This is a list of image metadata for the test suite.
The files follow naming pattern XX/XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.json where Xs are lowercase hex digits of SHA-1 of the image file being described. The first two digits are used for subdirectory name.
The JSON contains fields:
sha1— SHA-1 hash of the file being described (lowercase hex, required).lic— License (required). Seelicenses.json. One of:pd— Public Domaincc0— Creative Commons Zerocc-by-2.5,cc-by-3.0— Creative Commons Attributioncc-by-sa-2.5,cc-by-sa-3.0— Creative Commons Attribution + Share Alike
url— URL where the image has been downloaded from.urls— Array of alternative locations where this image may be downloaded from.from— Identifier/name of source where this image was obtained from (seesources.json)tags— Array of arbitrary tags that describe content of this image. Tags with:are reserved for special purposes.desc— Object with keys for image description (including attribution) in one or more formats:text— Plain text (with newlines)html— HTML-formattedurl— URL to the description
name— Short name/title of the image. Generally derived from filename.
There is download.js script included in the yr package that will download all the images in an optimal way.
npm i yr
node ./node_modules/yr/download.js