feat(starrynight): update modules and pipelines to comply with new spec

starrynight/src/starrynight/experiments/pcp_generic.py

@@ -5,6 +5,7 @@
 from typing import Self
 
 import polars as pl
+from cloudpathlib import CloudPath
 from pydantic import BaseModel, Field
 
 from starrynight.experiments.common import (
@@ -24,6 +25,7 @@ class SBSConfig(BaseModel):
     img_frame_type: ImageFrameType = Field(ImageFrameType.ROUND)
     channel_dict: dict
     acquisition_order: AcquisitionOrderType = Field(AcquisitionOrderType.SNAKE)
+    barcode_csv_path: Path | CloudPath
 
 
 class CPConfig(BaseModel):
@@ -36,15 +38,25 @@ class CPConfig(BaseModel):
     acquisition_order: AcquisitionOrderType = Field(AcquisitionOrderType.SNAKE)
 
 
+class PCPGenericInitConfig(BaseModel):
+    """PCP generic config for auto init from index."""
+
+    barcode_csv_path: Path | CloudPath
+
+
 class PCPGeneric(Experiment):
     """PCP experiment configuration."""
 
-    path_parser: Callable[[str], MeasuredInventory]
     sbs_config: SBSConfig
     cp_config: CPConfig
+    path_parser: Callable[[str], MeasuredInventory] | None
+
+    # users should not access it directly
+    _init_config: PCPGenericInitConfig | None
 
     @staticmethod
-    def from_index(index_path: Path) -> Self:
+    def from_index(index_path: Path, init_config: PCPGenericInitConfig) -> Self:
+        """Configure experiment with index."""
         if index_path.name.endswith(".csv"):
             index_df = pl.scan_csv(index_path)
         else:

starrynight/src/starrynight/modules/common.py

@@ -37,7 +37,12 @@ class Config:
 
     @staticmethod
     @abstractmethod
-    def from_config(experiment: Experiment, data: DataConfig, **kwargs: Unpack) -> Self:
+    def from_config(
+        experiment: Experiment,
+        data: DataConfig,
+        updated_spec_dict: dict[str, Container] = {},
+        **kwargs: Unpack,
+    ) -> Self:
         """Create module from experiment and data config."""
         raise NotImplementedError
 

starrynight/src/starrynight/modules/gen_index.py

@@ -9,8 +9,14 @@
 
 from starrynight.experiments.common import Experiment
 from starrynight.modules.common import StarrynightModule
-from starrynight.modules.schema import Container as SpecContainer
-from starrynight.modules.schema import ExecFunction, TypeCitations
+from starrynight.modules.schema import (
+    Container as SpecContainer,
+)
+from starrynight.modules.schema import (
+    ExecFunction,
+    TypeAlgorithmFromCitation,
+    TypeCitations,
+)
 from starrynight.schema import DataConfig
 
 
@@ -41,12 +47,14 @@ def create_work_unit_gen_index(out_dir: Path | CloudPath) -> list[UnitOfWork]:
 
 
 def create_pipe_gen_index(
-    inventory_path: Path | CloudPath, out_dir: Path | CloudPath
+    uid: str, inventory_path: Path | CloudPath, out_dir: Path | CloudPath
 ) -> Pipeline:
     """Create pipeline for gen index.
 
     Parameters
     ----------
+    uid: str
+        Module unique id.
     inventory_path : Path | CloudPath
         Inventory path. Can be local or cloud.
     out_dir : Path | CloudPath
@@ -61,7 +69,7 @@ def create_pipe_gen_index(
     gen_index_pipe = Seq(
         [
             Container(
-                "GenerateIndex",
+                name=uid,
                 input_paths={"inventory": [inventory_path.resolve().__str__()]},
                 output_paths={
                     "index": [out_dir.joinpath("index.parquet").resolve().__str__()]
@@ -113,9 +121,17 @@ def from_config(
                 ),
                 docker_image=None,
                 algorithm_folder_name=None,
-                citations=TypeCitations(algorithm=[]),
+                citations=TypeCitations(
+                    algorithm=[
+                        TypeAlgorithmFromCitation(
+                            name="Starrynight indexing module",
+                            description="This module generates an index for the dataset.",
+                        )
+                    ]
+                ),
             )
         pipe = create_pipe_gen_index(
+            uid=GenIndexModule.uid(),
             inventory_path=data.dataset_path,
             out_dir=data.storage_path.joinpath("inventory"),
         )

starrynight/src/starrynight/modules/gen_inv.py

@@ -9,9 +9,12 @@
 
 from starrynight.experiments.common import Experiment
 from starrynight.modules.common import StarrynightModule
-from starrynight.modules.schema import Container as SpecContainer
+from starrynight.modules.schema import (
+    Container as SpecContainer,
+)
 from starrynight.modules.schema import (
     ExecFunction,
+    TypeAlgorithmFromCitation,
     TypeCitations,
     TypeEnum,
     TypeInput,
@@ -47,12 +50,17 @@ def create_work_unit_gen_inv(out_dir: Path | CloudPath) -> list[UnitOfWork]:
 
 
 def create_pipe_gen_inv(
-    spec: SpecContainer, dataset_path: Path | CloudPath, out_dir: Path | CloudPath
+    uid: str,
+    spec: SpecContainer,
+    dataset_path: Path | CloudPath,
+    out_dir: Path | CloudPath,
 ) -> Pipeline:
     """Create pipeline for gen inv.
 
     Parameters
     ----------
+    uid: str
+        Module unique id.
     spec: SpecContainer
         GenInvModule specification.
     dataset_path : Path | CloudPath
@@ -83,7 +91,7 @@ def create_pipe_gen_inv(
     gen_inv_pipe = Seq(
         [
             Container(
-                "GenerateInventory",
+                name=uid,
                 input_paths={},
                 output_paths={"inventory": [spec.outputs[0].path]},
                 config=ContainerConfig(
@@ -145,7 +153,14 @@ def _spec() -> SpecContainer:
             ),
             docker_image=None,
             algorithm_folder_name=None,
-            citations=TypeCitations(algorithm=[]),
+            citations=TypeCitations(
+                algorithm=[
+                    TypeAlgorithmFromCitation(
+                        name="Starrynight inventory module",
+                        description="This module generates an inventory for the dataset.",
+                    )
+                ]
+            ),
         )
 
     @staticmethod
@@ -161,7 +176,7 @@ def from_config(
                 .resolve()
                 .__str__()
             )
-        pipe = create_pipe_gen_inv(spec=spec)
+        pipe = create_pipe_gen_inv(uid=GenInvModule.uid(), spec=spec)
         uow = create_work_unit_gen_inv(out_dir=data.storage_path.joinpath("inventory"))
 
         return GenInvModule(spec=spec, pipe=pipe, uow=uow)

starrynight/src/starrynight/modules/registry.py

@@ -4,7 +4,7 @@
 from starrynight.modules.gen_index import GenIndexModule
 from starrynight.modules.gen_inv import GenInvModule
 
-MODULE_REGISTRY: dict[str, StarrynightModule] = {
+MODULE_REGISTRY: dict[str, type(StarrynightModule)] = {
     GenInvModule.uid(): GenInvModule,
     GenIndexModule.uid(): GenIndexModule,
 }