updated tests and config nox

lang_main_config.toml

@@ -1,9 +1,12 @@
 # d-opt -- lang_main: config file
 
 [paths]
-inputs = '../lang-data/in/'
-results = '../lang-data/out/'
-models = '../lang-models'
+inputs = './lang-data/in/'
+results = './lang-data/out/'
+models = './lang-models'
+
+[models]
+use_large_model = true
 
 [logging]
 enabled = true

lang_main_config_old.toml

@@ -0,0 +1,56 @@
+# d-opt -- lang_main: config file
+
+[paths]
+inputs = '../lang-data/in/'
+results = '../lang-data/out/'
+models = '../lang-models'
+
+[logging]
+enabled = true
+stderr = true
+file = true
+
+# control which pipelines are executed
+[control]
+preprocessing_skip = false
+token_analysis_skip = false
+graph_postprocessing_skip = false
+graph_rescaling_skip = false
+graph_static_rendering_skip = true
+time_analysis_skip = true
+
+[preprocess]
+date_cols = [
+    "VorgangsDatum",
+    "ErledigungsDatum",
+    "Arbeitsbeginn",
+    "ErstellungsDatum",
+]
+target_feature = "VorgangsBeschreibung"
+threshold_amount_characters = 5
+threshold_similarity = 0.92
+
+[graph_postprocessing]
+max_edge_number = -1
+
+[time_analysis.uniqueness]
+threshold_unique_texts = 5
+criterion_feature = 'HObjektText'
+feature_name_obj_id = 'ObjektID'
+feature_name_obj_text = 'HObjektText'
+
+[time_analysis.preparation]
+name_delta_feat_to_repair = 'Zeitspanne bis zur Behebung [Tage]'
+name_delta_feat_to_next_failure = 'Zeitspanne bis zum nächsten Ereignis [Tage]'
+
+[time_analysis.model_input]
+input_features = [
+    'VorgangsBeschreibung',
+]
+activity_feature = 'VorgangsTypName'
+activity_types = [
+    'Reparaturauftrag (Portal)',
+    'Störungsmeldung',
+]
+threshold_num_activities = 1
+threshold_similarity = 0.8

noxfile.py

@@ -0,0 +1,41 @@
+import os
+import shlex
+from typing import Final
+
+import nox
+from nox import Session
+
+os.environ['PDM_IGNORE_SAVED_PYTHON'] = '1'
+PYTHON_VERS: Final[list[str]] = ['3.11']
+
+
+@nox.session(name='min', python=PYTHON_VERS, reuse_venv=True)
+def tests_wo_models_cyto(session: Session) -> None:
+    """Run all tests despite model loading and Cytoscape tests"""
+    cmd_raw = 'pdm install -G cytoscape,spacy-sm,spacy-md,spacy-trf -dG tests'
+    cmd = shlex.split(cmd_raw)
+    session.run_install(*cmd, external=True)
+    cmd_tests_raw = 'pytest -m "not mload and not cyto"'
+    cmd_tests = shlex.split(cmd_tests_raw)
+    session.run(*cmd_tests)
+
+
+@nox.session(name='all', python=PYTHON_VERS, reuse_venv=True, default=False)
+def tests_all(session: Session) -> None:
+    """Run all tests, including Cytoscape Docker container"""
+    cmd_raw = 'pdm install -G cytoscape,spacy-sm,spacy-md,spacy-trf -dG tests'
+    cmd = shlex.split(cmd_raw)
+    session.run_install(*cmd, external=True)
+
+    cmd_tests_raw = 'pytest -n 4'
+    cmd_tests = shlex.split(cmd_tests_raw)
+    session.run(*cmd_tests)
+
+    session.run('docker', 'desktop', 'start', external=True)
+    session.run('docker', 'start', 'cyrest', external=True)
+
+    cmd_tests_raw = 'pytest -m "cyto"'
+    cmd_tests = shlex.split(cmd_tests_raw)
+    session.run(*cmd_tests)
+
+    session.run('docker', 'stop', 'cyrest', external=True)

src/lang_main/config.py

@@ -36,7 +36,7 @@ BASE_FOLDERNAME: Final[str] = os.environ.get('LANG_MAIN_BASE_FOLDERNAME', 'lang-
 CONFIG_FILENAME: Final[str] = 'lang_main_config.toml'
 CYTO_STYLESHEET_FILENAME: Final[str] = r'cytoscape_config/lang_main.xml'
 PKG_DIR: Final[Path] = Path(__file__).parent
-STOP_FOLDER: Final[str] = os.environ.get('LANG_MAIN_STOP_SEARCH_FOLDERNAME', 'src')
+STOP_FOLDER: Final[str] = os.environ.get('LANG_MAIN_STOP_SEARCH_FOLDERNAME', 'lang-main')
 
 
 def load_toml_config(

src/lang_main/model_loader.py

@@ -115,11 +115,11 @@ def _preprocess_STFR_model_name(
         # !! without any user names
         folder_name = model_name.split('/')[-1]
         model_path = MODEL_BASE_FOLDER / folder_name
-        if not model_path.exists():
+        if not model_path.exists():  # pragma: no cover
             raise FileNotFoundError(
                 f'Target model >{model_name}< not found under {model_path}'
             )
-        model_name_or_path = str(model_path)  # pragma: no cover
+        model_name_or_path = str(model_path)
     else:
         model_name_or_path = model_name
 

tests/test_config.py

@@ -52,7 +52,7 @@ def test_load_cfg_func(monkeypatch, tmp_path):
         stop_folder_name=stop_folder,
         lookup_cwd=False,
     )
-    assert loaded_cfg['paths']['models'] == '../lang-models'
+    assert loaded_cfg['paths']['models'] == './lang-models'
 
     loaded_cfg = config.load_cfg(
         starting_path=pkg_dir,
@@ -60,4 +60,4 @@ def test_load_cfg_func(monkeypatch, tmp_path):
         stop_folder_name=stop_folder,
         lookup_cwd=True,
     )
-    assert loaded_cfg['paths']['models'] == '../lang-models'
+    assert loaded_cfg['paths']['models'] == './lang-models'

tests/test_model_loader.py

@@ -26,7 +26,7 @@ from lang_main.types import LanguageModels
     'model_name',
     [
         STFRModelTypes.ALL_MINI_LM_L6_V2,
-        STFRModelTypes.ALL_MPNET_BASE_V2,
+        STFRModelTypes.ALL_MINI_LM_L12_V2,
     ],
 )
 @pytest.mark.mload
@@ -58,10 +58,9 @@ def test_preprocess_STFR_model_name() -> None:
 
     model_name_exist = STFRModelTypes.E5_BASE_STS_EN_DE
     backend_exist = STFRBackends.ONNX
-    with pytest.raises(FileNotFoundError):
-        _ = model_loader._preprocess_STFR_model_name(
-            model_name=model_name_exist, backend=backend_exist, force_download=False
-        )
+    _ = model_loader._preprocess_STFR_model_name(
+        model_name=model_name_exist, backend=backend_exist, force_download=False
+    )
 
 
 @pytest.mark.parametrize(
@@ -75,7 +74,7 @@ def test_preprocess_STFR_model_name() -> None:
     'model_name',
     [
         STFRModelTypes.ALL_MINI_LM_L6_V2,
-        STFRModelTypes.ALL_MPNET_BASE_V2,
+        STFRModelTypes.ALL_MINI_LM_L12_V2,
     ],
 )
 @pytest.mark.mload
@@ -94,9 +93,7 @@ def test_load_sentence_transformer_onnx(model_name, similarity_func) -> None:
 @pytest.mark.parametrize(
     'model_name',
     [
-        SpacyModelTypes.DE_CORE_NEWS_SM,
         SpacyModelTypes.DE_CORE_NEWS_MD,
-        SpacyModelTypes.DE_CORE_NEWS_LG,
         SpacyModelTypes.DE_DEP_NEWS_TRF,
     ],
 )
@@ -111,7 +108,7 @@ def test_load_spacy_model(model_name):
 def test_load_spacy_model_fail():
     model_name = 'not_existing'
     with pytest.raises(LanguageModelNotFoundError):
-        model = model_loader.load_spacy(model_name)
+        _ = model_loader.load_spacy(model_name)
 
 
 @pytest.mark.mload