Tests for on-demand resource loading.

tests/test_mlp.py

 import json
+
 import pytest
 import regex as re
 
+from texta_mlp.entity_mapper import EntityMapper
 from texta_mlp.mlp import MLP
 
 
@@ -236,3 +238,22 @@ def test_parsing_empty_list_in_dictionary(mlp: MLP):
     result = mlp.process_docs([{"empty_list_field": []}], doc_paths=["empty_list_field"])
     for key in result:
         assert "mlp" not in key
+
+
+def test_that_models_are_loaded_on_demand():
+    pipeline = MLP(language_codes=["et", "en"], logging_level="info", use_gpu=False)
+    stanza_pipelines = pipeline._stanza_pipelines
+    assert len(stanza_pipelines.keys()) == 0
+    result = pipeline.process(raw_text="Tere, minu nimi on Joonas, kas saaksite öelda, mis kell praegu on?", analyzers=["lemmas"], lang="et")
+    assert "et" in pipeline._stanza_pipelines
+    assert len(pipeline._stanza_pipelines.keys()) == 1
+    result = pipeline.process(raw_text="Hello there, my name is Joonas, how do you do!?", analyzers=["lemmas"])
+    assert "en" in pipeline._stanza_pipelines
+    assert len(pipeline._stanza_pipelines) == 2
+
+
+def test_that_entity_mapper_is_loaded_on_demand():
+    pipeline = MLP(language_codes=["et"], logging_level="info", use_gpu=False)
+    assert pipeline._entity_mapper is None
+    pipeline.process(raw_text="Tere, minu nimi on Joonas, kas saaksite öelda, mis kell praegu on?", analyzers=["entities"], lang="et")
+    assert isinstance(pipeline._entity_mapper, EntityMapper)

texta_mlp/mlp.py

@@ -81,7 +81,7 @@ class MLP:
         self.use_default_lang = use_default_language_code
         self.resource_dir = resource_dir
 
-        self.__stanza_pipelines = {}
+        self._stanza_pipelines = {}
         self.logging_level = logging_level
         self.use_gpu = use_gpu
         self.stanza_resource_path = pathlib.Path(self.resource_dir) / "stanza"
@@ -93,7 +93,7 @@ class MLP:
 
         self.prepare_resources(refresh_data)
 
-        self.__entity_mapper = None
+        self._entity_mapper = None
         self.loaded_entity_files = []
 
         self.not_entities = self._load_not_entities()
@@ -235,15 +235,14 @@ class MLP:
 
 
     def get_entity_mapper(self):
-        if self.__entity_mapper is None:
-            self.__entity_mapper = self._load_entity_mapper()
-        return self.__entity_mapper
-
+        if self._entity_mapper is None:
+            self._entity_mapper = self._load_entity_mapper()
+        return self._entity_mapper
 
     def get_stanza_pipeline(self, lang: str):
-        if lang not in self.__stanza_pipelines:
+        if lang not in self._stanza_pipelines:
             try:
-                self.__stanza_pipelines[lang] = stanza.Pipeline(
+                self._stanza_pipelines[lang] = stanza.Pipeline(
                     lang=lang,
                     dir=str(self.stanza_resource_path),
                     processors=self._get_stanza_processors(lang),
@@ -253,7 +252,7 @@ class MLP:
 
             # This is for CUDA OOM exceptions. Fall back to CPU if needed.
             except RuntimeError:
-                self.__stanza_pipelines[lang] = stanza.Pipeline(
+                self._stanza_pipelines[lang] = stanza.Pipeline(
                     lang=lang,
                     dir=str(self.stanza_resource_path),
                     processors=self._get_stanza_processors(lang),
@@ -261,7 +260,7 @@ class MLP:
                     logging_level=self.logging_level,
                 )
 
-        return self.__stanza_pipelines[lang]
+        return self._stanza_pipelines[lang]
 
 
     def _get_stanza_tokens(self, lang: str, raw_text: str):