REST API

# Enviar
curl -s -X POST http://localhost:8080/parse \
  -F "file=@document.pdf" \
  -F "outputType=doclang"
# {"jobId":"6c4f1f0e-...","file":"document.pdf"}

# Consultar
curl -s http://localhost:8080/jobs/6c4f1f0e-...
# {"status":"successful","file":"6c4f1f0e-.../document.pdf.doclang"}

# Descargar
curl -OJ http://localhost:8080/jobs/6c4f1f0e-.../document.pdf.doclang

import time
import requests

BASE = "http://localhost:8080"

with open("document.pdf", "rb") as f:
    submitted = requests.post(
        f"{BASE}/parse",
        files={"file": f},
        data={"outputType": "doclang"},
    )
submitted.raise_for_status()
job_id = submitted.json()["jobId"]

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

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

result = requests.get(f"{BASE}/jobs/{job['file']}")
result.raise_for_status()
with open("document.doclang", "wb") as out:
    out.write(result.content)

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

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

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

const submitted = await fetch(`${BASE}/parse`, { method: "POST", body: form });
if (!submitted.ok) throw new Error(await submitted.text());
const { jobId } = await submitted.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 result = await fetch(`${BASE}/jobs/${job.file}`);
await writeFile("document.doclang", Buffer.from(await result.arrayBuffer()));