BrendonKoz

That was likely a conscious decision with their htaccess rules to prevent numerically traversing an entire forum (and instead requires parsing HTML to identify internal hyperlinks, so more processing power on the part of bad bots). Good find though! I much prefer shorter links. ?

I somehow managed to miss this reply. Thank you for both of your replies, @Robin S! I did manage to get this working, with some oddities here and there that had to be worked through due to how I was checking/setting access permissions. I chose not to explicitly set permission for each and every template per role since I'm using a field that's added to the templates to determine access, so it seemed doubly difficult to do that, and my hope was that inheritance would work properly most of the time. So far that's fairly true, with occasional exceptions (ex: nested repeaters since repeaters have their own system templates). It felt a little hack'ish but I checked against the template name and if strpos() found a match to "repeater_" against $page->template->name, then I'd allow access to the edit method (add didn't cause an issue the way I had it written). I wasn't sure how else to check if a template of a page being checked for access belonged to a repeatable fieldtype.

Disabling/Uninstalling SessionHandlerDB is supposed to log out any users - did you get logged out prior to seeing that message? I should've mentioned it would be a good idea to back up the database prior to disabling, just in case, but since you just moved hosts, I'd imagine whether you did or did not get a backup prior to disabling the module, you should have the ability to restore that table and its data, if need be - and/or restore to your prior host's backup state. I know a lot of people have had issue with the SessionHandlerDB, which is why I think it's a good idea to see if running the site with it disabled might help. Unfortunately, you also want to be able to use your site, and it seems something is holding on to a cache and/or expecting the table to exist. I'm not familiar with all of the modules you've mentioned, so maybe one of them needs it too?

I haven't personally messed much with the table salt things, but usually that's related to passwords. If you can log in successfully, I'm thinking that's OK (and I would not change it). If you have a feeling there may be an issue with SessionHandlerDB, try disabling it temporarily to see what happens.

I tried accessing your website (and the problematic page) last night and also encountered the error. The error message that I received was about a proxy server having issue, however. Is the current host using nginx or Apache? If there's a proxy server somewhere in there (if not nginx) then it's going to be beyond my experience/capabilities to know how to help. Otherwise, maybe look into why that particular page section (Newsboard) seems to cause trouble when other pages mostly load rather quickly. ProDevTools' ProfilerPro might be useful here, if identifying problem areas on your own ends up being difficult to find.

Are you calling your create() method from the constructor?

Thanks - that did it! Looks like forcing "isLocal" was what I needed here. In this case it's not a guest, though I'm guessing the permission level is equivalent to guest. (I haven't explicitly added any tracy related permissions to roles.)

I had thought I had Tracy enabled for all users (excluding guests) awhile back and disabled it. Now I can't seem to re-enable it now that I need to diagnose some things. I am not restricting non-superusers (or superusers for that matter), the Tracy Debugger is enabled, and I even tried to match the IP address using PCRE pattern of /.*/ with no luck.

I'm responding with no experience using custom page classes, so if my reply is just silly, I apologize. ? Perhaps I'm missing something, but I think my question here is also the question that ProcessWire is throwing an error about. Where do your ProcesswireTemplate::INDIVIDUAL_CAR_CHOICES and ProcesswireTemplate::INDIVIDUAL_CAR_CHOICE variables' values come from? In fact, where is the ProcessWireTemplate class located? There's a Template class in wire/core, but I'm not seeing a Processwire* class anywhere. If those classes/values don't exist, that could cause the errors to be thrown. If you're moving the code to another class, maybe it's not actually being called/used and ProcessWire is falling back to using the standard Page class?

This is really neat. If the action relates to a topic, can the topic name be linked to the actual topic? (I saw people viewing topics that I hadn't read in a long time and recall it being interesting; currently have to go back to the forum and search for it). Related to PWGeeks - is the "last updated" status the last time the related project was (determined to be) updated, or the last time PWGeeks checked the project's status? I ask because the very last page (279 as of right now) shows a last update time based on my current clock time. Ordering by "Recently Modified" shows FieldtypeFolderOptions on the first paginated page, but was last updated (Github) 4 years ago. I love the PWGeeks project, btw. ❤️

So far my solution to this seems unnecessarily complex. I have a primary RepeaterMatrix field (`content_body`). I've duplicated that field and called it "content_column". I created a Repeater (standard) field, added a simple number field (for column width, for now) and the "content_column" field. I've added "content_row" to "content_body" as a new item. I'm using field-based file rendering, so I have a file structure such as: fields: - content_body.php - content_body/field1.php - content_body/field2.php ...etc... So now I've got to duplicate the files for rendering the contents of the "content_column" RepeaterMatrix, which is basically just a copy/paste of all the individual fields of the "content_body" folder, and maintain changes between the duplicated files. I tried to add "content_body" to "content_row" as a subfield, but I kept getting 503 errors in the admin and ultimately had to revert the database to recover. ?

My difficulties were in understanding how to set up the fields to allow such a scenario, but not make it ridiculously complicated to maintain, or for the editing end-users. I suppose offering a numerical value for the column size (ex: 2-10) per matrix item would also be a valid solution, and far more customizable. As silly as it sounds, the reason I was thinking of offering varied custom sizes is because some of our users are extremely tech-averse, and will mess that up. ?

I have a single RepeaterMatrix field that I'm using to generate a bunch of available components. Most of the remaining components left to create are intended for columnar-layout options...in this case, equal width 2-column, 3-column, and 4-column options, as well as a 30/70, and 70/30 split. Ultimately I'd prefer to offer just the column option, and leave the content of the columns up to the end-user, rather than locking it into a "1 image, 1 textarea" style option. I'm having a hard time coming up with a way to offer the dynamic option, while also limiting the column count, and am curious how others might've tried solving this issue. It can't be all that unique! EDIT: Looks like I might (manually) recreate my RepeaterMatrix field and add it into a FieldsetOpen field along with a few settings (fields), such as a min/max number (2-4) and base the layout on that for the frontend. I don't need to force the columns to match the numbers of the column setting, the components would just flow into the number of columns as needed. Definitely not the most efficient, I'm sure, but it should work. It's too bad (at least without RockMigrations) that I can't easily create a duplicate of a RepeaterMatrix field! ?

Verified: It was our webhost's (Dreamhost) customized, and overly intrusive mod_security settings.

@lpa Reading over the ProcessWire Weekly, apparently there was definitely an issue that was fixed in the recent dev version. https://weekly.pw/issue/514/#latest-core-updates

Just a passing thought without testing or checking, but it looks like banner_file and video_file are string-based values, whereas is_html5 and html are boolean-based (true/false). If that's correct, it's possible that some stricter checking needed to be done for PHP 8.x compatibility, and you might need to break that selector up into multiple parts.

It seems as though ProcessPageAdd::getAllowedTemplates is called so many times that it doesn't actually always have a $parent value on each call (because of how my authentication is configured). I had to test for null on that value, and exit gracefully if null, and if not then I could enter into my logic. The manual assignment of values worked because it forced a value each time regardless of what was passed into the method as a $parent value for the $event object. EDIT: This now works for a superuser account, but not for a standard account. It seems there's something else in permissions that I'm missing. Updated hook method, without any added business logic of my own, is now: public function allowTemplates($event) { // Not all calls pass in an argument value if (is_null($event->arguments('parent'))) return; // Identify the intended parent template and fetch its allowed (template) children $parent_t = $this->wire->templates->get($event->arguments('parent')->template); $child_t = $parent_t->childTemplates; // Assign the allowed children to an array we can set to return $allowed_templates = []; $allowed_templates[$parent_t->id] = $parent_t; foreach ($child_t as $t) { $t = $this->wire->templates->get($t); $allowed_templates[$t->id] = $t; } // I like sorted values ^_^ ksort($allowed_templates); // Return the expected template array back to the calling process $event->return = $allowed_templates; } If I give "Edit Pages", and "Create Page" permission to the role for the child template, and "Add Children" permission to the parent template for the role, then I can create a Job Listing (child) under the Job List (parent). However, once I do that, everyone with the role can successfully do this, despite me trying to send FALSE (or some similar value) back through any available hooked method. I can't seem to override a strict parent/child relationship.

Not entirely - we haven't yet identified what on our webhost (Dreamhost) caused the file to become corrupted and only upload a portion of the file. What we do know, however, is that it's not a fault of ProcessWire. ? Because it's only occurred on one PDF for all the files we upload/share, it's entirely possible there was also some issue with the file itself. I'm going to wait to follow up with our host if/when we discover another file that shows this symptom. I appreciate the follow-up. ?

Backstory: I'm working on expanding access to permissions via a specific role, by using a FieldTypeTags on both user accounts and pages alike, thereby giving me (and others) the ability to control access per-page. The logic therein hasn't been fully fleshed out, but I keep getting stuck; for instance: on the Page List (admin home) screen, the "Add" button on hover shows up, but clicking on it threw an error that I didn't have permission (it was a parent-child relationship that the child was only allowed on that parent, and there's only ever allowed to be one parent). The role doesn't have page-create (or page-add) permission because the only template that is currently access-controlled is the home template...so theoretically, all templates can be created (I'd think?). I was having trouble where template permissions were seemingly overriding my hooks that were supposed to deny access, so I removed all but "home" role permissions. Anyhow... After finding Robin's TemplatesChildPages module it let me find a hook that I could use to forcibly override template permissions (saving/publishing does work). Unfortunately, I've only been able to get the "Add New" form to render when I've created the array in the ProcessPageAdd::getAllowedTemplates (after) hook when I create the return value's (template) array manually. (In this case it's a "job-list" parent template [indexes employment offerings], and 'job-listing' child template that describes the employment position.) So the following code allows the form to render and work as expected: $this->addHookAfter('ProcessPageAdd::getAllowedTemplates', $this, 'allowTemplates'); // ... public function allowTemplates($event) { $p = $this->wire->templates->get('job-list'); $t = $this->wire->templates->get('job-listing'); $event->return = [ $p->id => $p, $t->id => $t ]; ksort($event->return); } Obviously that will only work for my explicit example above while I test adding that particular template. So I went back and adjusted the hook to make it dynamic. Unfortunately, the following code throws an exception. public function allowTemplates($event) { $parent_t = $this->wire->templates->get($event->arguments('parent')->template); $child_t = $parent_t->childTemplates; $allowed_templates = []; $allowed_templates[$parent_t->id] = $parent_t; foreach ($child_t as $t) { $t = $this->wire->templates->get($t); $allowed_templates[$t->id] = $t; } ksort($allowed_templates); $event->return = $allowed_templates; } I've taken the $event_return value and processed it through a good old fashioned echo '<pre>'; var_dump($event->return); die('</pre>'); then taken the output (copy/paste) and saved it to a file, for each version above. I used WinMerge (to quickly compare) and the files are identical. The exception that is thrown, for the dynamic version, reads: ProcessWire: ProcessPageAdd: Template &quot;job-listing&quot; is not allowed here. But, as far as I can tell, the parent and child values exist in $event->return after my hook is run in both examples...yet only one seems to be working. Does anyone have any possible thoughts on this??

Looks like @Robin S had similar thoughts. ? https://processwire.com/modules/verify-links/

I can't seem to reproduce it on my development server (mirrored PW website), so it must be something related to the webhost. Thanks for having me take a closer look.

I am still using the outdated Xenu Link Sleuth to run a spider from my desktop against all links. It's fairly thorough, and you're essentially hitting your site with a decent amount of traffic all at once, but it mostly works (it still checks many external links - one level deeper than it should - even though I tell it not to). I do also have Ryan's FieldtypeVerifiedURL in use, but that - on its own - only checks links held within that fieldtype. A modern version of the same tool would be Screaming Frog's Link Checker. There is a free version, but it only checks up to 500 links. I run a check once a month for a single organization's web properties; it's a bit pricey for a once-a-month run for us, unfortunately. Otherwise, it seems like the best modern software-based option out there for the task.

I think the simplest would be to take advantage of some additional modules that can make this easier for you. Hanna Code (and add instructions for users in the CKeditor Notes field) TextFormatterOembed Or use a separate field, like FieldtypeOembed (since you'd usually want a video as either a link, or a full column fill anyway). Otherwise, you'd likely need to find, install, and configure custom CKeditor plugins that would handle this for you. This is also possible to do, but since the above is pretty easy, I didn't quite look this method up. ?

That might be another thing to test then, too - see if the automatic ZIP upload and decompression setting of a file field that ended up getting used in a, let's say FormBuilder form, which then creates pages for the user...would allow a PHP file to be created, and accessible, by a nefarious site user. In hindsight it seems silly to allow such a process to occur, but when you're so focused on the intended uses of something like this, the unintended sometimes is missed. (ex: Having a local writer's community website managed by ProcessWire, and allowing them to upload ZIP archives of text, html, photos, markdown, etc., to then be published for others to read/see - and then overlooking the potential for an HTML file to reference an external site via iframe that does some fun things - it's not specifically a ProcessWire issue, it's an oversight in development)

Has anyone experienced a scenario where a PDF (or other) file that's been uploaded, and shows as having been successfully uploaded, is reporting as an improper size, and then doesn't load correctly when attempted to be accessed/downloaded/viewed from the website? I have a PDF that, I believe, was generated from Canva. It is 4.5MB in size. During upload, the JS upload GUI reports the proper size in the progress bar, but then once complete, shows a filesize that is not the same as the original (2.5MB in this case). Unfortunately, it also seems to break and corrupt the file. It is indeed smaller. Thus far, I've manually uploaded, via FTP, the file overtop of the one processed by ProcessWire, but if this comes up again, I'm not sure what to look at. I have FileValidatorSVG installed, but the file extension here is definitely PDF. Also installed that might process files: WireRequestBlocker, SearchEngine File Indexer, WebP to JPG. I thought I'd recently seen a topic about this, but when trying to find it again, was unable.