Search the Community

Hi everyone, Yesterday I began working on a module to create a filesystem abstraction for files and images using the Flysytem library (http://flysystem.thephpleague.com), out of the necessity of having the images copied (and probably linked) on Amazon S3 and other places like Dropbox. There two reasons why I decided to tackle this: 1 - When I work on the project on my machine, I need a way to automatically sync the images from the server and/or the external storage. When an image is added on the server and the database is imported on my local env, PW shows a blank thumbnail of it. The idea for the module if to check if the page image has a width == 0 and if it exists on the server, add it to the local filesystem. 2 - In the past, I had to move a large website to a different server in a hurry and a lot of images were left behind (it was a mess). So I'm planning for a possible future worst-case scenario of the server exploding ? The code I quickly wrote is below (please bear with me that it's pretty raw at the moment). One thing I had to figure it out is why PW fires the Pageimage::size hook wherever a page is loaded in the admin, even though the thumbnails are already created. I was planning to save the image variations on S3 as well. Can someone clarify? I know that @LostKobrakai was working on a similar project, and so I would like to ask him and everyone else if you think I (and who may help) should evolve this idea into a full-featured module where the user can select which server (adapter) to use (AWS, Digital Ocean spaces, Dropbox etc.) <?php namespace ProcessWire; use Aws\S3\S3Client; use League\Flysystem\AwsS3v3\AwsS3Adapter; use League\Flysystem\Filesystem; use League\Flysystem\Adapter\Local; use Spatie\Dropbox\Client; use Spatie\FlysystemDropbox\DropboxAdapter; class ProcessFly extends WireData implements Module, ConfigurableModule { public static function getModuleInfo() { return array( 'title' => 'ProcessWire Flysystem Integration', 'version' => 001, 'summary' => 'Synchronize all the page assets uploaded through PW to a specified bucket in Amazon S3 and...', 'author' => 'Sérgio Jardim', 'singular' => true, 'autoload' => true, 'icon' => 'image' ); } public function init() { $this->client = S3Client::factory([ 'credentials' => [ 'key' => '', 'secret' => '', ], 'region' => 'sa-east-1', 'version' => 'latest', ]); $this->bucket_name = ""; $this->dir_name = "images"; $this->s3_adapter = new AwsS3Adapter($this->client, $this->bucket_name, $this->dir_name); $this->s3_filesystem = new Filesystem($this->s3_adapter); // DROPBOX $this->authorizationToken = ""; $this->dropbox_client = new Client($this->authorizationToken); $this->adapter_dropbox = new DropboxAdapter($this->dropbox_client); $this->dropbox_filesystem = new Filesystem($this->adapter_dropbox); $this->addHookAfter('Pages::saved', $this, 'checkImageS3'); // Download images that are not in local filesystem for whatever reason but are available in the remove one. $this->addHookAfter('InputfieldFile::fileAdded', $this, 'uploadImageS3'); // Fired when a file/image is added to a page in the admin // $this->addHookAfter('Pageimage::size', $this, 'uploadImageS3'); // Fired when a file is resized via API. Note: this hook is also called when the page is loaded in Admin. Need to know why. $this->addHookAfter('Pageimage::url', $this, 'redirectImageURL'); //Replace image URL for the S3 path } public function redirectImageURL($event){ if($event->page->template == 'admin') return; else $event->return = "https://s3-sa-east-1.amazonaws.com/[bucket name]/images/" . $event->object->page . "/" . $event->object->name; } // UPLOAD public function uploadImageS3($event){ if(count($event->return)) { //if it is a image resize event get image variation data $file = $event->return; } else { $file = $event->arguments(0); } $filename = $file->name; $filename_original = $file->url; $pathToFile = $file->page . "/" . $filename; $system_path = $file->filename(); try{ $file_on_s3 = $this->s3_filesystem->has($pathToFile); //check if file exists on AWS if(!$file_on_s3) { $contents = file_get_contents($system_path); $this->s3_filesystem->put($pathToFile, $contents, array( 'visibility' => 'public', )); //upload file with the same folder structure as in assets/files: page_id/file_name //Also add a copy to Dropbox (OPTIONAL) $this->dropbox_filesystem->put($pathToFile, $contents); } } catch (Exception $e){ throw new WireException("Error: Image not Added to S3: {$e->getMessage()}"); } } public function checkImageS3($event){ $page = $event->arguments(0); if(count($page->images)) { foreach($page->images as $file) { if($file->width == 0) { return $this->downloadImageS3($page, $file); }; } } } public function downloadImageS3($page, $file){ $pathToFile = $page->id . "/" . $file->name; $file_on_s3 = $this->s3_filesystem->has($pathToFile); //check if file exists on AWS if($file_on_s3) { $image = "https://s3-sa-east-1.amazonaws.com/[bucket name]/images/" . $pathToFile; // $page->images->remove($image); $page->images->add($image); $page->save(); } else { throw new WireException("Error: Image not found on S3: {$file->name}"); } } }

Hi Folks, I am currently building an angular app for creating and displaying polls/surveys: http://embed.plnkr.co/seQfprxQRvjg84bvi2Dy/ The data that needs to be stored in PW comes from JSON and looks like { "title": "My Survey", "questions": [ { "id": 1, "title": "First Question", "answers": [ { "id": 1, "title": "Answer 1" }, { "id": 2, "title": "Answer 2" } ] } ], "file": { "$ngfName": "long-beach-koh-payam.jpg", "$ngfOrigSize": 351901, "$ngfBlobUrl": "blob:http://localhost:8000/8b87de97-549a-4900-8579-71a0d87a9258" } } Now I need to decide how to best store that data with PW. I am also using Angular to build the forms where users can submit the poll from the JSON that is produced by the survey builder. So I would like to create 1 page per poll. The template holds title field, an image file field and one field that stores the data as JSON string. To me this seems an easier approach than having child pages for each question in the poll and again child pages for each answer to a question. Do you see any disadvantages that this approach might bring? And which field type should I use to store the JSON string, textarea?

Hello everyone, I'm creating a large scale application which will be used by thousands of people. Part of of the application requires storing temporary information about user's activities. For example, if user comments on certain topic, notification of same should be given to OP and people following the topic. I thought I'd create a subpage with a repeater field having few fields like user id, timestamp & notification status, read/unread (I just don't see why using page table is better in this case, if someone can enlighten me, that'd be great.). But I've read that every repeater item is same as new page, so I'm really confused whether I should just start keeping temporary pages or use a page with repeater field. But for every topic, there can be hundreds of notifications, which I think will be a problematic situation in future. So I am looking for any kind of temporary storage that will not included creating pages. Writing logs to file could be done but i'm not sure how good solution is that and how reliable it'll be. I'd really like a good conversation with someone who understands PW, so I'm here. Thank you all.

Hi all, I'm not sure how to best store multiple timestamps in a field for a page. As there can be 100 or more timestamps attached to the page, I need to find the most efficient way to save those timestamps in a field and later search through that field. My use case: I have pages for advertisements with start date, end date and start time, end time and frequency of publishing (every 15, 30, 60 minutes). From that data I calculate timestamps at which the advertisement will be available for viewing. One advertisement can have 100 or more timestamps attached to it. Example: title of page: Advertisement1 field "ad_publish" contains timestamps: 1413136860,1413136920,1413136980,1413137040,1413137100...(can be a lot) Now what would be the best way of storing these multiple timestamps. IOW: what field type to use? I need to take into consideration that the field needs to be searchable with a *= selector like $publishedAds = $pages->find("template=advertisement, ad_publish*={$timestamp}") 2 options that I can think of for my "ad_publish" field: 1. textarea field, store timestamps as comma delimited values. 2. page field: store each timestamp as a page What would be more efficient and faster when it comes to searching through hundreds of timestamps? I'm sure some of you had similar use cases in their projects and I would be happy if you could share your experience or give me pointers on how to best approach this. Thank you.

I want my site to have front end users able to register. Question: How should I handle registration information that is unvalidated? My idea was to make a sub group of pages called unvalidated users, and once they validate move them to validated users. Or I could name them unvalidated_username, and then rename it Or I could have a field checkbox for validatation. I guess in each of these scenarios, there would need to be some cleanup work... if a user stays unvalidated for x many weeks, delete the page? How would you do this. I expect to have thousands of front end users so I don't want to do anything manually. And yes, this is my first time building a custom dynamic website of any sort, so I have no prior experience with handling unvalidated users. Many thanks, aloha.