V8-archive: Expand EXIF and XMP metadata capabilities

[asitemade4u]

Feature Request

Currently, as written in the doc, Directus scrapes uploaded images for EXIF title, location, tags (keywords), and description.

1. This is NOT what I see happening. I attach an image with a lot of EXIF metadata (including Description), and yet none is scraped.

I also attach the metadata in the image, as output by EXIV2:
metadata.txt

1. Now, is there a way to scrape MORE metadata, notably XMP -- which is widely used by asset management software such as darktable or digikam?

What problem does this feature solve?

Use images themselves (and not XMP sidecar files) to convey information to the database, and ultimately to the Web publication (useful for copyright, captioning, etc.).

How do you think this should be implemented?

By allowing developers to surcharge the relevant interface by adding new metadata key-value pairs.

[rijkvanzanten]

@WoLfulus 🔔 really important to get this right for api-next

[rijkvanzanten]

To achieve better clarity/visibility, we are now tracking feature requests within the Feature Request project board.

This issue being closed does not mean it's not being considered.

[asitemade4u]

I can help as I have based the whole website development (which was
initially planned using WordPress) on the idea of saturating images with
information upstream, then transferring it downstream to the database
(there is a WP plugin for that).

On 3/20/20 5:50 PM, Rijk van Zanten wrote:
>

@WoLfulus https://github.com/WoLfulus 🔔 really important to get
this right for api-next

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
https://github.com/directus/api/issues/1813#issuecomment-601925006,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ACXN4CX5FXESUIHG2IFWSQ3RIPQI7ANCNFSM4LQVETDA.

[benhaynes]

Thanks @asitemade4u — and really good catch! I wonder what the overlap is between EXIF and IPTC metadata. To my knowledge, we scraped the latter.

In any event, we should be pulling as much information as possible from these sources automatically. We do have predefined fields for the key data, and I think adding the rest into the JSON metadata field (in Directus Files) might be a good option. Otherwise, we would need some sort of "mapping" system (or datatypes?) within Settings to map each value scraped to an individual field.