High-fidelity document parsing for AI, self-hosted

# 1. Submit. FineParser accepts the file and returns a job ID right away.
curl -s -X POST http://localhost:8080/parse \
  -F "file=@invoice.pdf" \
  -F "outputType=doclang"
# {"jobId":"6c4f1f0e-…","file":"invoice.pdf"}

# 2. Poll until status is "successful".
curl -s http://localhost:8080/jobs/6c4f1f0e-…
# {"status":"successful","file":"6c4f1f0e-…/invoice.pdf.doclang"}

# 3. Download the result.
curl -OJ http://localhost:8080/jobs/6c4f1f0e-…/invoice.pdf.doclang

import time
import requests

BASE = "http://localhost:8080"

with open("invoice.pdf", "rb") as f:
    job = requests.post(
        BASE + "/parse",
        files={"file": f},
        data={"outputType": "json"},
    ).json()

while True:
    status = requests.get(BASE + "/jobs/" + job["jobId"]).json()
    if status["status"] not in ("pending", "in-progress"):
        break
    time.sleep(1)

if status["status"] != "successful":
    raise RuntimeError(status.get("error", status["status"]))

document = requests.get(BASE + "/jobs/" + status["file"]).json()

import { openAsBlob } from "node:fs";
import { setTimeout as sleep } from "node:timers/promises";

const BASE = "http://localhost:8080";

const form = new FormData();
form.append("file", await openAsBlob("invoice.pdf"), "invoice.pdf");
form.append("outputType", "doclang");

const { jobId } = await (await fetch(BASE + "/parse", { method: "POST", body: form })).json();

let job;
do {
  await sleep(1000);
  job = await (await fetch(BASE + "/jobs/" + jobId)).json();
} while (job.status === "pending" || job.status === "in-progress");

if (job.status !== "successful") throw new Error(job.error ?? job.status);

const doclang = await (await fetch(BASE + "/jobs/" + job.file)).text();

# One-shot: parse a file and exit. Format comes from the output extension.
docker run --rm \
  -e FINEPARSER_LICENSE_DATA="$(cat acme.fineparserlicense)" \
  -v "$PWD:/data" \
  abbyyteam/fineparser /data/invoice.pdf /data/invoice.doclang