From c27cc505e150d5223e23ea291fd3120a94f7a826 Mon Sep 17 00:00:00 2001
From: "Gan, Jimmy" <jimmy.gan@adidas.com>
Date: Sun, 12 Apr 2026 11:33:24 +0800
Subject: [PATCH] fix: add batch processing for large conversation files

---
 claude-code-tracker/collector.py | 14 +++++++++-----
 claude-code-tracker/parser.py    | 10 ++++++----
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/claude-code-tracker/collector.py b/claude-code-tracker/collector.py
index b0705d4..d467395 100644
--- a/claude-code-tracker/collector.py
+++ b/claude-code-tracker/collector.py
@@ -52,9 +52,13 @@ async def process_file(file_path: str):
             # No new lines
             return 0
 
+        # Process in batches for large files (max 200 lines at a time)
+        batch_size = 200
+        end_line = min(start_line + batch_size, total_lines)
+
         # Parse new content
-        logger.info(f"Processing {file_path} from line {start_line} to {total_lines}")
-        conversation_data = parser.parse_conversation_file(file_path, start_line)
+        logger.info(f"Processing {file_path} from line {start_line} to {end_line} (total: {total_lines})")
+        conversation_data = parser.parse_conversation_file(file_path, start_line, end_line)
 
         if not conversation_data:
             logger.warning(f"No data extracted from {file_path}")
@@ -87,10 +91,10 @@ async def process_file(file_path: str):
         for tool_call in conversation_data['tool_calls']:
             await db.insert_tool_call(tool_call)
 
-        # Update checkpoint
-        await db.update_checkpoint(file_path, total_lines, file_mtime)
+        # Update checkpoint (use end_line instead of total_lines for batch processing)
+        await db.update_checkpoint(file_path, end_line, file_mtime)
 
-        logger.info(f"Processed {len(conversation_data['messages'])} messages from {file_path}")
+        logger.info(f"Processed {len(conversation_data['messages'])} messages from {file_path} (batch {start_line}-{end_line}/{total_lines})")
         return len(conversation_data['messages'])
 
     except Exception as e:
diff --git a/claude-code-tracker/parser.py b/claude-code-tracker/parser.py
index 16d8efd..57fa267 100644
--- a/claude-code-tracker/parser.py
+++ b/claude-code-tracker/parser.py
@@ -6,14 +6,16 @@ from typing import Dict, List, Any, Optional
 logger = logging.getLogger(__name__)
 
 
-def parse_jsonl_file(file_path: str, start_line: int = 0) -> List[Dict[str, Any]]:
-    """Parse JSONL file from a specific line number"""
+def parse_jsonl_file(file_path: str, start_line: int = 0, end_line: int = None) -> List[Dict[str, Any]]:
+    """Parse JSONL file from a specific line number to end_line (or EOF if None)"""
     messages = []
     try:
         with open(file_path, 'r', encoding='utf-8') as f:
             for i, line in enumerate(f):
                 if i < start_line:
                     continue
+                if end_line is not None and i >= end_line:
+                    break
                 if not line.strip():
                     continue
                 try:
@@ -169,9 +171,9 @@ def calculate_conversation_stats(messages: List[Dict[str, Any]]) -> Dict[str, An
     return stats
 
 
-def parse_conversation_file(file_path: str, start_line: int = 0) -> Dict[str, Any]:
+def parse_conversation_file(file_path: str, start_line: int = 0, end_line: int = None) -> Dict[str, Any]:
     """Parse a conversation file and extract all data"""
-    raw_messages = parse_jsonl_file(file_path, start_line)
+    raw_messages = parse_jsonl_file(file_path, start_line, end_line)
 
     if not raw_messages:
         return None