Merge conflict Gerrit cannot merge this change. It needs a rebase onto main.

Review main Work in progress Orphaned

91409[WIP FEATURE] Introduce a Digital Asset Provider API in TYPO3 Core

Presedo Roberto CR 0 V -1 Expert +2409 −0 updated 04/21/26, 15:16

Commit message

Currently, TYPO3 lacks a standard mechanism to
integrate external media libraries (Shutterstock,
Unsplash, etc.) or internal DAM/PIM systems.
Editors can only browse files available through
the File Abstraction Layer (FAL).

Third-party extensions attempt to fill this gap
but must re-implement Core logic (File Browser,
metadata handling, authentication), resulting in:
- Duplicated logic across extensions
- Inconsistent editor interfaces
- Fragile integrations hard to maintain

This change introduces a Digital Asset Provider
API that defines a generic, extensible interface
for connecting to external asset libraries while
maintaining a consistent Core-level UX.

Implementation:
- Add DigitalAssetProviderInterface for:
  * Searching assets with pagination/filters
  * Retrieving metadata (title, description,
    license, preview URL, etc.)
  * Retrieving or importing asset files
- Introduce provider registry mechanism via
  Symfony service tags
- Extend File Selector/Element Browser to show
  digital asset providers alongside FAL storages
- Ensure unified UX and metadata mapping
- Support two modes:
  * Import: download and store in FAL
  * Reference: store remote reference/metadata

Third-party extensions now only need to:
- Implement the provider interface
- Handle authentication and API communication
- Return standardized asset data

Core handles UI integration, preview rendering,
and metadata mapping automatically.

Benefits:
- Unified experience for all asset sources
- Reduced code duplication
- Easier development and maintenance
- Enables deprecation of stock-pictures and
  similar extensions

Test this change

Run a command in your clone of the TYPO3 Core. Then try what the change does and vote on Verified in Gerrit. How to review a patch

Check out patch set 2
git fetch https://review.typo3.org/Packages/TYPO3.CMS refs/changes/09/91409/2 && git checkout FETCH_HEAD
Cherry-pick onto main
git fetch https://review.typo3.org/Packages/TYPO3.CMS refs/changes/09/91409/2 && git cherry-pick FETCH_HEAD

Review

  • Code review CR 0 Needs a vote
  • Verified V -1 Blocked
    • core-ci -1 CI

Issues 1

Files 21

Show 1 more file

Review log 1012

  1. Presedo Roberto uploaded patch set 1 11/03/25, 14:07

  2. Mathias Brodala commented on patch set 1 11/03/25, 14:12

    Unresolved

    There must be a Feature RST explaining the new functionality. Also please check if you can add a DI compiler pass to autoregister providers which implement the relevant interface. This eliminates the need to manually register the providers.
  3. Mathias Brodala voted on patch set 1 CR -1 11/03/25, 14:12

  4. core-ci voted on patch set 1 V -1 11/03/25, 14:15 Core CI is not happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/88383

  5. Presedo Roberto updated the commit message in patch set 2 CR -1 V -1 11/03/25, 16:19

  6. core-ci voted on patch set 2 V -1 11/03/25, 16:27 Core CI is not happy: https://git.typo3.org/typo3/CI/cms/-/pipelines/88398

  7. Stefan Bürk commented on patch set 2 11/24/25, 20:20

    That's not true. TYPO3 calls this FAL (File Abstraction Layer). For example the 3rd party AWS S3 FAL Adapter can be used to have the same storage in TYPO3 and share it with PIM's, for example PIMCORE (which also supports S3 storages). If you want to adopt the PIMCORE DAM, you can implement a FAL driver consuming the PIMCORE API. Same counts for other systems. FAL is way more integrated throughout the whole TYPO3 system and API's, adding a new "dedicated and limited" API does not make much sense. There was somewhere a POC to make use of thepleague/flysystem (around FAL or as part of FAL) to have a more streamlined API integrating/connecting the already existing FAL parts ... that would be a more suitable aproach instead of adding a complete different API and hacking all the places to do additional stuff. It could be possible that I don't get the whole picture, maybe because of lacking a "real" implementation based on this interface (POC/demo) ? This approach also leaves permission (BE User/Group) out and similar things. Implementing a custom FAL driver would distribute the whole thing throughout the core without needing to adding everywhere additional handlings and hacks.
  8. Simon Schaufelberger commented on patch set 2 11/24/25, 20:27

    Did you implement everything yourself or did you get inspired by Neos? They have that feature already.
  9. Presedo Roberto commented on patch set 2 11/25/25, 07:28

    Reply Unresolved

    no, I didn't draw inspiration from Neos. This approach emerged independently from analyzing gaps in TYPO3's current editor experience.

    Reply

    You're absolutely right that FAL is deeply integrated throughout TYPO3, and extending it rather than creating a parallel API would be more maintainable. I acknowledge that my current approach of modifying Core in multiple places is not ideal. What I'm actually trying to solve: My primary goal isn't to replace FAL, but to unify the backend UI for asset selection. Currently, editors face fragmented experiences: separate "Add Image", "Add Video", "Add Shutterstock Media" buttons under each media field. My intention is to consolidate these into a single, unified asset picker that can source from multiple providers (FAL storages, external services, etc.). The secondary goal is to make it easier for third-party developers to integrate external asset libraries (Shutterstock, Unsplash, internal DAM systems) without re-implementing file browser logic or authentication each time. Editors should be able to search and select assets from these external libraries directly within the TYPO3 backend, without having to leave the interface. I've created a video demonstrating the concept: https://www.youtube.com/watch?v=6kiijNcfY1w Current limitations I'm aware of: - Permissions (BE User/Group) are not yet handled - This is still WIP and will require significant refinement - The approach of modifying Core in multiple places needs reconsideration I'm open to the FAL-first approach you suggest, but I'd like to understand: would a custom FAL driver truly solve the UI unification problem and enable seamless asset search/selection from external libraries? My concern is that extending FAL alone might not give us a cohesive backend experience without similar Core-level modifications. I'm discussing this work with Benni and welcome further architectural guidance. The video above and a more concrete POC/demo showing integration (e.g., Shutterstock) would likely clarify the approach and whether this should be FAL-driver-based or something complementary.
  10. Garvin Hicking commented on patch set 2 11/25/25, 08:53

    Unresolved

    Thanks for contributing such a large feature and improving our media experience. This looks like a huge effort and invest from you! I think this could be very helpful for streamlined UI integration. I am unsure how I could test your patch, do you have an example DAM integration with which we could test this? I haven't thoroughly looked into this yet but reading the code it feels like I would need an extension counterpart to utilize the new code? Also I believe the javascript would benefit from utilizing lit / web components instead of manually creating DOM elements, but that's not a domain I'm experienced in.
  11. Oli Bartsch commented on patch set 2 11/27/25, 14:50

    Reply Unresolved

    @stefan@buerk.tech As far as i understand, this is not about some new FAL driver implementation but to allow easier selection / uploading of files next to the standard "Add online media", "Add from your computer", etc. It is not about browsing / integrating a different driver.
  12. Benni Mack marked it work in progress 04/21/26, 15:16