fimidara can resize and convert images on the fly when you read a file. Pass transform query parameters to GET /v1/files/readFile/... — you do not need to store multiple sizes upfront.
Transforms apply only when the file is a processable image. Animated GIF/WebP and multi-page TIFF are not fully supported yet; requesting format=gif encodes a still image.
You can call readFile with:
Authorization: Bearer <agent-token-jwt> header, or<img src="..."> and public links.See also Introduction for how agent tokens and the Public permission group work.
| Param | Description |
|---|---|
w | Target width in pixels |
h | Target height in pixels |
fit | How to fit both dimensions: cover, contain, fill, inside, outside (default cover) |
pos | Anchor when fit is cover or contain (e.g. centre, top, entropy, attention) |
bg | Background color when fit is contain (e.g. %23ffffff for #ffffff) |
withoutEnlargement | If true, do not upscale smaller images |
format | Output format: jpeg, png, webp, avif, gif, tiff |
Full field definitions (including enums) are on the generated readFile GET docs.
With the JS SDK, build the same URL using:
const url = getFimidaraReadFileURL({
filepath: "my-workspace/photos/hero.png",
width: 600,
height: 400,
fit: "cover",
position: "centre",
background: "#ffffff",
withoutEnlargement: true,
format: "webp"
});Browser <img> tags cannot send an Authorization header. Issue a presigned path with issuePresignedPath, then use that id as the filepath in getFimidaraReadFileURL / readFile.
Note: duration on issuePresignedPath is in milliseconds (for example, 3_600_000 for one hour). Omit duration if the path should not expire.
In the fimidara app, open an image file menu and choose Image Transform to preview query parameters and copy a request URL.