When copying images from firefox to telegram, the latter does not insert the image data as a photo, but instead downloads the same image by url. This wastes twice the amount of traffic on re-downloading the picture.
In addition, this creates a problem when copying pictures from sites protected by authorization, because telegram does not have the same cookies - instead, it downloads the 403 page as a file and tries to send it.

Steps to reproduce (example 1)

1. Take, for example, large photo from wikipedia (~80 MB): https://upload.wikimedia.org/wikipedia/commons/e/ec/Mona_Lisa%2C_by_Leonardo_da_Vinci%2C_from_C2RMF_retouched.jpg
2. Right-click it and select "Copy image"
3. Paste it into telegram input field
4. Observe how long it is pasted

Steps to reproduce (example 2)

(You'll need account on pixiv.net - it's free)

1. Navigate to a pixiv image, e.g.: https://www.pixiv.net/en/artworks/79037838
2. Click on the artwork to enlarge it
3. Right-click it and select "Copy image"
4. Paste it into telegram input field
5. Image is not copy-pasted, instead a 146 bytes file is prompted to send, which contains "403 Forbidden" message

Expected behaviour

Image is copy-pasted without redownload.

Actual behaviour

Telegram forces it to download again.

Configuration

Operating system: Windows 10 20H2

Version of Telegram Desktop: 2.6.1

Installation source (Linux Only) - the official website / GitHub releases / flatpak / snap / distribution package: -

Used theme: -