Fix batch processing for large conversation files #48
@@ -52,9 +52,13 @@ async def process_file(file_path: str):
|
|||||||
# No new lines
|
# No new lines
|
||||||
return 0
|
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
|
# Parse new content
|
||||||
logger.info(f"Processing {file_path} from line {start_line} to {total_lines}")
|
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)
|
conversation_data = parser.parse_conversation_file(file_path, start_line, end_line)
|
||||||
|
|
||||||
if not conversation_data:
|
if not conversation_data:
|
||||||
logger.warning(f"No data extracted from {file_path}")
|
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']:
|
for tool_call in conversation_data['tool_calls']:
|
||||||
await db.insert_tool_call(tool_call)
|
await db.insert_tool_call(tool_call)
|
||||||
|
|
||||||
# Update checkpoint
|
# Update checkpoint (use end_line instead of total_lines for batch processing)
|
||||||
await db.update_checkpoint(file_path, total_lines, file_mtime)
|
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'])
|
return len(conversation_data['messages'])
|
||||||
|
|
||||||
except Exception as e:
|
except Exception as e:
|
||||||
|
|||||||
@@ -6,14 +6,16 @@ from typing import Dict, List, Any, Optional
|
|||||||
logger = logging.getLogger(__name__)
|
logger = logging.getLogger(__name__)
|
||||||
|
|
||||||
|
|
||||||
def parse_jsonl_file(file_path: str, start_line: int = 0) -> List[Dict[str, Any]]:
|
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"""
|
"""Parse JSONL file from a specific line number to end_line (or EOF if None)"""
|
||||||
messages = []
|
messages = []
|
||||||
try:
|
try:
|
||||||
with open(file_path, 'r', encoding='utf-8') as f:
|
with open(file_path, 'r', encoding='utf-8') as f:
|
||||||
for i, line in enumerate(f):
|
for i, line in enumerate(f):
|
||||||
if i < start_line:
|
if i < start_line:
|
||||||
continue
|
continue
|
||||||
|
if end_line is not None and i >= end_line:
|
||||||
|
break
|
||||||
if not line.strip():
|
if not line.strip():
|
||||||
continue
|
continue
|
||||||
try:
|
try:
|
||||||
@@ -169,9 +171,9 @@ def calculate_conversation_stats(messages: List[Dict[str, Any]]) -> Dict[str, An
|
|||||||
return stats
|
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"""
|
"""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:
|
if not raw_messages:
|
||||||
return None
|
return None
|
||||||
|
|||||||
Reference in New Issue
Block a user