# Passing files between jobs

`inputs` gives a job the result files of your earlier finished jobs without downloading and uploading them again.

## Use the results of another job

**Python**

```
from wbxsandbox import run_job, job_input

prep = run_job(command="python prepare.py", save_artifacts=True).wait()
report = run_job(command="python report.py input/data.csv",
                 inputs=[job_input(prep.id, "output/data.csv", "data.csv")])
```

**curl**

```
curl -s -X POST https://sandbox.wbx.cz/v1/jobs -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"command": "python report.py input/data.csv",
       "inputs": [{"job_id": "JOB_ID", "name": "output/data.csv", "as": "data.csv"}]}'
```

## How it works

- Each input is `{"job_id": "…", "name": "output/data.csv", "as": "data.csv"}`; `as` is optional (the base name is used).
- A job in a virtual machine finds them in `./input/` of its working directory, a [WebAssembly job](https://sandbox.wbx.cz/docs/jobs/wasm.md) in `/input/`.
- Up to 32 files and 64 MB per job. Need more? Pack the files into one archive in the earlier job.
- The SHA-256 is checked when the job is accepted and again when it starts, and the job's [receipt](https://sandbox.wbx.cz/docs/jobs/receipts.md) lists each input file with its SHA-256.
- Only your own jobs can be used, and only while their results are kept (7 days).
- Inside a [batch](https://sandbox.wbx.cz/docs/jobs/batches.md), `from_index` instead of `job_id` refers to another job of the same batch, and the job waits for it.

Source: https://sandbox.wbx.cz/docs/jobs/inputs
