From 965f9e98bf73ab3c30c67a84061f7f84f3837085 Mon Sep 17 00:00:00 2001
From: Chris Weaver <25087905+Weves@users.noreply.github.com>
Date: Mon, 10 Mar 2025 08:50:50 -0700
Subject: [PATCH] Eliminate extremely long log line for large checkpointds
 (#4236)

* Eliminate extremely long log line for large checkpointds

* address greptile
---
 backend/onyx/connectors/models.py | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/backend/onyx/connectors/models.py b/backend/onyx/connectors/models.py
index 6211970f7..6c5ba7691 100644
--- a/backend/onyx/connectors/models.py
+++ b/backend/onyx/connectors/models.py
@@ -1,3 +1,4 @@
+import json
 from datetime import datetime
 from enum import Enum
 from typing import Any
@@ -204,6 +205,15 @@ class ConnectorCheckpoint(BaseModel):
     def build_dummy_checkpoint(cls) -> "ConnectorCheckpoint":
         return ConnectorCheckpoint(checkpoint_content={}, has_more=True)
 
+    def __str__(self) -> str:
+        """String representation of the checkpoint, with truncation for large checkpoint content."""
+        MAX_CHECKPOINT_CONTENT_CHARS = 1000
+
+        content_str = json.dumps(self.checkpoint_content)
+        if len(content_str) > MAX_CHECKPOINT_CONTENT_CHARS:
+            content_str = content_str[: MAX_CHECKPOINT_CONTENT_CHARS - 3] + "..."
+        return f"ConnectorCheckpoint(checkpoint_content={content_str}, has_more={self.has_more})"
+
 
 class DocumentFailure(BaseModel):
     document_id: str