Leaderboard

Hi all, Just wanted to share a tip that can be useful if you want to debug a core file before Tracy has been loaded. Suppose you were debugging ProcessWire::getHttpHost() and for some reason wanted to dump the $host variable just before it is returned. If you add the line... bd($host, 'host'); ...then you'll get a fatal error when the code runs because Tracy has not yet been loaded and so the bd() function is undefined. When debugging core files you'll find a number of places like this where the Tracy dump functions are not available. A workaround for this is to set the variable that you want to dump as a custom property on the Wire object. You want to be careful not to use a property name that clashes with any existing Wire properties or API variables - adding a prefix like "td_" is a safe bet. $this->wire()->set('td_host', $host); Now in /site/ready.php you can dump the variable that you stored on the Wire object: bd($wire->td_host, 'host'); ...or shorter (thanks @adrian)... bd($td_host, 'host');

Really great idea @Robin S ! You can make it even less work by using d($td_host) in the Console panel:

@Robin S - what do you think about this idea - I am contemplating having Tracy check $wire for any variables prefixed with "bd_" and automatically bd() them without the need to manually make a bd() or d() call. The title would be automatically set to the name of the variable after the "bd_" prefix. PS - remember you can set $wire variables like this: $this->wire('bd_host', $host); The shorter the better when adding debug statements ?

webp support is added in PW 3.0.132

@pwired — there was a time when I'd agree with you, but there are quite a bit of options for I ❤️ PW devs out there - for flat file CMS at least. So in the end I'd say that ProcessWire and its community are better served by focusing at its one thing.

Just tested it locally in two different ProcessWire instances and it worked as expected. Used my admin user and I saw the images, was able to edit/delete them. ProcessWire 3.0.133 and 3.0.98. Can you actually view these images by taking their URLs? Maybe it's just a cache or browser glitch in the backend somehow. Are there 404 or other warnings in the console > network tab?

I've created two identical pages, with only the page-name being different. An ordinary query like $res = $pages->find("template=basic-page, title|summary|body|sidebar%=muster"); d($res); correctly returned two matches. The entire DB is innoDB as well. So that surely can't be the culprit.

Can you post your selector query? What version of ProcessWire are you using? Is one of the pages hidden? Is there something special with your pages or templates? Are these pages not visible to guests or admin users or everyone?

You're welcome! I wish I could help more.

hahaha thanks a lot for you help and patience! I just pinged on an nginx thread to check if anyone's got some more info.

Care to elaborate about such projects? i.e. some insights into the pros/cons of PW, or just general infos (what worked well, what didn't; what general dev route you were taking instead, bottlenecks or overhead etc).

@bernhard's post gives you the clue - it's to do with PW's built-in caching. $page->children() is a method that queries the database. You want to minimise the number of requests to the database because going to the DB is a lot slower than working with objects in memory. In your code you call $page->children() five times, so the expectation is that this would mean five trips to the DB... except that PW tries to help you out and caches the results to avoid unnecessary DB queries. But the exact workings of this caching are undocumented (as far as I know) and so I'd say it's best practice not to rely on it. It would be better to consciously minimise the DB requests in your code - it would make things clearer (i.e. avoid the confusion you're getting from the behind-the-scenes caching) and it's a good habit to get into for when you're working with other PHP projects that might not cache DB query results like PW does. You could rewrite your code like this: $layout->setTemplate('sidebar'); $selector = [ 'template' => 'product', 'sort' => 'title', 'categories' => page('id'), 'limit' => 9 ]; // Just get the children from the DB once // I assume you need all the children at some point later in the code because you dump page()->children() in your example. // If you don't need all the children then just get the categories directly with page()->children('template=category') $children = page()->children(); // Get the categories from the children PageArray (happens in memory) $categories = $children->find('template=category'); // If there are any categories then modify $selector if($categories->count) { $categories->add(page()); $selector['categories'] = $categories; } bd($children); bd($categories); And this might be getting into the zone of micro-optimisation but you could consider if you actually need full Page objects in a PageArray or if you only need the IDs (reduces the memory footprint): $category_ids = page()->children('template=category', ['findIDs' => 1]);; if(count($category_ids)) { $category_ids[] = page()->id; $selector['categories'] = $category_ids; }

@CrazyEnimal, please insert your code inside a code block in forum posts. You can use SQL in ProcessWire when it suits you. That's what the $database API variable is for. Here is one way you could get a listing of manufacturers with the number of occurrences within a selection of cars. // Get the IDs of the cars $car_ids = $pages->findIDs("parent=$parent, template=page_car, dealer=$dealer_id"); // Get table for manufacturer Page Reference field $table = $fields->get('manufacturer')->getTable(); // Get manufacturers that are selected in the car pages $query = $database->query("SELECT data FROM $table WHERE pages_id IN (" . implode(',', $car_ids) . ")"); $manufacturer_ids = $query->fetchAll(\PDO::FETCH_COLUMN); // Count how many times each manufacturer occurs in the results $manufacturer_occurrences = array_count_values($manufacturer_ids); // Sort the results in order of occurrences, highest to lowest arsort($manufacturer_occurrences); // Get the manufacturer pages $manufacturers = $pages->getById(array_keys($manufacturer_occurrences)); // Output a list of manufacturer titles and the number of occurrences foreach($manufacturers as $manufacturer) { echo "<p>{$manufacturer->title} ({$manufacturer_occurrences[$manufacturers->id]})</p>"; }

Actually, it's pretty nice running your API call through the Tracy Console - save it as a snippet so you always have it on hand. This way you can easily see the results of the selector as well as the SQL query used to generate them. PS I think this is the thread that was mentioned regarding teppo's version: https://processwire.com/talk/topic/9408-is-there-a-way-to-convert-a-selector-in-sql-using-pw-engine/ although I think it's actually longer.