'true') ? DIR_WS_HTTPS_CATALOG : DIR_WS_CATALOG;
171,182c173,180
<         // Any param? No? return page.extension
<         if(empty($parameters)){
<             if(empty($sid))
<             return $link.$page.(trim(SSU_FILE_EXTENSION)=='' ? '' : '.'.SSU_FILE_EXTENSION);
<             else
<             return $link.$page.'/'.$sid;
<         }
< 
<         $parameters = zen_output_string($parameters);
<         $link .= $page.'/'.$parameters;
< 
<         while (substr($link, -1) == '/') $link = substr($link, 0, -1);
---
>         // Any param? 
>         if(!empty($parameters)){
>         $parameters = zen_output_string($parameters);
>           $link .= $parameters.'/';
>         }
>         
>         // return page.extension
>       $link .= $page.((trim(SSU_FILE_EXTENSION)=='') ? '' : '.'.trim(SSU_FILE_EXTENSION)); 
200c198
<         return 'c'.SSU_ID_DELIMITER.$content.SSU_ID_DELIMITER.$cPath;
---
>         return $file_name.SSU_ID_DELIMITER.$content;
225c223
<         return "c$cPath";
---
>         return 'c'. str_replace('_','s',$cPath);
227c225
< 
---
>     
234c232
<         return 'p'.SSU_ID_DELIMITER.$content.SSU_ID_DELIMITER.$products_id;
---
>         return $file_name.SSU_ID_DELIMITER.$content; 

[yellow1912]

Here is what I'm thinking:
we either: remove cpath completely from product_info links, rebuild later
or: we make sure the cPath will always be the same, and always be there for a specific product.

[skaimauve]

My take is that we've disquised cPath in the URL so far to make things prettier while and not sacrifice speed. If we can do without cPath in the URL, then great!

The question is to choose the ultimate URL format. The code will follow.