Info & bulk

Two helpers go with the 10 tools: inspect an image, and run one operation over many images at once.

get_image_info

get_image_info(input, &CommonOptions) -> ImageInfo

Language	Name
Rust	get_image_info
JavaScript	getImageInfo
Python	get_image_info
CLI	dusrelimage info <input> --json

ImageInfo

type ImageInfo = {
  format: string;       // "jpeg" | "png" | "webp" | "avif" | "gif" | "bmp" | "tiff"
  width: number;        // pixels
  height: number;       // pixels
  has_alpha: boolean;   // does it carry transparency?
  colour_type: string;  // e.g. "rgb8", "rgba8", "luma8"
  file_size: number;    // input bytes
};

Examples

import { getImageInfo } from 'dusrelimage';

const info = getImageInfo(input);
console.log(`${info.format} ${info.width}×${info.height}, alpha: ${info.has_alpha}`);

$ dusrelimage info photo.jpg --json
{"format":"jpeg","width":4000,"height":3000,"has_alpha":false,"colour_type":"rgb8","file_size":2456789}

bulk_process

bulk_process(inputs: &[Vec<u8>], op: &BulkOp) -> Vec<BulkItemResult>

One call, one result per input, same order. A failure on one item is reported for that item only — it never aborts the batch.

Language	Name
Rust	bulk_process
JavaScript	bulkProcess
Python	bulk_process
CLI	dusrelimage compress *.jpg --quality 70 (shell glob; one result per file)

BulkResult

type BulkResult = {
  ok: boolean;
  output?: Uint8Array;   // present when ok
  error?: {              // present when !ok
    code: string;        // e.g. "INVALID_IMAGE"
    message_en: string;
    message_hi: string;
  };
};

Example

import { bulkProcess } from 'dusrelimage';

const results = bulkProcess(files, 'compress', { quality: 70 });
results.forEach((r, i) => {
  if (r.ok) writeFile(`out-${i}.jpg`, r.output);
  else console.warn(`file ${i} failed: ${r.error.code} — ${r.error.message_en}`);
});

from dusrelimage import bulk_process

results = bulk_process(files, "resize", {"width": 800})
good = [r["output"] for r in results if r["ok"]]
bad = [(i, r["error"]["code"]) for i, r in enumerate(results) if not r["ok"]]

mock — generate test images

The CLI’s mock subcommand generates deterministic test images (solid, gradient, or photo-like noise) so you can try the tools with no download:

dusrelimage mock --kind gradient -o test.png
dusrelimage mock --kind solid --width 800 --height 600 -o solid.png
dusrelimage mock --kind noise --seed 42 -o noise.png

Fixtures for the shared test vectors (shared/test-vectors.json) are generated this way with a fixed seed — never committed as binary blobs (except 2 tiny PNGs, under 50 KB total).

Partial failure is the point

bulk_process exists so a batch job over user uploads does not die on one corrupt file. If you need all-or-nothing behaviour, check every result’s ok flag yourself and abort — the core will not do it for you.

Errors

Code	When
INVALID_IMAGE	get_image_info on undecodable input.
IMAGE_TOO_LARGE	Over max_pixels (default 120,000,000).
EMPTY_INPUT	Empty input bytes. (An empty input list for bulk_process is not an error — it returns an empty result list.)