Difference between revisions of "PVT API"
| Line 26: | Line 26: | ||
| DELETE || /pvt-calculator/{token} || Delete the job | | DELETE || /pvt-calculator/{token} || Delete the job | ||
|} | |} | ||
| + | |||
| + | Use /calc (synchronous) for normal interactive requests — a single PVT model computes in well under a second and the result comes back in the HTTP response. | ||
| + | |||
| + | Use /calc-async only for batch/heavy workloads where you would rather not hold an open HTTP connection. The async path enqueues an api_task row that is picked up by the CalcDaemon worker; you then poll status and fetch output-data. The async job runs the same calculatePvt() code as the sync endpoint, so the result is identical for identical input. | ||
| + | |||
[[Category:API]] | [[Category:API]] | ||
Revision as of 17:03, 14 July 2026
Contents
PVT API
The PVT calculator models reservoir fluid properties (oil, gas, water) over a pressure range using various correlations, and returns property curves (plots), single‑point reservoir‑condition values (solution), and a stepped property table (results).
Endpoints
All endpoints require the Authorization: Bearer <api_key> header and are subject to rate limiting. The response envelope is always { "success": <bool>, "data": <payload> }
| Method | Path | Purpose |
|---|---|---|
| POST | /pvt-calculator/calc | Synchronous — compute and return the result in the same request |
| POST | /pvt-calculator/calc-async | Asynchronous — enqueue a job, return a {token} |
| GET | /pvt-calculator/{token}/status | Job status and progress |
| GET | /pvt-calculator/{token}/output-data | Job result (same shape as the sync data) |
| GET | /pvt-calculator/{token}/input-data | The input the job was created with |
| POST | /pvt-calculator/{token}/start | Reset and re-run the job |
| POST | /pvt-calculator/{token}/stop | Stop a running job |
| DELETE | /pvt-calculator/{token} | Delete the job |
Use /calc (synchronous) for normal interactive requests — a single PVT model computes in well under a second and the result comes back in the HTTP response.
Use /calc-async only for batch/heavy workloads where you would rather not hold an open HTTP connection. The async path enqueues an api_task row that is picked up by the CalcDaemon worker; you then poll status and fetch output-data. The async job runs the same calculatePvt() code as the sync endpoint, so the result is identical for identical input.
