Merge pull request #15 from gabrielguarisa/f/update-pydantic

Update pydantic to v2

.github/workflows/checks.yml

@@ -16,7 +16,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: [3.7.16, 3.11]
+        python-version: [3.9, 3.11]
         poetry-version: [1.4.2]
         os: [ubuntu-latest]
     runs-on: ${{ matrix.os }}
@@ -47,7 +47,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: [3.9]
+        python-version: [3.11]
         poetry-version: [1.4.2]
         os: [ubuntu-latest]
     runs-on: ${{ matrix.os }}

.github/workflows/release.yml

@@ -9,7 +9,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: [3.7.16]
+        python-version: [3.11]
         poetry-version: [1.4.2]
         os: [ubuntu-latest]
     runs-on: ${{ matrix.os }}

.gitignore

@@ -282,3 +282,4 @@ $RECYCLE.BIN/
 
 # Custom rules (everything added below won't be overriden by 'Generate .gitignore File' if you use 'Update' option)
 
+.vscode/settings.json

Makefile

@@ -4,13 +4,11 @@ init:
 
 .PHONY: formatting
 formatting:
-	poetry run isort --settings-path pyproject.toml ./
-	poetry run black --config pyproject.toml ./
+	poetry run ruff format .
 
 .PHONY: check-formatting
 check-formatting:
-	poetry run isort --settings-path pyproject.toml --check-only ./
-	poetry run black --config pyproject.toml --check ./
+	poetry run ruff check .
 
 .PHONY: tests
 tests:

README.md

@@ -40,14 +40,16 @@ import retrack
 parser = retrack.Parser(rule)
 
 # Create a runner
-runner = retrack.Runner(parser)
+runner = retrack.Runner(parser, name="your-rule")
 
 # Run the rule/model passing the data
 runner.execute(data)
 ```
 
 The `Parser` class parses the rule/model and creates a graph of nodes. The `Runner` class runs the rule/model using the data passed to the runner. The `data` is a dictionary or a list of dictionaries containing the data that will be used to evaluate the conditions and execute the actions. To see wich data is required for the given rule/model, check the `runner.request_model` property that is a pydantic model used to validate the data.
 
+Optionally you can name the rule by passing the `name` field to the `retrack.Runner` constructor. This is useful to identify the rule when exceptions are raised.
+
 ### Creating a rule/model
 
 A rule is a set of conditions and actions that are executed when the conditions are met. The conditions are evaluated using the data passed to the runner. The actions are executed when the conditions are met.

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "retrack"
-version = "0.11.1"
+version = "1.0.0-alpha.1"
 description = "A business rules engine"
 authors = ["Gabriel Guarisa <gabrielguarisa@gmail.com>", "Nathalia Trotte <nathaliatrotte@gmail.com>"]
 license = "MIT"
@@ -10,90 +10,21 @@ homepage = "https://github.com/gabrielguarisa/retrack"
 keywords = ["rules", "models", "business", "node", "graph"]
 
 [tool.poetry.dependencies]
-python = "^3.7.16"
-pandas = [
-  { version = "1.2.0", python = "<3.8" },
-  { version = "^1.2.0", python = ">=3.8" }
-]
-numpy = [
-  { version = "1.19.5", python = "<3.8" },
-  { version = "^1.19.5", python = ">=3.8" }
-]
-pydantic = "^1.10.4"
-networkx = [
-  { version = "2.6.3", python = "<3.8" },
-  { version = "^2.6.3", python = ">=3.8" }
-]
+python = ">=3.9,<4.0.0"
+pandas = "^1.2.0"
+numpy = "^1.19.5"
+pydantic = "2.4.2"
+networkx = "^2.6.3"
+pandera = "^0.17.2"
 
 [tool.poetry.dev-dependencies]
-pytest = [
-  { version = "6.2.2", python = "<3.8" },
-  { version = "^6.2.4", python = ">=3.8" }
-]
-pytest-cov = [
-  { version = "2.11.1", python = "<3.8" },
-  { version = "^3.0.0", python = ">=3.8" }
-]
-black = [
-  { version = "20.8b1", python = "<3.8" },
-  { version = "^22.6.0", python = ">=3.8" }
-]
-isort = {extras = ["colors"], version = "*"}
-pytest-mock = [
-  { version = "3.5.1", python = "<3.8" },
-  { version = "^3.10.0", python = ">=3.8" }
-]
-
-[tool.black]
-# https://github.com/psf/black
-target-version = ["py37"]
-line-length = 88
-color = true
-
-exclude = '''
-/(
-    \.git
-    | \.hg
-    | \.mypy_cache
-    | \.tox
-    | \.venv
-    | _build
-    | buck-out
-    | build
-    | dist
-    | env
-    | venv
-)/
-'''
-
-[tool.isort]
-# https://github.com/timothycrosley/isort/
-py_version = 37
-line_length = 88
-
-known_typing = [
-  "typing",
-  "types",
-  "typing_extensions",
-  "mypy",
-  "mypy_extensions",
-]
-sections = [
-  "FUTURE",
-  "TYPING",
-  "STDLIB",
-  "THIRDPARTY",
-  "FIRSTPARTY",
-  "LOCALFOLDER",
-]
-include_trailing_comma = true
-profile = "black"
-multi_line_output = 3
-indent = 4
-color_output = true
+pytest = "*"
+pytest-cov = "*"
+ruff = "^0.1.2"
+pytest-mock = "*"
 
 [tool.pytest.ini_options]
-addopts = "--junitxml=pytest.xml -p no:warnings --cov-report term-missing:skip-covered --cov=retrack"
+addopts = "-vv --junitxml=pytest.xml -p no:warnings --cov-report term-missing:skip-covered --cov=retrack"
 
 [build-system]
 requires = ["poetry-core>=1.0.0"]

retrack/engine/parser.py

@@ -1,7 +1,10 @@
 import typing
 
+import hashlib
+
 from retrack import nodes, validators
 from retrack.utils.registry import Registry
+import json
 
 
 class Parser:
@@ -28,11 +31,16 @@ def __init__(
         self._set_indexes_by_kind_map()
         self._set_execution_order()
         self._set_indexes_by_memory_type_map()
+        self._set_version()
 
     @property
     def graph_data(self) -> dict:
         return self.__graph_data
 
+    @property
+    def version(self) -> str:
+        return self._version
+
     @staticmethod
     def _check_input_data(data: dict):
         if not isinstance(data, dict):
@@ -166,7 +174,7 @@ def _set_execution_order(self):
     def get_node_connections(
         self, node_id: str, is_input: bool = True, filter_by_connector=None
     ):
-        node_dict = self.get_by_id(node_id).dict(by_alias=True)
+        node_dict = self.get_by_id(node_id).model_dump(by_alias=True)
 
         connectors = node_dict.get("inputs" if is_input else "outputs", {})
         result = []
@@ -200,3 +208,24 @@ def _walk(self, actual_id: str, skiped_ids: list):
                     self._walk(next_id, skiped_ids)
 
         return skiped_ids
+
+    def _set_version(self):
+        self._version = self.graph_data.get("version", None)
+
+        graph_json_content = (
+            json.dumps(self.graph_data["nodes"])
+            .replace(": ", ":")
+            .replace(", ", ",")
+            .encode("utf-8")
+        )
+        calculated_hash = hashlib.sha256(graph_json_content).hexdigest()[:10]
+
+        if self.version is None:
+            self._version = f"{calculated_hash}.dynamic"
+        else:
+            file_version_hash = self.version.split(".")[0]
+
+            if file_version_hash != calculated_hash:
+                raise ValueError(
+                    f"Invalid version. Graph data has changed and the hash is different: {calculated_hash} != {file_version_hash}"
+                )

retrack/engine/request_manager.py

@@ -1,5 +1,7 @@
 import typing
 
+import pandas as pd
+import pandera
 import pydantic
 
 from retrack.nodes.base import BaseNode, NodeKind
@@ -8,6 +10,7 @@
 class RequestManager:
     def __init__(self, inputs: typing.List[BaseNode]):
         self._model = None
+        self._dataframe_model = None
         self.inputs = inputs
 
     @property
@@ -41,13 +44,19 @@ def inputs(self, inputs: typing.List[BaseNode]):
 
         if len(self.inputs) > 0:
             self._model = self.__create_model()
+            self._dataframe_model = self.__create_dataframe_model()
         else:
             self._model = None
+            self._dataframe_model = None
 
     @property
     def model(self) -> typing.Type[pydantic.BaseModel]:
         return self._model
 
+    @property
+    def dataframe_model(self) -> pandera.DataFrameSchema:
+        return self._dataframe_model
+
     def __create_model(
         self, model_name: str = "RequestModel"
     ) -> typing.Type[pydantic.BaseModel]:
@@ -62,26 +71,45 @@ def __create_model(
         fields = {}
         for input_field in self.inputs:
             fields[input_field.data.name] = (
-                (str, ...)
-                if input_field.data.default is None
-                else (str, input_field.data.default)
+                str,
+                pydantic.Field(
+                    default=Ellipsis
+                    if input_field.data.default is None
+                    else input_field.data.default,
+                ),
             )
 
         return pydantic.create_model(
             model_name,
             **fields,
         )
 
+    def __create_dataframe_model(self) -> pandera.DataFrameSchema:
+        """Create a pydantic model from the RequestManager's inputs"""
+        fields = {}
+        for input_field in self.inputs:
+            fields[input_field.data.name] = pandera.Column(
+                str,
+                nullable=input_field.data.default is not None,
+                coerce=True,
+                default=input_field.data.default,
+            )
+
+        return pandera.DataFrameSchema(
+            fields,
+            index=pandera.Index(int),
+            strict=True,
+            coerce=True,
+        )
+
     def validate(
         self,
-        payload: typing.Union[
-            typing.Dict[str, str], typing.List[typing.Dict[str, str]]
-        ],
+        payload: pd.DataFrame,
     ) -> typing.List[pydantic.BaseModel]:
         """Validate the payload against the RequestManager's model
 
         Args:
-            payload (typing.Union[typing.Dict[str, str], typing.List[typing.Dict[str, str]]]): The payload to validate
+            payload (pandas.DataFrame): The payload to validate
 
         Raises:
             ValueError: If the RequestManager has no model
@@ -92,7 +120,7 @@ def validate(
         if self.model is None:
             raise ValueError("No inputs found")
 
-        if not isinstance(payload, list):
-            payload = [payload]
+        if not isinstance(payload, pd.DataFrame):
+            raise TypeError(f"payload must be a pandas.DataFrame, not {type(payload)}")
 
-        return pydantic.parse_obj_as(typing.List[self.model], payload)
+        return self.dataframe_model.validate(payload)