Omitting a folder with more than one sub-level

Kirby 3 Solved ✅
1

Hi there,
i have a route i am using to omit a top level folder. I got this from the docs on how to omit the blog/ folder:

'routes' => [
    [
        'pattern' => '(:any)',
        'action'  => function($uid) {
            $page = page($uid);
            if(!$page) $page = page('blog/' . $uid);
            if(!$page) $page = site()->errorPage();
            return site()->visit($page);
        }
    ],
    [
        'pattern' => 'blog/(:any)',
        'action'  => function($uid) {
            go($uid);
        }
    ]
]];

This works fine if there is just one level of subpages in the blog folder, but it doesn’t with more sublevels. Is there a way to employ the "Mighty mighty URI-Objects (I can’t find this article anymore on the kirby site) to do this?
I am searching for an expression that would say: ‘blog/’. page URI minus the first folder…

2

You can use the (:all) instead of the (:any) placeholder for a catch all route.

3

Thanks for the quick reaction. Changing from any to all didn’t really change anything on my end. I thought it has more to do with this line:
if(!$page) $page = page(‘blog/’ . $uid);
because it only constructs a url with “blog” and the next subfolder uid, while all the folders in between the current UID and “blog” get ignored…

4

Well, yes, of course you have to handle other pages as well, the easiest is probably this (not tested):

'action'  => function($uri) {
            $page = page($uri);
            if(!$page) $page = page('blog/' . $uri);
            if ($page) {
               return site()->visit($page);
            }
            $this->next();
        }
5

thanks, @texnixe. This doesn’t work either, and now i am starting to loose track on what it means.
but what i think shouldn’t work is this: ‘blog/’.$uri should create a url like this:
“blog/blog/currentpage” because the $uri object already contains the blog part, right? i was just wondering if there is a way to construct a path with “blog”.$uri-without-the-first-level.

But i am not sure, since i can’t really read and understand even the original blog function… I can only assume that the line
if(!$page) $page = page(‘blog/’ . $uid);
basically lays out the case of a subpage of “blog”.

6

now i got it to work, just by trying:

'routes' => [
    [
        'pattern' => '(:all)',
        'action'  => function($uid) {
        $page = page($uri); // $uri is not defined!!
        if(!$page) $page = page('blog/' . $uid);
        if ($page) {
           return site()->visit($page);
        }
        $this->next();
    }
    ],
    [
        'pattern' => 'blog/(:all)',
        'action'  => function($uid) {
            go($uri); // $uri is not defined!
        }
    ]
]];

seems to work on my end. i change the uri back to uid and it worked!!!

7

That doesn’t make sense, you mix both $uid and $uri now (see my comments above), you have to stick with the variable name you pass to the anonymous function, no matter what you call it.

8

The problem is the order of the routes when using the catch all, this should work:

    'routes' => [
        [
            'pattern' => 'notes/(:all)',
            'action'  => function($uid) {
                return go($uid);
            }
        ],
        [
            'pattern' => '(:all)',
            'action'  => function($uid) {
                $page = page($uid);
                if(!$page) $page = page('notes/' . $uid);
                if(!$page) $page = site()->errorPage();
                return $page;
            }
        ],
    ]
9

Thanks a lot, @texnixe
This seems to work as well, and i was already wondering about the error page bit…

It it always a pleasure to seek help in this forum. Thanks again!!!

1 Like