PyGeoModel 1.0.6__tar.gz → 1.0.8__tar.gz

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/PKG-INFO

@@ -1,12 +1,11 @@
 Metadata-Version: 2.4
 Name: PyGeoModel
-Version: 1.0.6
+Version: 1.0.8
 Summary: A Python package for integrating OpenGMS geographic model services.
 Home-page: https://github.com/MpLebron/PyGeoModel
 Author: Peilong Ma
 Author-email: mpl_gis@nnu.edu.cn
 Keywords: geographic modeling,GIS,OpenGMS,model services,geospatial analysis,jupyter
-Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Science/Research
 Classifier: Topic :: Scientific/Engineering :: GIS
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -91,20 +90,6 @@ result = modeler.invoke(
 result.to_json("execution_record.json")
 ```
 
-OpenGMS execution uses the bundled public demo token by default. For stable or private use, configure your own OpenGMS token:
-
-```bash
-export OGMS_TOKEN="your-token"
-```
-
-Optional endpoint overrides:
-
-```bash
-export OGMS_BASE_PORTAL_URL="http://222.192.7.75"
-export OGMS_BASE_MANAGER_URL="http://222.192.7.75/managerServer"
-export OGMS_BASE_DATA_URL="http://222.192.7.75/dataTransferServer"
-```
-
 ## Notebook Interface
 
 ```python
@@ -124,18 +109,8 @@ answer = modeler.ask_model("SWAT_Model", "What input data are required?")
 answer.to_json("qa_record.json")
 ```
 
-The recommendation service automatically builds notebook/data context and calls the configured Dify workflow. `DIFY_API_KEY` is required for `suggest_model()` because the context-aware recommendation logic is implemented in that workflow. Q&A uses OpenGMS model metadata and, when configured, an OpenAI-compatible web-enabled model.
+The recommendation service automatically builds notebook/data context and calls the configured recommendation workflow. Q&A uses OpenGMS model metadata and an OpenAI-compatible web-enabled model. The main notebook workflow is designed to run out of the box for demonstration use.
 
 ## Relation to OpenGMS
 
 OpenGMS provides the model-service platform and online execution infrastructure. PyGeoModel is a Python client package that exposes OpenGMS model-service discovery, metadata inspection, task invocation, and result records to Python and notebook workflows.
-
-## Development
-
-Run the lightweight test suite with:
-
-```bash
-python -m unittest discover -s tests
-```
-
-The source distribution should not include real API keys, local `.env` files, bytecode caches, generated C files, `.pyd` binaries, or previous build artifacts.

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/PyGeoModel.egg-info/PKG-INFO

@@ -1,12 +1,11 @@
 Metadata-Version: 2.4
 Name: PyGeoModel
-Version: 1.0.6
+Version: 1.0.8
 Summary: A Python package for integrating OpenGMS geographic model services.
 Home-page: https://github.com/MpLebron/PyGeoModel
 Author: Peilong Ma
 Author-email: mpl_gis@nnu.edu.cn
 Keywords: geographic modeling,GIS,OpenGMS,model services,geospatial analysis,jupyter
-Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Science/Research
 Classifier: Topic :: Scientific/Engineering :: GIS
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -91,20 +90,6 @@ result = modeler.invoke(
 result.to_json("execution_record.json")
 ```
 
-OpenGMS execution uses the bundled public demo token by default. For stable or private use, configure your own OpenGMS token:
-
-```bash
-export OGMS_TOKEN="your-token"
-```
-
-Optional endpoint overrides:
-
-```bash
-export OGMS_BASE_PORTAL_URL="http://222.192.7.75"
-export OGMS_BASE_MANAGER_URL="http://222.192.7.75/managerServer"
-export OGMS_BASE_DATA_URL="http://222.192.7.75/dataTransferServer"
-```
-
 ## Notebook Interface
 
 ```python
@@ -124,18 +109,8 @@ answer = modeler.ask_model("SWAT_Model", "What input data are required?")
 answer.to_json("qa_record.json")
 ```
 
-The recommendation service automatically builds notebook/data context and calls the configured Dify workflow. `DIFY_API_KEY` is required for `suggest_model()` because the context-aware recommendation logic is implemented in that workflow. Q&A uses OpenGMS model metadata and, when configured, an OpenAI-compatible web-enabled model.
+The recommendation service automatically builds notebook/data context and calls the configured recommendation workflow. Q&A uses OpenGMS model metadata and an OpenAI-compatible web-enabled model. The main notebook workflow is designed to run out of the box for demonstration use.
 
 ## Relation to OpenGMS
 
 OpenGMS provides the model-service platform and online execution infrastructure. PyGeoModel is a Python client package that exposes OpenGMS model-service discovery, metadata inspection, task invocation, and result records to Python and notebook workflows.
-
-## Development
-
-Run the lightweight test suite with:
-
-```bash
-python -m unittest discover -s tests
-```
-
-The source distribution should not include real API keys, local `.env` files, bytecode caches, generated C files, `.pyd` binaries, or previous build artifacts.

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/README.md

@@ -45,20 +45,6 @@ result = modeler.invoke(
 result.to_json("execution_record.json")
 ```
 
-OpenGMS execution uses the bundled public demo token by default. For stable or private use, configure your own OpenGMS token:
-
-```bash
-export OGMS_TOKEN="your-token"
-```
-
-Optional endpoint overrides:
-
-```bash
-export OGMS_BASE_PORTAL_URL="http://222.192.7.75"
-export OGMS_BASE_MANAGER_URL="http://222.192.7.75/managerServer"
-export OGMS_BASE_DATA_URL="http://222.192.7.75/dataTransferServer"
-```
-
 ## Notebook Interface
 
 ```python
@@ -78,18 +64,8 @@ answer = modeler.ask_model("SWAT_Model", "What input data are required?")
 answer.to_json("qa_record.json")
 ```
 
-The recommendation service automatically builds notebook/data context and calls the configured Dify workflow. `DIFY_API_KEY` is required for `suggest_model()` because the context-aware recommendation logic is implemented in that workflow. Q&A uses OpenGMS model metadata and, when configured, an OpenAI-compatible web-enabled model.
+The recommendation service automatically builds notebook/data context and calls the configured recommendation workflow. Q&A uses OpenGMS model metadata and an OpenAI-compatible web-enabled model. The main notebook workflow is designed to run out of the box for demonstration use.
 
 ## Relation to OpenGMS
 
 OpenGMS provides the model-service platform and online execution infrastructure. PyGeoModel is a Python client package that exposes OpenGMS model-service discovery, metadata inspection, task invocation, and result records to Python and notebook workflows.
-
-## Development
-
-Run the lightweight test suite with:
-
-```bash
-python -m unittest discover -s tests
-```
-
-The source distribution should not include real API keys, local `.env` files, bytecode caches, generated C files, `.pyd` binaries, or previous build artifacts.

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/pygeomodel/__init__.py

@@ -5,7 +5,7 @@ from .modeler import GeoModeler
 from .models import ModelInput, ModelOutput, ModelService, ModelSummary
 from .results import QAResult, RecommendationResult, TaskResult
 
-__version__ = "1.0.6"
+__version__ = "1.0.8"
 
 __all__ = [
     "GeoModeler",

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/pygeomodel/config.py

@@ -9,6 +9,11 @@ from pathlib import Path
 
 
 PUBLIC_DEMO_OGMS_TOKEN = '6U3O1Sy5696I5ryJFaYCYVjcIV7rhd1MKK0QGX9A7zafogi8xTdvejl6ISUP1lEs'
+PUBLIC_DEMO_DIFY_API_KEY = 'app-CuNONc6hSct2ap07nmUgcaw9'
+PUBLIC_DEMO_DIFY_BASE_URL = 'https://api.dify.ai/v1'
+PUBLIC_DEMO_OPENAI_API_KEY = 'sk-4bp5a1DcdLSHCiw1401270055f47424b9eA58cAd587266A3'
+PUBLIC_DEMO_OPENAI_BASE_URL = 'https://aihubmix.com/v1'
+PUBLIC_DEMO_OPENAI_MODEL = 'gpt-5.2-low'
 
 
 @dataclass(frozen=True)
@@ -47,11 +52,11 @@ def get_llm_config() -> LLMConfig:
     dify = local.get("dify", {})
     consensus = local.get("consensus", {})
     return LLMConfig(
-        openai_api_key=openai.get("api_key") or os.environ.get("OPENAI_API_KEY"),
-        openai_base_url=openai.get("base_url") or os.environ.get("OPENAI_BASE_URL") or LLMConfig.openai_base_url,
-        openai_model=openai.get("model") or os.environ.get("OPENAI_MODEL") or LLMConfig.openai_model,
-        dify_api_key=dify.get("api_key") or os.environ.get("DIFY_API_KEY"),
-        dify_base_url=dify.get("base_url") or os.environ.get("DIFY_BASE_URL") or LLMConfig.dify_base_url,
+        openai_api_key=os.environ.get("OPENAI_API_KEY") or openai.get("api_key") or PUBLIC_DEMO_OPENAI_API_KEY,
+        openai_base_url=os.environ.get("OPENAI_BASE_URL") or openai.get("base_url") or PUBLIC_DEMO_OPENAI_BASE_URL,
+        openai_model=os.environ.get("OPENAI_MODEL") or openai.get("model") or PUBLIC_DEMO_OPENAI_MODEL,
+        dify_api_key=os.environ.get("DIFY_API_KEY") or dify.get("api_key") or PUBLIC_DEMO_DIFY_API_KEY,
+        dify_base_url=os.environ.get("DIFY_BASE_URL") or dify.get("base_url") or PUBLIC_DEMO_DIFY_BASE_URL,
         consensus_api_key=consensus.get("api_key") or os.environ.get("CONSENSUS_API_KEY"),
         consensus_base_url=consensus.get("base_url") or os.environ.get("CONSENSUS_BASE_URL") or LLMConfig.consensus_base_url,
     )

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/pygeomodel/results.py

@@ -29,7 +29,7 @@ class TaskResult:
     params: dict[str, Any] = field(default_factory=dict)
     uploaded_inputs: dict[str, Any] = field(default_factory=dict)
     endpoint: str | None = None
-    pygeomodel_version: str = "1.0.6"
+    pygeomodel_version: str = "1.0.8"
     execution_time: float | None = None
     created_at: float = field(default_factory=time.time)
 

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/setup.py

@@ -11,7 +11,7 @@ def read_readme():
 
 setup(
     name="PyGeoModel",
-    version="1.0.6",
+    version="1.0.8",
     author="Peilong Ma",
     author_email="mpl_gis@nnu.edu.cn",
     description="A Python package for integrating OpenGMS geographic model services.",
@@ -41,7 +41,6 @@ setup(
     },
     python_requires=">=3.8",
     classifiers=[
-        "Development Status :: 4 - Beta",
         "Intended Audience :: Science/Research",
         "Topic :: Scientific/Engineering :: GIS",
         "Topic :: Scientific/Engineering :: Artificial Intelligence",

{pygeomodel-1.0.6 → pygeomodel-1.0.8}/tests/test_core_api.py

@@ -323,6 +323,52 @@ class CoreApiTests(unittest.TestCase):
             else:
                 os.environ["OGMS_TOKEN"] = original_env
 
+    def test_llm_config_uses_public_demo_credentials_as_fallback(self):
+        import pygeomodel.config as config_module
+
+        env_keys = ["DIFY_API_KEY", "DIFY_BASE_URL", "OPENAI_API_KEY", "OPENAI_BASE_URL", "OPENAI_MODEL"]
+        original_env = {key: os.environ.get(key) for key in env_keys}
+        original_loader = config_module._load_local_api_keys
+        try:
+            config_module._load_local_api_keys = lambda: {}
+            for key in env_keys:
+                os.environ.pop(key, None)
+
+            cfg = config_module.get_llm_config()
+            self.assertEqual(cfg.dify_api_key, config_module.PUBLIC_DEMO_DIFY_API_KEY)
+            self.assertEqual(cfg.dify_base_url, config_module.PUBLIC_DEMO_DIFY_BASE_URL)
+            self.assertEqual(cfg.openai_api_key, config_module.PUBLIC_DEMO_OPENAI_API_KEY)
+            self.assertEqual(cfg.openai_base_url, config_module.PUBLIC_DEMO_OPENAI_BASE_URL)
+            self.assertEqual(cfg.openai_model, config_module.PUBLIC_DEMO_OPENAI_MODEL)
+
+            config_module._load_local_api_keys = lambda: {
+                "dify": {"api_key": "local-dify", "base_url": "https://local-dify.example/v1"},
+                "openai": {
+                    "api_key": "local-openai",
+                    "base_url": "https://local-openai.example/v1",
+                    "model": "local-model",
+                },
+            }
+            cfg = config_module.get_llm_config()
+            self.assertEqual(cfg.dify_api_key, "local-dify")
+            self.assertEqual(cfg.openai_api_key, "local-openai")
+            self.assertEqual(cfg.openai_model, "local-model")
+
+            os.environ["DIFY_API_KEY"] = "env-dify"
+            os.environ["OPENAI_API_KEY"] = "env-openai"
+            os.environ["OPENAI_MODEL"] = "env-model"
+            cfg = config_module.get_llm_config()
+            self.assertEqual(cfg.dify_api_key, "env-dify")
+            self.assertEqual(cfg.openai_api_key, "env-openai")
+            self.assertEqual(cfg.openai_model, "env-model")
+        finally:
+            config_module._load_local_api_keys = original_loader
+            for key, value in original_env.items():
+                if value is None:
+                    os.environ.pop(key, None)
+                else:
+                    os.environ[key] = value
+
     def test_recommendation_result_has_notebook_rich_display(self):
         from pygeomodel import RecommendationResult