# Projects and Libraries

Monata organizes reusable circuit designs with a Library/Cell/View hierarchy.
This is the main framework boundary: simulator backends and model compilers are
attached to this structure, but they do not replace it.

## Library Registry

`LibraryRegistry` is the top-level entry point. It keeps track of library
directories and gives you dictionary-style access to libraries.

```python
from monata import LibraryRegistry

reg = LibraryRegistry()
lib = reg.create_library(
    path="work/cmos_cells",
    name="cmos_cells",
    tech_model_paths=["models/ptm_45nm_nmos.lib", "models/ptm_45nm_pmos.lib"],
)

assert "cmos_cells" in reg
lib = reg["cmos_cells"]
```

## Library Layout

A Monata library is a directory with a `lib.toml` file and one directory per
cell.

```text
cmos_cells/
  lib.toml
  inverter/
    cell.toml
    schematic.py
    testbench.py
    symbol.toml
    netlist.scs
```

`lib.toml` stores library-level metadata and technology settings:

```toml
[library]
name = "cmos_cells"
description = "CMOS standard-cell experiments"

[technology]
model_paths = ["models/ptm_45nm_nmos.lib", "models/ptm_45nm_pmos.lib"]
simulator = "ngspice-subprocess"
node_type = "bulk"
```

## Cells

A `Cell` represents one design unit, such as an inverter, buffer, gate,
amplifier, or test macro.

```python
cell = lib.create_cell("inverter", description="CMOS inverter")
assert "inverter" in lib
```

Each cell owns its views. The idiomatic access pattern is:

```python
cell = reg["cmos_cells"]["inverter"]
schematic = cell["schematic"]
```

## Views

Monata currently recognizes these view types:

| View | Purpose |
| --- | --- |
| `schematic` | Native Python `monata.netlist` source for the circuit topology |
| `testbench` | Python function that wraps a circuit in sources, loads, and analyses |
| `netlist` | Generated or hand-written SPICE/Spectre-compatible netlist |
| `symbol` | Pin interface for hierarchical use |

Views are declared in `cell.toml`:

```toml
[cell]
name = "inverter"
description = "CMOS inverter"

[views]
schematic = { entry = "schematic.py", class = "Inverter" }
testbench = { entry = "testbench.py", function = "main" }
symbol = { entry = "symbol.toml", generated = true }
netlist = { entry = "netlist.scs", generated = true }
```

Generated views are protected. If a generated netlist or symbol was edited by
hand, regenerate with `force=True` only when you intend to overwrite it.

```python
cell.generate_symbol()
cell.generate_netlist()

cell.generate_netlist(force=True)
```