Merge pull request #28 from freelawproject/increase_query_max_len

fix(setting): increase max query length

Author: mlissner

.env.example

@@ -5,7 +5,7 @@ MAX_TOKENS=512
 OVERLAP_RATIO=0.004
 MIN_TEXT_LENGTH=1
 MAX_TEXT_LENGTH = 10000000
-MAX_QUERY_LENGTH = 100
+MAX_QUERY_LENGTH = 1000
 MAX_BATCH_SIZE=100
 PROCESSING_BATCH_SIZE = 8
 POOL_TIMEOUT=3600

README.md

@@ -63,7 +63,7 @@ Model Settings:
 
 - `MAX_QUERY_LENGTH`
 
-    Default: `100`
+    Default: `1000`
 
     The maximum allowable length (in characters) for a query text.
 

inception/config.py

@@ -21,7 +21,7 @@ class Settings(BaseSettings):
         description="Ratio to calculate number of sentence overlap between chunks",
     )
     min_text_length: int = 1
-    max_query_length: int = 100
+    max_query_length: int = 1000
     max_text_length: int = 10_000_000
     max_batch_size: int = 100
     processing_batch_size: int = 8

tests/test_embedding_service.py

@@ -133,7 +133,10 @@ class TestInputValidation:
     def test_query_embedding_validation(self, client):
         """Test query endpoint input validation."""
 
-        long_query = "Pellentesque tellus felis cursus id velit ac feugiat rutrum massa Mauris dapibus fermentum sagittis Donec viverra mauris a velit"
+        long_query = (
+            "Pellentesque tellus felis cursus id velit ac feugiat rutrum massa Mauris dapibus fermentum sagittis Donec viverra mauris a velit ac quam consectetur, a facilisis enim eleifend."
+            * 10
+        )
         test_cases = [
             {
                 "name": "short text",