Do not store word filters index to file. Not necessary for now

- It's more of a hassle to not let word filter go stale on entry
  updates
- Generating index on 120K lines of notes takes 1s. Loading from file
  takes 0.2s. For less content load time difference will be even smaller
- Let go of startup time improvement for simplicity for now

tests/conftest.py

@@ -58,7 +58,7 @@ def model_dir(search_config: SearchConfig):
         compressed_jsonl = model_dir.joinpath('notes.jsonl.gz'),
         embeddings_file = model_dir.joinpath('note_embeddings.pt'))
 
-    filters = [DateFilter(), WordFilter(model_dir, search_type=SearchType.Org), FileFilter()]
+    filters = [DateFilter(), WordFilter(), FileFilter()]
     text_search.setup(org_to_jsonl, content_config.org, search_config.asymmetric, regenerate=False, filters=filters)
 
     return model_dir

tests/test_client.py

@@ -132,7 +132,7 @@ def test_notes_search(content_config: ContentConfig, search_config: SearchConfig
 # ----------------------------------------------------------------------------------------------------
 def test_notes_search_with_include_filter(content_config: ContentConfig, search_config: SearchConfig):
     # Arrange
-    filters = [WordFilter(content_config.org.compressed_jsonl.parent, search_type=SearchType.Org)]
+    filters = [WordFilter()]
     model.orgmode_search = text_search.setup(org_to_jsonl, content_config.org, search_config.asymmetric, regenerate=False, filters=filters)
     user_query = 'How to git install application? +"Emacs"'
 
@@ -149,7 +149,7 @@ def test_notes_search_with_include_filter(content_config: ContentConfig, search_
 # ----------------------------------------------------------------------------------------------------
 def test_notes_search_with_exclude_filter(content_config: ContentConfig, search_config: SearchConfig):
     # Arrange
-    filters = [WordFilter(content_config.org.compressed_jsonl.parent, search_type=SearchType.Org)]
+    filters = [WordFilter()]
     model.orgmode_search = text_search.setup(org_to_jsonl, content_config.org, search_config.asymmetric, regenerate=False, filters=filters)
     user_query = 'How to git install application? -"clone"'
 

tests/test_word_filter.py

@@ -1,15 +1,12 @@
-# External Packages
-import torch
-
 # Application Packages
 from src.search_filter.word_filter import WordFilter
 from src.utils.config import SearchType
 
 
-def test_no_word_filter(tmp_path):
+def test_no_word_filter():
     # Arrange
-    word_filter = WordFilter(tmp_path, SearchType.Org)
-    embeddings, entries = arrange_content()
+    word_filter = WordFilter()
+    entries = arrange_content()
     q_with_no_filter = 'head tail'
 
     # Act
@@ -22,10 +19,10 @@ def test_no_word_filter(tmp_path):
     assert entry_indices == {0, 1, 2, 3}
 
 
-def test_word_exclude_filter(tmp_path):
+def test_word_exclude_filter():
     # Arrange
-    word_filter = WordFilter(tmp_path, SearchType.Org)
-    embeddings, entries = arrange_content()
+    word_filter = WordFilter()
+    entries = arrange_content()
     q_with_exclude_filter = 'head -"exclude_word" tail'
 
     # Act
@@ -38,10 +35,10 @@ def test_word_exclude_filter(tmp_path):
     assert entry_indices == {0, 2}
 
 
-def test_word_include_filter(tmp_path):
+def test_word_include_filter():
     # Arrange
-    word_filter = WordFilter(tmp_path, SearchType.Org)
-    embeddings, entries = arrange_content()
+    word_filter = WordFilter()
+    entries = arrange_content()
     query_with_include_filter = 'head +"include_word" tail'
 
     # Act
@@ -54,10 +51,10 @@ def test_word_include_filter(tmp_path):
     assert entry_indices == {2, 3}
 
 
-def test_word_include_and_exclude_filter(tmp_path):
+def test_word_include_and_exclude_filter():
     # Arrange
-    word_filter = WordFilter(tmp_path, SearchType.Org)
-    embeddings, entries = arrange_content()
+    word_filter = WordFilter()
+    entries = arrange_content()
     query_with_include_and_exclude_filter = 'head +"include_word" -"exclude_word" tail'
 
     # Act
@@ -71,11 +68,10 @@ def test_word_include_and_exclude_filter(tmp_path):
 
 
 def arrange_content():
-    embeddings = torch.randn(4, 10)
     entries = [
         {'compiled': '', 'raw': 'Minimal Entry'},
         {'compiled': '', 'raw': 'Entry with exclude_word'},
         {'compiled': '', 'raw': 'Entry with include_word'},
         {'compiled': '', 'raw': 'Entry with include_word and exclude_word'}]
 
-    return embeddings, entries
+    return entries