You shouldn't ever need a million rows. Query what's needed and paginate server side.

Not really, eloquent is just sql in the end, and to order or filter by another table, you need to do a join. Writing joins in eloquent/query builder isn't always pretty, at least not if there are multiple tables involved. One option I like to use is to push the query down into the database, and use a database view, and the use eloquent on that view just as I would a single table. I wrote a blog post on that a few years back, and you can find it here.

https://tray2.se/posts/use-a-view-instead-of-a-complex-eloquent-query-in-your-laravel-application

@jeffreyway Can you take a look at this issue?

It depends a little, but as a rule of thumb, you should always leave the code a little cleaner than it was before.

Take this small JavaScript example.

Original code

 function setItemHtml(item, html) {
    let el = document.querySelector(item);
    el.innerHTML = html;
 }

Iterration 1

 function getElement(selector) {
   return document.querySelector(selector);
 }

 function setItemHtml(item, html) {
    let el = getElement(item);
    el.innerHTML = html;
 }

Itteration 2

 function getElement(selector) {
   return document.querySelector(selector);
 }

 function setItemHtml(item, html) {
    getElement(item).innerHTML = html;
 }

This is the way I would do it, for two reasons, one: reuse, two: cleaner code. You can argue that using a wrapper for document.querySelector() is unnecessary, I don't agree, getElement is much shorter and at least to me makes more sense. The inlining of the getElement and the assign of innerHTML is a good way to make the function smaller.

Exactly, tenant first, and then if some company says they need their own db, host the application separately for them.

It depends, if you have a single controller for it, then it makes sense to have a single test file, but if you use single action controllers, then it makes more sense to have separate test files.

That is of course not the only thing to take into consideration, the size of the file, and groupings of tests, for example separating the validation tests from the route tests would be an option as well.

Only when it's a legal requirement, and if that is the case, I would host the applications separately as well. One lemp stack for each client.

In the first example you are loading hundreds of records from the database, and you are checking if you are allowed to delete it for each record, while in the other you are just checking if a variable is true or false.

I would save the latest payload in the database to display when and if the API fails, then display a last updated message.

I do however have some suggestions.

1. Don't be cheap and use a flaky service like that, find a proper one.
2. Pay for a service to ensure a good uptime.

I will put it like this.

A database table can without problem contain hundred of millions of records, and if indexing isn't enough, you can always partition the table, and that is possible to to per tenant, just give every table a tenant_id, and you should be good to go.

Scope all queries per tenant, then you have the isolation as well.

It's all about database structure, indexes, and proper queries.

Also make sure you normalize your database

https://en.wikipedia.org/wiki/Database_normalization

My recommendation is to use integers as primary and foreign keys, they are usually faster, and to avoid using json columns, since they aren't really indexable.

I would suggest against it. Use a single db, it will make your life a lot easier.

The short answer is YES.

The slightly longer answer is also yes, the integrity in the database is very important, without it you can't control your application's data in a good way.

Starting out with a column as nullable isn't in my opinion a good idea, it will cause issues when later making it not null. If you manage to do that and keep the database empty until the launch of the app, then it's better, but not good.

I would suggest spending more time on the initial database model.

I wrote these posts a few years ago, but they are still valid.

https://tray2.se/posts/database-design

https://tray2.se/posts/database-design-part-2

https://tray2.se/posts/properly-formed-foreign-keys-are-your-best-friends

I usually use the dialog tag for that, it's very easy to use, and has a lot of native functionality.

https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/Elements/dialog

You can't really prevent it, people can change the metadata. The easiest I think would be to remove any meta, get a checksum for the image, and compare it to what is stored in the database.

@reshadman My feeling is, if Taylor and the Laravel team wanted validation at the model layer, it would be already.

I will put it plainly, unless the data is hashed, the data will be accessible by the person having the root access.

There is no way to encrypt the data and store know the key, the only way for that to work is to allow the tenant to use some kind of secret key, that they need to enter to decrypt the data, and even that isn't fool proof, it probably need to be either passed in every request, or stored temporarily on the session.

I've used mohammedmanssour/form-request-tester for these kinds of tests.

The syntax is very clean

use App\Http\Requests\BookFormRequest;
use Carbon\Carbon;
use MohammedManssour\FormRequestTester\TestsFormRequests;

uses(TestsFormRequests::class);

it('fails if the title is missing', function () {
    $this->formRequest(BookFormRequest::class)
        ->post(['title' => ''])
        ->assertValidationFailed()
        ->assertValidationMessages(['The title field is required.']);
});

it('fails if the title is not a string', function () {
    $this->formRequest(BookFormRequest::class)
        ->post(['title' => ['Some title']])
        ->assertValidationFailed()
        ->assertValidationMessages(['The title must be a string.']);
});

I think it might have something to do with the change of hosting for the videos. The old host worked fine with the correct file names, and the new one gave then a generic name like download.mp4 or something like that. So that combined with other stuff, it makes sense to remove it.

@lukask If I'm not mistaken, they are only for subscribers.

You do know that this thread is 11 years old right?

While AI is a good tool to use in many tasks, writing code that you don't understand or know exactly what is does, is not only a risk for bugs, it might also be a security risk.

You can probably get away with not understanding all of the code it writes for a while, but sooner or later it will bite you in the ass.

The smarter the AI tool becomes, the better of a programmer you need to be to spot the mistakes.

A few years back it was quite easy to spot an AI generated image, it could have six fingers or something similarly visible, but now it is really hard to spot one, the same goes for code, it used to be easy to spot the mistakes made by AI, now it isn't as easy.

Not sure exactly what it is you want to track here, but a table trigger might be an alternative if you want to track changes made byt the user.

https://tray2.se/posts/using-table-triggers-to-log-changes-in-your-database-tables

In that case I would use a sequence from the database.

https://mariadb.com/docs/server/reference/sql-structure/sequences/create-sequence

I would even go so far as to why UUID at all if you don't need to make it impossible to guess.

check https://packgist.org

I would say that it's generally reliable, the thing is most likely that things sometimes takes a little longer to run/render that you would suspect. It could be a db query, or and API hit that causes the delay, and if something is missing when you try to access it, it might look like random errors.

I've been using Cypress and Playwright to run browser tests, and I had to add extra delays in multiple places to get the tests to green every time.

@Tray2 thank you sir you saved me after 4 hours i dont know what was happening getting crazy !!!

@terdoo-mzer Simple you can consider the routes a list of addresses, and it select the first matching one.

Yes, you can develop a blog with laravel, you can basically create any kind of web application with it.

Start here https://laracasts.com/series/laravel-from-scratch-2026

Easiest is to use github, other than that you can use a mounted share, or a regular usb stick.

Looks like that senior used to be a Cobol programmer.

I guess there are no tests written for it either, but if there are, I would ctrl-a delete, and start from scratch.

If not, the I would start by extraction each segment of the if statements to it's own method.

For example

    if ($data['user']['send_email']) {
                $skipResetPassword = array_key_exists('skip_reset_password', $data['user']);
                $withDelay = (array_key_exists('invite_email_with_delay', $data['user']) ? $data['user']['invite_email_with_delay'] : true);
                
                $this->sendInviteEmail($modelResource, $setInviteFlag, $skipResetPassword, $withDelay);
            }

would become something like this

function sendInviteEmailIfNeeded($modelResource, $setInviteFlag, $data)
{
  if (! $data['user']['send_email']) return;

  $skipResetPassword = array_key_exists('skip_reset_password', $data['user']);
  $withDelay = (array_key_exists('invite_email_with_delay', $data['user']) ? $data['user']['invite_email_with_delay'] : true);
  $this->sendInviteEmail($modelResource, $setInviteFlag, $skipResetPassword, $withDelay);
}