oceanprotocol-job-details 0.3.1__tar.gz → 0.3.3__tar.gz

{oceanprotocol_job_details-0.3.1 → oceanprotocol_job_details-0.3.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: oceanprotocol-job-details
-Version: 0.3.1
+Version: 0.3.3
 Summary: A Python package to get details from OceanProtocol jobs
 Project-URL: Homepage, https://github.com/AgrospAI/oceanprotocol-job-details
 Project-URL: Issues, https://github.com/AgrospAI/oceanprotocol-job-details/issues
@@ -30,24 +30,29 @@ A Python package to get details from OceanProtocol jobs
 
 ## Installation
 
-```
+```bash
 pip install oceanprotocol-job-details
 ```
 
-## Usage 
+```bash
+uv add oceanprotocol-job-details
+```
+
+## Usage
 
-As a simple library, we only need to import `JobDetails` and load it, it will:
+As a simple library, we only need to import `load_job_details` and run it. It will:
 
-1. Fetch the needed parameters to populate the `JobDetails` instance from the environment variables or use the passed values to the `load()` method.
+1. Fetch the needed parameters to populate the `JobDetails` instance from the environment variables or use the passed values to the function.
 1. Look for the files corresponding to the passed DIDs in the filesystem according to the [Ocean Protocol Structure](#oceanprotocol-structure) and load them into the `JobDetails` instance.
 
-
 ### Minimal Example
 
 ```python
-from oceanprotocol_job_details import JobDetails
+from oceanprotocol_job_details import load_job_details
 
-job_details = JobDetails.load()
+class InputParameters(BaseModel): ...
+
+job_details = load_job_details({}, InputParameters)
 ```
 
 ### Custom Input Parameters
@@ -55,58 +60,40 @@ job_details = JobDetails.load()
 If our algorithm has custom input parameters and we want to load them into our algorithm, we can do it as follows:
 
 ```python
-from dataclasses import dataclass
-from oceanprotocol_job_details import JobDetails
-
-
-@dataclass
-class InputParameters:
-    foobar: str
-
-
-job_details = JobDetails[InputParameters].load(InputParameters)
-
-# Usage
-job_details.input_parameters.foobar
-```
-
-```python
-from dataclasses import dataclass
-from oceanprotocol_job_details import JobDetails
+from pydantic import BaseModel
+from oceanprotocol_job_details import load_job_details
 
 
-@dataclass
-class Foo:
+class Foo(BaseModel):
     bar: str
 
 
-@dataclass
-class InputParameters:
+class InputParameters(BaseModel):
     # Allows for nested types
     foo: Foo
 
 
-job_details = JobDetails[InputParameters].load(InputParameters)
+job_details = load_job_details({}, InputParameters)
 
 # Usage
+job_details.input_parameters.foo
 job_details.input_parameters.foo.bar
 ```
 
-The values to fill the custom `InputParameters` will be parsed from the `algoCustomData.json` located next to the input data directories. 
+The values to fill the custom `InputParameters` will be parsed from the `algoCustomData.json` located next to the input data directories.
 
 ### Iterating Input Files the clean way
 
 ```python
-from oceanprotocol_job_details import JobDetails
+from oceanprotocol_job_details import load_job_details
 
 
-job_details = JobDetails.load()
+job_details = load_job_details
 
-for idx, file_path in job_details.next_file():
+for idx, file_path in job_details.inputs():
     ...
 
-# Or if you just want one file path
-_, file_path = job_details.next_file()
+_, file_path = next(job_details.inputs())
 ```
 
 ## OceanProtocol Structure
@@ -124,4 +111,4 @@ data        # Root /data directory
 └── outputs # Algorithm output files dir
 ```
 
-> **_Note:_** Even though it's possible that the algorithm is passed multiple datasets, right now the implementation only allows to use **one dataset** per algorithm execution, so **normally** the executing job will only have **one ddo**, **one dir** inside inputs, and **one data file** named `0`.
+> **_Note:_** Even though it's possible that the algorithm is passed multiple datasets, right now the implementation only allows to use **one dataset** per algorithm execution, so **normally** the executing job will only have **one ddo**, **one dir** inside inputs, and **one data file** named `0`.

{oceanprotocol_job_details-0.3.1 → oceanprotocol_job_details-0.3.3}/README.md

@@ -4,24 +4,29 @@ A Python package to get details from OceanProtocol jobs
 
 ## Installation
 
-```
+```bash
 pip install oceanprotocol-job-details
 ```
 
-## Usage 
+```bash
+uv add oceanprotocol-job-details
+```
+
+## Usage
 
-As a simple library, we only need to import `JobDetails` and load it, it will:
+As a simple library, we only need to import `load_job_details` and run it. It will:
 
-1. Fetch the needed parameters to populate the `JobDetails` instance from the environment variables or use the passed values to the `load()` method.
+1. Fetch the needed parameters to populate the `JobDetails` instance from the environment variables or use the passed values to the function.
 1. Look for the files corresponding to the passed DIDs in the filesystem according to the [Ocean Protocol Structure](#oceanprotocol-structure) and load them into the `JobDetails` instance.
 
-
 ### Minimal Example
 
 ```python
-from oceanprotocol_job_details import JobDetails
+from oceanprotocol_job_details import load_job_details
 
-job_details = JobDetails.load()
+class InputParameters(BaseModel): ...
+
+job_details = load_job_details({}, InputParameters)
 ```
 
 ### Custom Input Parameters
@@ -29,58 +34,40 @@ job_details = JobDetails.load()
 If our algorithm has custom input parameters and we want to load them into our algorithm, we can do it as follows:
 
 ```python
-from dataclasses import dataclass
-from oceanprotocol_job_details import JobDetails
-
-
-@dataclass
-class InputParameters:
-    foobar: str
-
-
-job_details = JobDetails[InputParameters].load(InputParameters)
-
-# Usage
-job_details.input_parameters.foobar
-```
-
-```python
-from dataclasses import dataclass
-from oceanprotocol_job_details import JobDetails
+from pydantic import BaseModel
+from oceanprotocol_job_details import load_job_details
 
 
-@dataclass
-class Foo:
+class Foo(BaseModel):
     bar: str
 
 
-@dataclass
-class InputParameters:
+class InputParameters(BaseModel):
     # Allows for nested types
     foo: Foo
 
 
-job_details = JobDetails[InputParameters].load(InputParameters)
+job_details = load_job_details({}, InputParameters)
 
 # Usage
+job_details.input_parameters.foo
 job_details.input_parameters.foo.bar
 ```
 
-The values to fill the custom `InputParameters` will be parsed from the `algoCustomData.json` located next to the input data directories. 
+The values to fill the custom `InputParameters` will be parsed from the `algoCustomData.json` located next to the input data directories.
 
 ### Iterating Input Files the clean way
 
 ```python
-from oceanprotocol_job_details import JobDetails
+from oceanprotocol_job_details import load_job_details
 
 
-job_details = JobDetails.load()
+job_details = load_job_details
 
-for idx, file_path in job_details.next_file():
+for idx, file_path in job_details.inputs():
     ...
 
-# Or if you just want one file path
-_, file_path = job_details.next_file()
+_, file_path = next(job_details.inputs())
 ```
 
 ## OceanProtocol Structure
@@ -98,4 +85,4 @@ data        # Root /data directory
 └── outputs # Algorithm output files dir
 ```
 
-> **_Note:_** Even though it's possible that the algorithm is passed multiple datasets, right now the implementation only allows to use **one dataset** per algorithm execution, so **normally** the executing job will only have **one ddo**, **one dir** inside inputs, and **one data file** named `0`.
+> **_Note:_** Even though it's possible that the algorithm is passed multiple datasets, right now the implementation only allows to use **one dataset** per algorithm execution, so **normally** the executing job will only have **one ddo**, **one dir** inside inputs, and **one data file** named `0`.

oceanprotocol_job_details-0.3.3/oceanprotocol_job_details/__init__.py

@@ -0,0 +1,4 @@
+from .helpers import create_container, load_job_details
+from .ocean import JobDetails
+
+__all__ = [JobDetails, load_job_details, create_container]  # type: ignore

oceanprotocol_job_details-0.3.3/oceanprotocol_job_details/helpers.py

@@ -0,0 +1,29 @@
+from typing import Any, Dict, Type, TypeVar
+
+from pydantic import BaseModel, JsonValue
+
+from oceanprotocol_job_details.di import Container
+from oceanprotocol_job_details.ocean import JobDetails
+from oceanprotocol_job_details.settings import JobSettings
+
+InputParametersT = TypeVar("InputParametersT", bound=BaseModel)
+
+
+def create_container(config: Dict[str, Any]) -> Container[InputParametersT]:  # type: ignore[explicit-any]
+    """Return a fully configured Container from a config dict."""
+    container = Container[InputParametersT]()
+    settings = JobSettings(**config)
+    container.config.from_pydantic(settings)
+    return container
+
+
+def load_job_details(
+    config: Dict[str, JsonValue],
+    input_type: Type[InputParametersT],
+) -> JobDetails[InputParametersT]:
+    """
+    Load JobDetails for a given input_type using the config.
+    Returns a fully initialized JobDetails instance.
+    """
+    container: Container[InputParametersT] = create_container(config)
+    return container.job_details_loader(input_type=input_type).load()

{oceanprotocol_job_details-0.3.1 → oceanprotocol_job_details-0.3.3}/oceanprotocol_job_details/loaders/impl/files.py

@@ -20,14 +20,10 @@ class FilesLoader:
     dids: list[str]
     """Input DIDs"""
 
-    transformation_did: InitVar[str | None] = None
+    transformation_did: str
     """DID for the transformation algorithm"""
 
-    _transformation_did: str = field(init=False)
-
-    def __post_init__(self, transformation_did: str | None) -> None:
-        object.__setattr__(self, "_transformation_did", transformation_did)
-
+    def __post_init__(self) -> None:
         assert self.dids, "Missing input DIDs"
 
     def calculate_path(self, did: str, path_type: Literal["input", "ddo"]) -> Path:

{oceanprotocol_job_details-0.3.1 → oceanprotocol_job_details-0.3.3}/oceanprotocol_job_details/settings.py

@@ -1,22 +1,22 @@
 # mypy: disable-error-code=call-overload
 from logging import Logger, getLogger
 from pathlib import Path
+from typing import Self
 
 import orjson
-from pydantic import Field, field_validator
+from pydantic import Field, field_validator, model_validator
 from pydantic_settings import BaseSettings, SettingsConfigDict
 
 
 class JobSettings(BaseSettings):  # type: ignore[explicit-any]
     base_dir: Path = Field(alias="BASE_DIR")
-    dids: list[str] = Field(alias="DIDS")
+    dids: list[str] = Field(default_factory=list, alias="DIDS")
     transformation_did: str = Field(alias="TRANSFORMATION_DID")
     secret: str | None = Field(default=None, alias="SECRET")
     logger: Logger = Field(default_factory=lambda: getLogger(__name__))
 
     model_config = SettingsConfigDict(
         extra="forbid",
-        validate_default=True,
         populate_by_name=True,
         arbitrary_types_allowed=True,
     )
@@ -29,3 +29,9 @@ class JobSettings(BaseSettings):  # type: ignore[explicit-any]
             assert isinstance(data, list)
             return data
         return v
+
+    @model_validator(mode="after")
+    def validate_dids(self) -> Self:
+        if not self.dids:
+            self.dids.extend([f.name for f in (self.base_dir / "ddos").glob("*")])
+        return self

{oceanprotocol_job_details-0.3.1 → oceanprotocol_job_details-0.3.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "oceanprotocol-job-details"
-version = "0.3.1"
+version = "0.3.3"
 description = "A Python package to get details from OceanProtocol jobs"
 authors = [
     { name = "Agrospai", email = "agrospai@udl.cat" },

oceanprotocol_job_details-0.3.1/oceanprotocol_job_details/__init__.py

@@ -1,4 +0,0 @@
-from .ocean import JobDetails
-
-
-__all__ = [JobDetails]  # type: ignore