Need to confirm that re-enabling `\Illuminate\Foundation\Http\Middleware\ConvertEmptyStringsToNull::class,` won’t mangle anything. I know we ran into some issues when testing a long time ago, but not sure those issues apply anymore, and I can’t remember what they were.
* Added AWS url to example env
* Upgrader - added check for new storage path and attempt to move
* Ignore symlink
* Updated paths for models
* Moved copy methods
* Added AWS_URL support
For some reasin, Flysystem was generating the wrong AWS url (with a region included)
* Switch to Flysystem for image uploads
* Nicer display of image preview
* Updated image preview on edit blades to use Flysystem
* Twiddled some more paths
* Working filesystems config
* Updated Asset Models and Departments to use Flysystem
* Janky workaround for differing S3/local urls/paths
* Try to smartly use S3 as public disk if S3 is configured
* Use public disk Storage options for public files
* Additional transformer edits for Flysystem
* Removed debugging
* Added missing use Storage directive
* Updated seeders to use Flysystem
* Default logo
* Set a default width
We can potentially override this in settings later
* Use Flysystem for logo upload
* Update downloadFile to use Flysystem
* Updated AssetFilesController to use Flysystem
* Updated acceptance signatures to use Flysystem
* Updated signature view to use Flysystem
This isn’t working 100% yet
* Use Flysystem facade for displaying asset image
* Set assets path
Should clean all these up when we’re done here
* Added Rackspace support for Flysystem
* Added Flysystem migrator console command
* Added use Storage directive for categories
* Added user avatars to Flysystem
* Added profile avatar to Flysystem
* Added the option to delete local files with the migrator
* Added a check to prevent people from trying to move from local to local
* Fixed the selectlists for Flysystem
* Fixed the getImageUrl method to reflect Flysystem
* Fixed AWS copy process
* Fixed models path
* More selectlist updates for Flysystem
* Updated example .envs with updated env variable names
* *sigh*
* Updated non-asset getImageUrl() methods to use Flysystem
* Removed S3 hardcoding
* Use Flysystem in email headers
* Fixed typo
* Removed camera support from asset file upload
We’ll find a way to add this in later (and add that support to all of the other image uploads as well)
* Fixed path for categories
* WIP - Switched to standard handleImages for asset upload.
This is currently broken as I refact the handleImages method. Because the assets store/create methods use their own Form Request, the handleImages method doesn’t exist in that Form Request so it wil error now.
* Fixed css URL error
* Updated Debugbar to latest version (#6265)
v3.2 adds support for Laravel 5.7
* Fixed: Missing CSS file in basic.blade.php (#6264)
* Fixed missing CSS file in basic.blade.php
* Added
* Changed stylesheet import for authorize.blade.php
* Updated composer lock
* Added AWS_BUCKET_ROOT as env variable
* Use nicer image preview for logo upload
* Removed AssetRequest form request
* Removed asset form request, moved custom field validation into model
* Added additional help text for logo upload
* Increased the size of the image resize - should make this a setting tho
* Few more formatting tweaks to logo section of branding blade preview
* Use Flysystem for asset/license file uploads
* Use Flysystem for removing images from models that have been deleted
* Enable backups to use Flysystem
This only handles part of the problem. This just makes it so we can ship files to S3 if we want, but does not account for how we backup files that are hosted on S3
* Use Flysystem to download license files
* Updated audits to use Flysystem
* Extract a handlesimages trait to centralize logic for parsing/storing images on upload in create/edit methods.
* Use same image upload/layout in accessories as consum+components.
* Monster: Cleanup/Refactor http controllers.
This cleans up docblocks, pulls most non-crudy actions into their own
controllers, and does general cleanup/logic refactoring. There /should/
be no functional changes, but we all know how should works..
Extract checkin/checkout functions to a separate controller for accessories.
Move controllers to subdirectory.
Cleanup AssetModelsController
Extract component checkin/checkout
Assorted cleanups/doc/formatting in controllers.
Refactor LicenseController.
Refactor UsersController
Update viewassetscontroller.
* Codacy cleanups
* More codacy cleanups. Extract a LicenseCheckout Form request as well.
* A bit more refactor/cleaning of the license checkout method.
* Review Related Cleanups
* Fix most of the item_not_found translations. In many cases, the
string being generated did not even use the id parameter. Where it
does, pass it as id instead of as a different value.
* Remove some old $data arrays from when we manually sent emails from
the controllers. This has been superseeded by the notification system
(yay!)
* Bugfix: Only log the checkin of an accessory if the checkin completes sucessfully.
* adds permission checks for companies
* adds permission checks for depreciations
* adds permission check for all reports
* fixes permissions for departments
* fixes permission naming (edit -> update)
* fixes authorization checking wrong permission in API
The authorization was checking for the non-existent „edit“ method where it should have checked for the „update“ method.
* adds authorization checks for select2 lists
* adds missing authorization checks for api
* fixes user authorization check for creating users
* adds additional check viewing assets on showing a users assets
* Removes authorization checks for select2 lists
Reference: https://github.com/snipe/snipe-it/pull/5807#pullrequestreview-136018755
* Added de-norm counter migration for assets
* Renaming counter columns, since Eloquent has a magical *_count helper
* Added artisan command to sync counters (one-off)
* Update API to use de-normed fields
* Increment counters for checkin;/checkout
* Derp.
* Added request increment/decrementer
* Move increment for checkout to the Asset::checkout method
* Added “could take a while” message
* Fixes#4445: prevents assigned assets from being checked out in bulk checkout
* Updates data attribute to more versatile 'data-asset-status-type'
* Fixes broken unit test
* WIP - beginning of improved requested assets
- Use Ajax tables for faster loading
- Use new notifications for requesting an asset
TODO:
- Use ajax tables for requestable asset models
- Use new notifications for canceling an asset request
- Expire requests once the asset has been checked out to the requesting user
* Only show asset name in email if it has one
* Refactor requested method to only include non-canceled requests
* Refactored requestable assets to log request and cancelation
* Added softdeletes on checkout requests
* Differentiate between canceling and deleting requests
* Added asset request cancelation notification
* Added timestamps and corrected unique key on requests table
* Improved requests view
* Re-use blade for cancel/request email
* Refactored BS table formatter for requested assets
* Location name min reduced to 2
* Added PAT test as maintenance option
This needs to be refactored into database-driven options with a UI
* Better slack message
* Added getImageUrl method for assets
* Include qty in request notifications
TODO:
- Try to pull requested info from original request for cancelation, otherwise it will default to 1
* Removed old asset request/cancel emails
* Added user profile asset request routes
* Added profile controller requested assets method
* Added blade link to requested assets for profile view
* Sort user history desc
* Added requested assets blade
* Added canceled at to checkoutRequest method
* Include qty in request
* Fixed comment, removed allowed_columns
* Removed Queable methods, since we don’t use a queue
* Fixed return type in method doc
* Fixed version number
* Changed id to user_id for clarity
* Added “show fields in email” to custom fields
* Added “show images in email” to settings
* Added nicer HTML emails
* Break notifications out into their own, instead of trying to mash them all together
* Remove old notification for accessory checkout
* Janky fix for #5076 - “The asset you have attempted to accept was not checked out to you”
* Add method for image url for accessories
* Added accessory checkout email blade
* Make accessory email notification on checkout screen consistent with assets
* Added native consumables notifications
* Fixes for asset notification
* Updated notification blades with correct-er fields
* Updated notifications
* License checkin notification - does not work yet
Need to figure out whether the license seat is assigned to a person or an asset before we can pass the target
* Added alternate “cc” email for admins
* Only try to trigger notifications if the target is a user
* Fix tests
* Fixed consumable URL
* Removed unused notification
* Pass target type in params
* Show slack status
* Pass additional parameters
There is a logic bug in this :( Will send to slack twice, since the admin CC and the user are both using the same notification. Fuckity fuck fuck fuck.
* Pass a variable to the notification to supress the duplicate slack message
* Slack is broken :( Trying to fix
Will try a git bisect
* Put preview back into checkout
* Pulled old archaic mail
* Removed debugging
* Fixed wrong email title
* Fixed slack endpoint not firing
* Poobot, we hardly knew ye.
* Removed old, manual mail from API
* Typo :-/
* Code cleanup
* Use defined formatted date in JSON
* Use static properties for checkin/checkout notifiers for cleaner code
* Removed debugging
* Use date formatter
* Fixed target_type
* Fixed language in consumable email
* Allow auto incrementing asset_id from the API when the setting is enabled
* Cleaned up the if else statement
* Added prefix to the orWhereRaw which causes a database error if the configuration uses a prefix
* Auto incrementing through the API
* Fix some n+1 problems
* Use route in notification dropdown to make sure we link to correct page
* Work on better UI support for checkout to non-user. Fix links on index bootstrap table, work towards eliminating assignedUser
* Remove Asset::assigneduser() relationship. Instead add a checkedOutToUser() method and/or port to using assignedTo()
* Adjust string to fit new reality
* Fix#3780. Move the consumables getDataView method to the ApiController. Not entirely RESTful, but it's a weird method that probably doesn't need its own controller and the functionality would be strange to stack on the userscontroller...
* Fix file uploads to assets and restore the delete route.
* Add asset maintence edit action to index.
* Suppliers asset list should link to the related asset, not to the supplier with same ID.
* Asset models page should use polymorphic formatter on assigned to to better handle assorted item types.
* Comment out more assigneduser fallacy until we figure out the query builder approach to searching for location text.
* Refactored AssetsTransformer
Casted all ids to int, escaped all text values,
* Added warranty_expires attribute to Asset model
$asset->warranty_expires contains a Carbon object with the warranty
expiration date. Returns null when either purchase_date or
warranty_months are not set.
* Ignoring php-cs cache files
* Restored asset tests expectations
Work in progress - tests still fail
* API controller refactoring, fixed HTTP status codes in responses
* Restored $request->get - debugging
* Added further checks in ApiAssetsCest::updateAssetWithPatch
* /api_tests: Fixed bugs in update() method + code formatting
* /api_tests: Minor code formatting
* /api_tests: Mirrored test code for PATCH and PUT methods
* Removed repeated code
* Test cleanup
* Fixed issues with update() through PATCH and PUT methods
* Added API test suite to .travis.yml
* Refactored AssetsTransformer
Casted all ids to int, escaped all text values,
* Added warranty_expires attribute to Asset model
$asset->warranty_expires contains a Carbon object with the warranty
expiration date. Returns null when either purchase_date or
warranty_months are not set.
* Ignoring php-cs cache files
* Restored asset tests expectations
Work in progress - tests still fail
* API controller refactoring, fixed HTTP status codes in responses
* Restored $request->get - debugging
* Added further checks in ApiAssetsCest::updateAssetWithPatch
* Begin work on vueifying the importer
* Beginning work on migrating the importer to use a vue/components for future interactivity
Update JS
More importer work. Move to a vue based modal, begin handling of processing. Still need to port error messages.
More importer work. Move to a vue based modal, begin handling of processing. Still need to port error messages.
Update importer. Add error display. Fix modal, update vue-strap to vue2
More progress. Add select2 vue bits.
* Move to querying the db to find importer matches. It scales better on large datasets.
Fix select2 related issues. We were trying to initialize it twice, which led to the custom data being overwritten.
* Better error handling on uploads and deletion of files. Restore progressbar on upload.
* Add support for generic exception reporting if app.debug is enabled.
* Handle Http 500 errors better. Display errors if debug is enabled. Assorted cleanups.
* Fix codacy issues, remove unused methods.
* Only bind vue to the importer for now.
* Load vue for passport as well.
