From ebd4403c2f74bf6f65f717ad389b89e869a7efc2 Mon Sep 17 00:00:00 2001
From: hajdul88 <52442977+hajdul88@users.noreply.github.com>
Date: Wed, 13 Aug 2025 12:14:42 +0200
Subject: [PATCH] adds new error classes to keywordextration error and defines
 new error

---
 .../data/exceptions/__init__.py               |  9 ++++++++
 .../data/exceptions/exceptions.py             | 22 +++++++++++++++++++
 .../data/utils/extract_keywords.py            |  4 ++--
 3 files changed, 33 insertions(+), 2 deletions(-)
 create mode 100644 cognee/infrastructure/data/exceptions/__init__.py
 create mode 100644 cognee/infrastructure/data/exceptions/exceptions.py

diff --git a/cognee/infrastructure/data/exceptions/__init__.py b/cognee/infrastructure/data/exceptions/__init__.py
new file mode 100644
index 000000000..6735200ed
--- /dev/null
+++ b/cognee/infrastructure/data/exceptions/__init__.py
@@ -0,0 +1,9 @@
+"""
+Custom exceptions for the Cognee API.
+
+This module defines a set of exceptions for handling various data errors
+"""
+
+from .exceptions import (
+    KeywordExtractionError
+)
diff --git a/cognee/infrastructure/data/exceptions/exceptions.py b/cognee/infrastructure/data/exceptions/exceptions.py
new file mode 100644
index 000000000..5c36d6754
--- /dev/null
+++ b/cognee/infrastructure/data/exceptions/exceptions.py
@@ -0,0 +1,22 @@
+from cognee.exceptions import (
+    CogneeValidationError,
+)
+from fastapi import status
+
+
+class KeywordExtractionError(CogneeValidationError):
+    """
+    Raised when a provided value is syntactically valid but semantically unacceptable
+    for the given operation.
+
+    Example:
+        - Passing an empty string to a keyword extraction function.
+    """
+
+    def __init__(
+        self,
+        message: str = "Extract_keywords cannot extract keywords from empty text.",
+        name: str = "KeywordExtractionError",
+        status_code: int = status.HTTP_400_BAD_REQUEST,
+    ):
+        super().__init__(message, name, status_code)
diff --git a/cognee/infrastructure/data/utils/extract_keywords.py b/cognee/infrastructure/data/utils/extract_keywords.py
index bd4fedd56..2915131a4 100644
--- a/cognee/infrastructure/data/utils/extract_keywords.py
+++ b/cognee/infrastructure/data/utils/extract_keywords.py
@@ -1,6 +1,6 @@
 from sklearn.feature_extraction.text import TfidfVectorizer
 
-from cognee.exceptions import InvalidValueError
+from cognee.infrastructure.data.exceptions.exceptions import KeywordExtractionError
 from cognee.shared.utils import extract_pos_tags
 
 
@@ -25,7 +25,7 @@ def extract_keywords(text: str) -> list[str]:
           with more than 3 characters.
     """
     if len(text) == 0:
-        raise InvalidValueError(message="extract_keywords cannot extract keywords from empty text.")
+        raise KeywordExtractionError()
 
     tags = extract_pos_tags(text)
     nouns = [word for (word, tag) in tags if tag == "NN"]