readme.md

image database

vkdt's database backend is fairly minimal. in particular:

• it does not copy around your original data.
• it does not write to your original data.
• it merely writes *.cfg sidecar files with processing info.
  in fact these are the actual input files to a loaded directory. there is, however, a fallback to known raw files which will spawn default .cfg files by simply appending the suffix to the full name.
• it writes a minimal vkdt.db file to each directory, containing
  information about rating and labels.

thumbnails

vkdt stores thumbnails for lighttable view in .cache/vkdt/<murmur3-hash>.bc1. that is, they are compressed in bc1 format on the fly and also stored as such on disk. this is good for fast and compact display on gpu.

tags/collections

you can assign tags or images to named collections in lighttable mode. this will create (if it doesn't already exist) a directory in .config/vkdt/tags/<tagname>/ and create a symlink .cfg file in this directory, pointing to the images you assigned the tag to. you can then open all images with the given tag by pointing vkdt to this directory.

a collection created this way has its own vkdt.db file, so you can assign a different rating or labels when working on this collection. this means there is no concept of global rating/labels, but these are relative to the directory you are working on. this facilitates relative ratings (best out of a collection of all 5-star images for instance).

currently there is no (fast) way to see all labels assigned to a particular image.