Image Pipeline Playground
"""Generate standalone Python scripts from a playground graph."""
from __future__ import annotations
from app.engine.executor import topological_sort, validate_graph
from app.engine.registry import registry
from app.models.graph import Graph
from app.nodes import register_builtin_nodes
def generate_python(graph: Graph, seed: int = 0) -> str:
register_builtin_nodes()
validate_graph(graph, registry)
order = topological_sort(graph)
nodes = {node.id: node for node in graph.nodes}
lines: list[str] = [
'"""Auto-generated Image Pipes pipeline."""',
"",
"from __future__ import annotations",
"",
"from pathlib import Path",
"",
"import cv2",
"import numpy as np",
"",
"",
f"def run(seed: int = {seed}) -> None:",
]
outputs_by_node: dict[str, dict[str, str]] = {}
for node_id in order:
instance = nodes[node_id]
impl = registry.get(instance.type)
params = impl.validate_params(instance.params)
safe_id = node_id.replace("-", "_")
input_vars: dict[str, str] = {}
for edge in graph.edges:
if edge.target != node_id:
continue
source_outputs = outputs_by_node.get(edge.source, {})
source_var = source_outputs.get(edge.source_port)
if source_var:
input_vars[edge.target_port] = source_var
output_ports = [port.id for port in impl.ports if port.direction.value == "output"]
if not output_ports:
output_ports = ["image"]
output_vars = {port: f"{safe_id}_{port}" for port in output_ports}
outputs_by_node[node_id] = output_vars
lines.append(f" # node: {node_id} ({instance.type})")
emitted = impl.emit_python(node_id, params, input_vars, output_vars)
for statement in emitted:
lines.append(f" {statement}")
primary = output_vars.get("image")
if primary:
lines.append(f" print('produced', '{node_id}', {primary}.shape)")
else:
for port, var_name in output_vars.items():
lines.append(f" print('produced', '{node_id}:{port}', {var_name}.shape)")
lines.extend(
[
"",
"",
'if __name__ == "__main__":',
" run()",
"",
]
)
return "\n".join(lines)