# Check
Pipeline validation catches structural errors before execution. The `check()` method on `PipelineInfo` walks the pipeline and verifies several invariants.
## Usage
```rust,ignore
let steps = pipeline(&catalog, ¶ms);
steps.check()?; // Result<(), CheckError>
```
Or via the CLI:
```sh
$ my_app check
Pipeline is valid.
```
## What check validates
### Sequential ordering
A node must not read a dataset that is only produced by a **later** node. Datasets that no node produces are treated as external inputs (valid).
```rust,ignore
// Valid: n1 produces a, n2 reads a
(
Node { name: "n1", input: (¶m,), output: (&a,), .. },
Node { name: "n2", input: (&a,), output: (&b,), .. },
)
// Invalid: n1 reads b, but b is produced by n2
(
Node { name: "n1", input: (&b,), output: (&a,), .. },
Node { name: "n2", input: (¶m,), output: (&b,), .. },
)
// → CheckError::InputNotProduced { node_name: "n1", .. }
```
### No duplicate outputs
A dataset must not be produced by more than one node:
```rust,ignore
(
Node { name: "n1", input: (¶m,), output: (&a,), .. },
Node { name: "n2", input: (¶m,), output: (&a,), .. }, // same output!
)
// → CheckError::DuplicateOutput { node_name: "n2", .. }
```
### Params are read-only
No node may write to a `Param` dataset:
```rust,ignore
(
Node { name: "n1", func: || ((),), input: (), output: (¶m,) },
)
// → CheckError::ParamWritten { node_name: "n1", .. }
```
### Pipeline contracts
For `Pipeline` structs, the declared inputs and outputs must match what the children actually consume and produce. See [Pipeline](./pipeline.md).
## CheckError variants
| `InputNotProduced` | Node reads a dataset produced by a later node |
| `DuplicateOutput` | Two nodes produce the same dataset |
| `ParamWritten` | A node writes to a `Param` |
| `UnusedPipelineInput` | Pipeline declares an input its children don't consume |
| `UnproducedPipelineOutput` | Pipeline declares an output its children don't produce |
| `CapacityExceeded` | Internal dataset buffer overflow (see below) |
## Capacity
`check()` uses a fixed-capacity buffer (default 20 datasets) for `no_std` compatibility. If your pipeline has more than 20 unique datasets, use `check_with_capacity`:
```rust,ignore
steps.check_with_capacity::<64>()?;
```
## no_std compatibility
`check()` works in `no_std` environments. It uses no allocation — all dataset tracking is done in fixed-size arrays on the stack.