Added inline documentation for the following filter widget events:
backend.filter.extendQuery
backend.filter.extendScopesBefore
backend.filter.extendScopes
Fixes#3627. This changes the logic for hiding tabs when all of their fields are hidden to specifically only target direct children of the tab pane when looking for visible fields because repeaters will have fields present that are only hidden by virtue of the parent field being hidden however the previous behaviour would consider it visible.
Fix pathing issue in media manager
Currently there is a way to manipulate folder names using the request headers for the medafinder folder path. since `validatePath` does not validate file names but rather only validates the existence of a folder in a parent/subdirectory. We can utilize `validateFileName` to verify if the name of the folder path being created is valid during the folder creation process and the name applying process.
Documented in e56468a95b. Credit to @manogi.
Added an option to change from the "search on every keystroke" functionality to one where the serach is only fired when hitting the enter key. This makes sense when having larger datasets or CMS users who are slow typers.
This changes Backend::dateTime to return an empty string instead of the current date when dealing with a null date. It should only affect the date field preview instance. Credit to @pvullioud
Fixes#3260. Enables setting the target field for the trigger API to a form context an arbitrary number of levels above the current field by prepending the field name with `^` for each level to go up. For instance, if you had a field [allow_pictures] in the base form and a [records] repeater with [image] fields, and you set the target field for the trigger on [image] to [^allow_pictures] then it would work properly instead of looking for an [allow_pictures] field on the [records] field. Credit to @MartynasJanu. Documented in edc78b584b
Credit @mcharytoniuk.
While importing over 250k records ImportModel kept running into various problems. One of them was too big memory usage - `ImportModel` loaded the complete file upfront (`$reader->fetchAll()`). Simple one-line change to `$reader->fetch()` makes `ImportModel` import CSV file row-by-row and returning an iterator which limits memory usage and allows data to be imported. This change optimizes memory usage and allows much simpler importing of larger files.
If the parent of the file attachment relationship already exists, then attach the file directly to it in this request instead of going through deferred bindings. Will continue to attach files through deferred binding if the parent relation doesn't exist.
Adds support to retrieve protected files using the class they are defined to be using if that class differs from the default System\Models\File class. This makes it possible to use a custom class extending the base file model class that does output processing on the file data (for example, an EncryptedFile class that has to decrypt the file contents before they can be output to the browser).
Stores the last visited page of list widgets in the session to restore to on next page load. Does not apply when filters / searches are applied. Credit to @tobias-kuendig
