Shivamyadav wrote a reply+100 XP
3d ago
Shivamyadav wrote a reply+100 XP
1w ago
Shivamyadav liked a comment+100 XP
1w ago
@shivamyadav I see you didn’t take my advice (that I’m sure I mentioned a couple of times when you’ve posted about this Google Drive-related project) of moving your Google Client creation logic to a service provider, so that classes then using that client (controllers, etc) just need to type-hint it and it will be automatically built and resolved:
public function register(): void
{
$this->app->singleton(Client::class, function (Application $app) {
return new Client([
'client_id' => $app['config']['services.google.client_id'],
'client_secret' => $app['config']['services.google.client_secret'],
'scopes' => 'https://www.googleapis.com/auth/drive',
'redirect_uri' => $app['config']['services.google.redirect_uri'],
'prompt' => 'consent',
'access_type' => 'offline',
]);
});
}
For refreshing access tokens, I have a scheduled task that fires daily to refresh any soon-to-expire access tokens. But again, I’m sure I’ve mentioned this before:
Schedule::command('access-token:refresh')->daily();
If you find you’re unable to refresh an access token (because it’s been revoked by the owner), then you can delete it, and send a notification to the owner telling them they need to re-connect their Google account.
Shivamyadav wrote a reply+100 XP
1w ago
Shivamyadav liked a comment+100 XP
1w ago
Your logic is actually fine. The problem is almost certainly your database migration.
Google access tokens are frequently longer than 255 characters. If you used $table->string('access_token') in your migration, Laravel defaults to a VARCHAR(255) and is silently truncating your token when saving it to the database.
Here is exactly why your test behaved that way:
When you omit the created key, the Google client assumes the token is expired and forces a refresh. It then uses the full, fresh token directly from memory, which is why it works. However, when you include created, the client checks the timestamp, sees it hasn't been an hour yet, and sends the truncated (corrupted) token from the database to Google. Google rejects it, triggering the 401 Unauthorized.
Change your columns to text in your database migration:
$table->text('access_token');
$table->text('refresh_token');
Clear out your oauth_credentials table and re-authenticate to store a fresh, full-length token.
Avoid putting heavy external API calls inside a __construct(). If Google's API goes down, times out, or rate-limits you, your entire class fails to instantiate and it can crash Laravel's service container resolution. Move the token initialization to a dedicated connect() or boot() method that you call when you actually need to interact with the Drive API.
Shivamyadav wrote a reply+100 XP
3w ago
Shivamyadav wrote a reply+100 XP
3w ago
Shivamyadav started a new conversation+100 XP
3w ago
I’m building a lightweight Laravel-based attendance system where users can check in/out only when they are physically present within a strict 10–20 meter radius of the workplace using live GPS validation.
- ⚙️ Core Features Implemented 2.📍 Real-time geofencing (strict 10–20m radius enforcement)
- 🎯 GPS accuracy filtering (reject low-accuracy readings)
- ⏱️ Live check-in only (no delayed or backdated submissions)
- 🚫 Mock location detection (Android-based checks)
- 📱 Device tracking (basic fingerprinting for consistency)
- 🤳 Selfie Verification Approach (No AI)
To keep the system lightweight and privacy-friendly, I’m intentionally avoiding AI/ML solutions.
Current approach:
- Capture a selfie during check-in
- Compare it with a previously stored reference image
- Use pixel-level comparison techniques, such as:
- Image difference
- Hashing (e.g., perceptual hash)
- Similarity scoring
❓ Feedback I’m Looking For
I’d really appreciate insights from experienced developers on the following:
- Reliability How reliable is pixel-based image comparison in real-world usage?
- Real-world Variations How should I handle: 1. lighting differences 2. face angle changes 3. different camera quality
- Alternatives (Without AI) Are there any better non-AI approaches for basic face verification?
- Feasibility Is this approach fundamentally flawed for identity verification?
- Improvements Would capturing multiple selfies (2–3 per check-in) improve accuracy or just add noise? 🎯 Goal
The aim is to build a system that is:
- ✅ Simple to integrate
- ⚡ Lightweight (no heavy dependencies)
- 🔒 Privacy-friendly (no AI processing)
- 🛡️ Reasonably resistant to misuse
🙏 Closing
Any suggestions, critiques, or alternative approaches would be greatly appreciated. Looking forward to learning from your experience!
Shivamyadav wrote a reply+100 XP
1mo ago
Try this as the route ziggy package does not provide the reactive data on page navigation. You need to use the vue's reactivity with the route of the ziggy like this :
<script setup>
import { useSlots, h, computed } from 'vue'
import { usePage } from '@inertiajs/vue3'
const slots = useSlots()
const page = usePage()
// reactive current URL
const currentUrl = computed(() => page.url)
const renderFunc = () => {
const children = slots.collapseChild?.() || []
return children.map(child => {
return h(
"li",
{
class: {
active: child.props?.href === currentUrl.value
}
},
child
)
})
}
</script>
<template>
<ul>
<component :is="renderFunc" />
</ul>
</template>
Shivamyadav started a new conversation+100 XP
1mo ago
I have a GoogleDriveService to mange the work for the drive related.
Earlier it was setup by me to work like get the new refresh access token for each request and it was working fine.
Yesterday my senior suggested me to use this way: 1. To add the logic only get the new access token if it is expired. 2. I have added this logic but now I am getting 401 (Unauthorized) error.
My constructor code
public function __construct()
{
$client = new \Google\Client();
$client->setClientId(config('google.drive.client_id'));
$client->setClientSecret(config('google.drive.client_secret'));
$client->setRedirectUri(config('google.drive.redirect_uri'));
$client->addScope(\Google\Service\Drive::DRIVE);
$client->setAccessType('offline');
$client->setPrompt('consent');
// INFO: Get the oauth credentials as an array
$accessToken = OauthCredentials::first()->toArray();
// dd($accessToken);
Log::info('Db token', $accessToken);
if (!isset($accessToken['access_token'])) {
throw new \Exception('Failed to fetch access token. Check your refresh token!');
}
// INFO: Set the access token with all the details
$client->setAccessToken([
'access_token' => $accessToken['access_token'],
'refresh_token' => $accessToken['refresh_token'],
'token_type' => $accessToken['token_type'],
'expires_in' => (int) $accessToken['expires_in'],
'created' => (int) $accessToken['token_created_at'], // Unix timestamp
]);
// dump($client->isAccessTokenExpired());
/** Chech if the token is expired or not and refresh if needed */
if ($client->isAccessTokenExpired()) {
Log::info("Access token expired. Refreshing...");
// INFO: Use the db refresh token to fetch the new token
$newToken = $client->fetchAccessTokenWithRefreshToken($accessToken['refresh_token']);
// INFO: Set the refresh token if the new token doesn't have it
if (!isset($newToken['refresh_token'])) {
$newToken['refresh_token'] = $accessToken['refresh_token'];
}
/** Store the credentials and token related meta data to the db */
OauthCredentials::updateOrCreate(
['id' => 1],
[
'refresh_token' => $newToken['refresh_token'],
'access_token' => $newToken['access_token'],
'token_type' => $newToken['token_type'] ?? 'Bearer',
'expires_in' => $newToken['expires_in'],
'token_created_at' => $newToken['created'],
]
);
/** Store the credentials and token related meta data to the db */
Log::info('New token', $client->getAccessToken());
Log::info("New access token generated : ", $newToken);
// Set new token
$client->setAccessToken($newToken);
Log::info("New access token generated");
}
$this->drive = new \Google\Service\Drive($client);
$this->parentFolderId = config('google.drive.parent_folder_id');
}
As soon i removed this below line form the top (first) set access token
'created' => (int) $accessToken['token_created_at'], // Unix timestamp
It works but now each time it fetches the new access token for it.
Main error i am getting
"error": {
"code": 401,
"message": "Request had invalid authentication credentials. Expected OAuth 2 access token, login cookie or other valid authentication credential. See https://developers.google.com/identity/sign-in/web/devconsole-project.",
"errors": [
{
"message": "Invalid Credentials",
"domain": "global",
"reason": "authError",
"location": "Authorization",
"locationType": "header"
}
],
"status": "UNAUTHENTICATED"
}
}
Shivamyadav started a new conversation+100 XP
1mo ago
For the first time when I click to my sidebar notifications button it does two things:
- Navigate to the notifications page.
- Clears the notification badges for the new emails.
And then I tried some filter to the notifications and when I am done with the filtering and click on the notifications button again it works again for step 1 and step 2 but the issue arrives here in the form of the not letting the search and other inputs gets cleared but the data is at the initial state without applied any filter and url is also clean.
I have tried with the preserveState to be false for the below function markNewMailsRead() but did not worked.
Is there something I am missing or any solution for this issue.
My sidebar link component
<SidebarLink @click="markNewMailsRead" :href="route('notifications.index')"
:active="isActiveLink('notifications.index')">
<div class="flex items-center justify-between w-full">
<!-- Left -->
<div class="flex items-center gap-4">
<Bell class="size-5" />
<span>Notifications</span>
</div>
<!-- Badge -->
<div v-if="totalNewMails" class="relative">
<!-- Pulse ring -->
<span class="absolute inset-0 rounded-full bg-primary animate-ping"></span>
<!-- Count -->
<span class="relative flex items-center justify-center
w-6 h-6 p-1
rounded-full bg-primary
text-xs font-bold text-black">
{{ totalNewMails > 99 ? '99+' : totalNewMails }}
</span>
</div>
</div>
</SidebarLink>
// INFO: New Mail mark as read ( to just make the live new count 0)
const markNewMailsRead = () => {
router.post(route('notifications.new-mails.read'), {}, {
preserveScroll: true,
preserveState: true,
});
};
Shivamyadav wrote a reply+100 XP
1mo ago
Shivamyadav liked a comment+100 XP
1mo ago
@shivamyadav As @ian_h says, you should not be using the env helper in your code. The Laravel docs also tell you not to do this.
Instead, you should be binding that class in the service container where the configuration values are passed to your constructor:
public function register(): void
{
$this->app->singleton(GoogleDriveService::class, function () {
return new GoogleDriveService(
clientId: $this->app['config']['services.google.drive.client_id'],
clientSecret: $this->app['config']['services.google.drive.client_secret'],
redirectUri: $this->app['config']['services.google.drive.redirect'],
);
});
}
You can then just type-hint the service class, and it will be resolved (and configured) by the service container:
class SomeController extends Controller
{
protected GoogleDriveService $googleDrive;
public function __construct(GoogleDriveService $googleDrive)
{
$this->googleDrive = $googleDrive;
}
}
As for the error, it’s telling you what the problem is: service accounts don’t have a storage quote, so you can’t use service accounts for storage-related operations like you’re trying to do. You need to use an OAuth approach so you’re authenticating as a user, who will have a storage quota. You can use Socialite to create (and refresh) Google access tokens.
Shivamyadav wrote a reply+100 XP
1mo ago
Shivamyadav liked a comment+100 XP
1mo ago
Shivamyadav wrote a reply+100 XP
1mo ago
<?php
namespace App\Services;
use Google\Client;
use Google\Service\Drive;
use Google\Service\Drive\DriveFile;
class GoogleDriveService
{
protected Drive $drive;
protected string $parentFolderId;
public function __construct()
{
$client = new Client();
$client->setClientId(env('GOOGLE_DRIVE_CLIENT_ID'));
$client->setClientSecret(env('GOOGLE_DRIVE_CLIENT_SECRET'));
$client->setRedirectUri(env('GOOGLE_DRIVE_REDIRECT_URI'));
$client->addScope(Drive::DRIVE);
$client->refreshToken(env('GOOGLE_DRIVE_REFRESH_TOKEN'));
$this->drive = new Drive($client);
// Use parent folder from config
$this->parentFolderId = config('drive.parent_folder_id');
}
/**
* Upload a file into a subfolder based on the config mapping
*/
public function uploadByPosition(string $filePath, string $fileName, string $position): array
{
// Get folder name from config mapping, default to position if not mapped
$folderMap = config('drive.position_folders', []);
$folderName = $folderMap[$position] ?? $position;
$subFolderId = $this->getOrCreateSubFolder($folderName);
$fileMetadata = new DriveFile([
'name' => $fileName,
'parents' => [$subFolderId]
]);
$content = file_get_contents($filePath);
$file = $this->drive->files->create($fileMetadata, [
'data' => $content,
'mimeType' => mime_content_type($filePath),
'uploadType' => 'multipart',
'supportsAllDrives' => true,
]);
return [
'subfolder' => $folderName,
'file_id' => $file->id,
'preview_url' => $this->getFileUrl($file->id)
];
}
/**
* Get or create a subfolder under parent
*/
protected function getOrCreateSubFolder(string $folderName): string
{
$response = $this->drive->files->listFiles([
'q' => "name='{$folderName}' and mimeType='application/vnd.google-apps.folder' and '{$this->parentFolderId}' in parents and trashed=false",
'spaces' => 'drive',
'fields' => 'files(id, name)',
'supportsAllDrives' => true,
]);
if (count($response->files) > 0) {
return $response->files[0]->id;
}
// Folder not found → create it
$folderMetadata = new DriveFile([
'name' => $folderName,
'mimeType' => 'application/vnd.google-apps.folder',
'parents' => [$this->parentFolderId],
]);
$folder = $this->drive->files->create($folderMetadata, [
'fields' => 'id',
'supportsAllDrives' => true,
]);
return $folder->id;
}
/**
* Generate a preview URL for Google Drive
*/
public function getFileUrl(string $fileId): string
{
return "https://drive.google.com/file/d/{$fileId}/view";
}
}
Shivamyadav started a new conversation+100 XP
1mo ago
Currently I am using the service account and setup the testing route with the credentials and getting this error:
Route::get('/drive-test', function () {
$service = new \App\Services\GoogleDriveService();
$fileId = $service->upload(
storage_path('1770402265.shivam-resume (1).pdf'),
'test.pdf'
);
return $service->getFileUrl($fileId);
});
Google \ Service \ Exception (403)
{ "error": { "code": 403, "message": "Service Accounts do not have storage quota. Leverage shared drives (https://developers.google.com/workspace/drive/api/guides/about-shareddrives), or use OAuth delegation (http://support.google.com/a/answer/7281227) instead.", "errors": [ { "message": "Service Accounts do not have storage quota. Leverage shared drives (https://developers.google.com/workspace/drive/api/guides/about-shareddrives), or use OAuth delegation (http://support.google.com/a/answer/7281227) instead.", "domain": "usageLimits", "reason": "storageQuotaExceeded" } ] } }
Shivamyadav wrote a reply+100 XP
2mos ago
After reading your reply. I thought yeah this is the way I can achieve it.
Also I can check the every pagination data listing with that source of truth and mark the checkbox for every visiting page to checked the box to provide the feedback to the user and got the some help from the AI to unselected paginations data listing could be added to the another source of truth like. excluded_records which will hold the ID's which does not required to be deleted.
Shivamyadav liked a comment+100 XP
2mos ago
Shivamyadav started a new conversation+100 XP
2mos ago
I have created a notification listing with the pagination but I want to delete all the notifications or only the selected one to delete it.
- I was using the checkbox for each record and the top header select all the records form the listing and it selects all the records but only from which ever page I select the header checkbox to select all the listing.
- When I go to the 2nd page it's record listings was not selected how to get the all the listing selected within the pagination too?
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav liked a comment+100 XP
2mos ago
Shivamyadav liked a comment+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav liked a comment+100 XP
2mos ago
Laravel 4.1 is ancient and I wouldn't try to migrate it step by step.
- The intermediate versions won't work with modern PHP versions. Laravel 4 doesn't support PHP 7+.
- The intermediate versions will require obsolete or completely removed extensions, as you found out with Mcrypt.
- You'll have to rewrite parts of the app completely along the way. Laravel 4 doesn't even have middleware. It used filters instead, which were removed in L5.
If I were you, I'd create a fresh Laravel 12 project with PHP 8.x, and start moving the business logic from the old app to the new one, piece by piece.
You'll still have to read every upgrade guide along the way, especially the major version ones.
Shivamyadav liked a comment+100 XP
2mos ago
Why only migrate to 4.2 ? This could be done by following the upgrade guide.
But I suggest you to migrate to Laravel 12.
For such a migration, I suggest you to create a new Laravel application and copy / paste the code from your old application to your new one. Then you will have to change some code to be compatible with the recent PHP versions and with the new structure for the application folder.
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav started a new conversation+100 XP
2mos ago
Currently I have assigned a project to migrate it. As this is the first time I am working on the migrating the older project.
Current Working state:
- I have installed a new fresh
laravel 4.2project and according to the upgrade guide I have makes the necessary changes and put the copied existing older project withlaravel version 4.1. - New fresh project is using the php 7.3 and mysql 9.
- I have copied the the older project composer.json to the new project but kept the laravel version to the 4.2 and other things.
- Now when I ran any artisan command I get this error
Mcrypt PHP extension required.
How to fix it. I have tried to resolve it from the AI too but It seems not working.
Any help would be appreciated.
Shivamyadav was awarded Best Answer+1000 XP
2mos ago
Here is the key thing...
// Searched on Google and got this load missing() used case.
Laravel's loadMissing() method is used to eager load a relationship on a model or a collection of models only if that relationship has not already been loaded.
-
withPivot() only applies when the relationship is loaded for the first time.
-
If groups.articles was already loaded earlier without the formula column, loadMissing() will not re-query or add new pivot fields.
-
Eloquent treats the relationship as already loaded, so this behavior fails silently by design.
I think you can define the 2 relationship one for without formula and other with formula
Like this ..
//Without formula relationship
public function articles(): BelongsToMany
{
return $this->belongsToMany(Article::class, 'group_article')
->using(GroupArticle::class)
->withPivot('rounding_rule', 'precision');
}
// With formula needed
public function articlesWithFormula(): BelongsToMany
{
return $this->belongsToMany(Article::class, 'group_article')
->using(GroupArticle::class)
->withPivot('formula', 'rounding_rule', 'precision');
}
Shivamyadav wrote a reply+100 XP
2mos ago
Here is the key thing...
// Searched on Google and got this load missing() used case.
Laravel's loadMissing() method is used to eager load a relationship on a model or a collection of models only if that relationship has not already been loaded.
-
withPivot() only applies when the relationship is loaded for the first time.
-
If groups.articles was already loaded earlier without the formula column, loadMissing() will not re-query or add new pivot fields.
-
Eloquent treats the relationship as already loaded, so this behavior fails silently by design.
I think you can define the 2 relationship one for without formula and other with formula
Like this ..
//Without formula relationship
public function articles(): BelongsToMany
{
return $this->belongsToMany(Article::class, 'group_article')
->using(GroupArticle::class)
->withPivot('rounding_rule', 'precision');
}
// With formula needed
public function articlesWithFormula(): BelongsToMany
{
return $this->belongsToMany(Article::class, 'group_article')
->using(GroupArticle::class)
->withPivot('formula', 'rounding_rule', 'precision');
}
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav started a new conversation+100 XP
2mos ago
In my projects / Index component i have this code and when I search anything and get's the expected result and try to navigate back to the same projects/ index url on i had searched earlier the input search component get's clear and even also from the url params.
my code
const page = usePage();
const projects = computed(() => page.props.projects);
const breadcrumbLinks = [
{ name: 'Dashboard', href: route('admin.dashboard') },
{ name: 'Projects', href: route('projects.index') },
];
// Create form with search field
const form = useForm({
search: page.props.search ?? ''
});
// Debounced request for search
const debouncedSearch = debounce(() => {
form.get(route('projects.index'), {
replace: true,
preserveScroll: true,
preserveState: true
});
}, 500);
// Watch form.search directly
watch(() => form.search, () => {
debouncedSearch();
});
// Cancel debounce on unmount
onUnmounted(() => {
debouncedSearch.cancel();
});
The main issue comes here in the Notifications / Index component I was expecting the same behaviour but did not get the input cleared.
- The url search query gets removed but the input search has the searched value.
- Event using the smae approach for the both components.
my code
const page = usePage();
const notifications = computed(() => page.props.notifications);
// Form (search + filter + date)
const form = useForm({
search: page.props.search ?? "",
filter: page.props.filter ?? "all",
from_date: page.props.from_date ?? "",
to_date: page.props.to_date ?? "",
});
// Debounced request
const debouncedSearch = debounce(() => {
form.get(route("notifications.index"), {
replace: true,
preserveScroll: true,
preserveState: true,
});
}, 500);
// Watch search, filter, and dates
watch(
() => [form.search, form.filter, form.from_date, form.to_date],
() => {
debouncedSearch();
}
);
// Cleanup
onUnmounted(() => {
debouncedSearch.cancel();
});
Is there anything I am missing ?
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav was awarded Best Answer+1000 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
If you really want to use the facade:
@php
use Illuminate\Support\Facades\Session;
@endphp
@if ($message = Session::get('flash'))
<div class="alert alert-success">
{{ $message }}
</div>
@endif
Or just use the session helper function
@if (session('success'))
<div class="alert alert-success">
{{ session('success') }}
</div>
@endif
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav liked a comment+100 XP
2mos ago
@shivamyadav That makes absolutely no difference. I use SQS myself. You don’t need to store failed jobs in a custom table just to then present them in a UI.
You can query the failed_jobs table just like any other table:
$failedJobs = DB::table('failed_jobs')->paginate();
@foreach($failedJobs as $failedJob)
<tr>
<td>{{ $failedJob->uuid }}</td>
<td>
<!-- Button to re-try failed job -->
</td>
</tr>
@endforeach
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav liked a comment+100 XP
2mos ago
Yes, I think he just was a list of all failed jobs in the ui listing
@shivamyadav Your senior is literally re-inventing the wheel.
You could have just used Horizon for Redis-based queues. If you’re not using Redis, you still don’t need to push jobs to an entirely new table. You could have just queried Laravel’s native failed_jobs table. I’ve done it myself; including with a button to re-try failed jobs; no custom table needed.
Given the questions you’ve asked in such a short span of time, I don’t feel this “senior” is really someone you want to be learning habits from, as they seem to be senior in name only.
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav wrote a reply+100 XP
2mos ago
Shivamyadav started a new conversation+100 XP
2mos ago
My app service provider queue event store the failed jobs to the custom table
Queue::failing(function (JobFailed $event) use ($queueJobStatusRepository) {
/** Extracting the current job class name */
$job = $event->job;
$payload = $job->payload();
$action = $payload['displayName'];
/** Extracting the current job class name */
$queueJobStatusRepository->updateOrCreate(
['job_id' => $event->job->uuid()], // INFO: Uniquely identify to create or update
[
'action' => $action, // INFO: Get the current Event / Job Class name
'status' => 4, // INFO: Failed
'exception' => str()->limit($event->exception->getMessage(), 2000)
]
);
});
I want to get the failed job uuid and based on that I could push that failed jobs to the queue again to process it.
Shivamyadav started a new conversation+100 XP
2mos ago
My senior uses the repository patterns to work with the controller business logics. In the function he has the create and update method to create/ update the resource and other side effects to perform actions whenever resource gets created or updated and both were approximately 900 of lines.
That was really too hard to understand it and overwhelming for me to just looking at the no of the lines in the function.
How long no of lines any function could have?
Shivamyadav wrote a reply+100 XP
2mos ago
Sorry 😔.
I will provide the information once I am on my machine.
For now my senior has written the whole code and and was optimising it and find that the creation of the user was the problem with many things like creating user stripe account and divisions, experts etc.
So I just put that in a queue and imported the user's again and now it's taking 8 minutes.
Without the job (as if I commented the job) it was taking one 1.20 min and added the job it's taking that 8 minutes like resolving the jobs dependency and serealizing the constructor payload and I am using the AWS SQS.