diff --git a/backend/services/billing.py b/backend/services/billing.py index db1c7dba..7bae3c9f 100644 --- a/backend/services/billing.py +++ b/backend/services/billing.py @@ -225,74 +225,220 @@ async def get_user_subscription(user_id: str) -> Optional[Dict]: async def calculate_monthly_usage(client, user_id: str) -> float: """Calculate total agent run minutes for the current month for a user.""" + start_time = time.time() + + # Use get_usage_logs to fetch all usage data (it already handles the date filtering and batching) + total_cost = 0.0 + page = 0 + items_per_page = 1000 + + while True: + # Get usage logs for this page + usage_result = await get_usage_logs(client, user_id, page, items_per_page) + + if not usage_result['logs']: + break + + # Sum up the estimated costs from this page + for log_entry in usage_result['logs']: + total_cost += log_entry['estimated_cost'] + + # If there are no more pages, break + if not usage_result['has_more']: + break + + page += 1 + + end_time = time.time() + execution_time = end_time - start_time + logger.info(f"Calculate monthly usage took {execution_time:.3f} seconds, total cost: {total_cost}") + + return total_cost + + +async def get_usage_logs(client, user_id: str, page: int = 0, items_per_page: int = 1000) -> Dict: + """Get detailed usage logs for a user with pagination.""" # Get start of current month in UTC now = datetime.now(timezone.utc) start_of_month = datetime(now.year, now.month, 1, tzinfo=timezone.utc) - # Use fixed cutoff date: June 27, 2025 midnight UTC + # Use fixed cutoff date: June 26, 2025 midnight UTC # Ignore all token counts before this date - cutoff_date = datetime(2025, 6, 27, 0, 0, 0, tzinfo=timezone.utc) + cutoff_date = datetime(2025, 6, 26, 0, 0, 0, tzinfo=timezone.utc) - # Use the later of the two dates (start of month or cutoff date) start_of_month = max(start_of_month, cutoff_date) - # First get all threads for this user - threads_result = await client.table('threads') \ - .select('thread_id') \ - .eq('account_id', user_id) \ - .execute() + # First get all threads for this user in batches + batch_size = 1000 + offset = 0 + all_threads = [] - if not threads_result.data: - return 0.0 + while True: + threads_batch = await client.table('threads') \ + .select('thread_id') \ + .eq('account_id', user_id) \ + .gte('created_at', start_of_month.isoformat()) \ + .range(offset, offset + batch_size - 1) \ + .execute() + + if not threads_batch.data: + break + + all_threads.extend(threads_batch.data) + + # If we got less than batch_size, we've reached the end + if len(threads_batch.data) < batch_size: + break + + offset += batch_size - thread_ids = [t['thread_id'] for t in threads_result.data] - + if not all_threads: + return {"logs": [], "has_more": False} + thread_ids = [t['thread_id'] for t in all_threads] + + # Fetch usage messages with pagination, including thread project info start_time = time.time() - token_messages = await client.table('messages') \ - .select('content') \ + messages_result = await client.table('messages') \ + .select( + 'message_id, thread_id, created_at, content, threads!inner(project_id)' + ) \ .in_('thread_id', thread_ids) \ - .gte('created_at', start_of_month.isoformat()) \ .eq('type', 'assistant_response_end') \ + .gte('created_at', start_of_month.isoformat()) \ + .order('created_at', desc=True) \ + .range(page * items_per_page, (page + 1) * items_per_page - 1) \ .execute() + end_time = time.time() execution_time = end_time - start_time - logger.info(f"Database query for token messages took {execution_time:.3f} seconds") + logger.info(f"Database query for usage logs took {execution_time:.3f} seconds") - if not token_messages.data: - return 0.0 + if not messages_result.data: + return {"logs": [], "has_more": False} - # Calculate total cost per message (to handle different models correctly) - total_cost = 0.0 + # Process messages into usage log entries + processed_logs = [] - for run in token_messages.data: - prompt_tokens = run['content']['usage']['prompt_tokens'] - completion_tokens = run['content']['usage']['completion_tokens'] - model = run['content']['model'] + for message in messages_result.data: + try: + # Safely extract usage data with defaults + content = message.get('content', {}) + usage = content.get('usage', {}) + + # Ensure usage has required fields with safe defaults + prompt_tokens = usage.get('prompt_tokens', 0) + completion_tokens = usage.get('completion_tokens', 0) + model = content.get('model', 'unknown') + + # Safely calculate total tokens + total_tokens = (prompt_tokens or 0) + (completion_tokens or 0) + + # Calculate estimated cost using the same logic as calculate_monthly_usage + estimated_cost = calculate_token_cost( + prompt_tokens, + completion_tokens, + model + ) + + # Safely extract project_id from threads relationship + project_id = 'unknown' + if message.get('threads') and isinstance(message['threads'], list) and len(message['threads']) > 0: + project_id = message['threads'][0].get('project_id', 'unknown') + + processed_logs.append({ + 'message_id': message.get('message_id', 'unknown'), + 'thread_id': message.get('thread_id', 'unknown'), + 'created_at': message.get('created_at', None), + 'content': { + 'usage': { + 'prompt_tokens': prompt_tokens, + 'completion_tokens': completion_tokens + }, + 'model': model + }, + 'total_tokens': total_tokens, + 'estimated_cost': estimated_cost, + 'project_id': project_id + }) + except Exception as e: + logger.warning(f"Error processing usage log entry for message {message.get('message_id', 'unknown')}: {str(e)}") + continue + + # Check if there are more results + has_more = len(processed_logs) == items_per_page + + return { + "logs": processed_logs, + "has_more": has_more + } - # Check if we have hardcoded pricing for this model - hardcoded_pricing = get_model_pricing(model) + +def calculate_token_cost(prompt_tokens: int, completion_tokens: int, model: str) -> float: + """Calculate the cost for tokens using the same logic as the monthly usage calculation.""" + try: + # Ensure tokens are valid integers + prompt_tokens = int(prompt_tokens) if prompt_tokens is not None else 0 + completion_tokens = int(completion_tokens) if completion_tokens is not None else 0 + + # Try to resolve the model name using MODEL_NAME_ALIASES first + resolved_model = MODEL_NAME_ALIASES.get(model, model) + + # Check if we have hardcoded pricing for this model (try both original and resolved) + hardcoded_pricing = get_model_pricing(model) or get_model_pricing(resolved_model) if hardcoded_pricing: input_cost_per_million, output_cost_per_million = hardcoded_pricing input_cost = (prompt_tokens / 1_000_000) * input_cost_per_million output_cost = (completion_tokens / 1_000_000) * output_cost_per_million message_cost = input_cost + output_cost else: - # Use litellm pricing as fallback + # Use litellm pricing as fallback - try multiple variations try: - prompt_token_cost, completion_token_cost = cost_per_token(model, int(prompt_tokens), int(completion_tokens)) - message_cost = prompt_token_cost + completion_token_cost + models_to_try = [model] + + # Add resolved model if different + if resolved_model != model: + models_to_try.append(resolved_model) + + # Try without provider prefix if it has one + if '/' in model: + models_to_try.append(model.split('/', 1)[1]) + if '/' in resolved_model and resolved_model != model: + models_to_try.append(resolved_model.split('/', 1)[1]) + + # Special handling for Google models accessed via OpenRouter + if model.startswith('openrouter/google/'): + google_model_name = model.replace('openrouter/', '') + models_to_try.append(google_model_name) + if resolved_model.startswith('openrouter/google/'): + google_model_name = resolved_model.replace('openrouter/', '') + models_to_try.append(google_model_name) + + # Try each model name variation until we find one that works + message_cost = None + for model_name in models_to_try: + try: + prompt_token_cost, completion_token_cost = cost_per_token(model_name, prompt_tokens, completion_tokens) + if prompt_token_cost is not None and completion_token_cost is not None: + message_cost = prompt_token_cost + completion_token_cost + break + except Exception as e: + logger.debug(f"Failed to get pricing for model variation {model_name}: {str(e)}") + continue + + if message_cost is None: + logger.warning(f"Could not get pricing for model {model} (resolved: {resolved_model}), returning 0 cost") + return 0.0 + except Exception as e: - logger.warning(f"Could not get pricing for model {model}: {str(e)}, skipping message") - continue + logger.warning(f"Could not get pricing for model {model} (resolved: {resolved_model}): {str(e)}, returning 0 cost") + return 0.0 - total_cost += message_cost - - # Return total cost * TOKEN_PRICE_MULTIPLIER (as per original logic) - total_cost = total_cost * TOKEN_PRICE_MULTIPLIER - logger.info(f"Total cost for user {user_id}: {total_cost}") - - return total_cost + # Apply the TOKEN_PRICE_MULTIPLIER + return message_cost * TOKEN_PRICE_MULTIPLIER + except Exception as e: + logger.error(f"Error calculating token cost for model {model}: {str(e)}") + return 0.0 async def get_allowed_models_for_user(client, user_id: str): """ @@ -797,7 +943,7 @@ async def get_subscription( price_id=free_tier_id, minutes_limit=free_tier_info.get('minutes') if free_tier_info else 0, cost_limit=free_tier_info.get('cost') if free_tier_info else 0, - current_usage=round(current_usage, 2) + current_usage=current_usage ) # Extract current plan details @@ -818,7 +964,7 @@ async def get_subscription( trial_end=datetime.fromtimestamp(subscription['trial_end'], tz=timezone.utc) if subscription.get('trial_end') else None, minutes_limit=current_tier_info['minutes'], cost_limit=current_tier_info['cost'], - current_usage=round(current_usage, 2), + current_usage=current_usage, has_schedule=False # Default ) @@ -1135,4 +1281,43 @@ async def get_available_models( except Exception as e: logger.error(f"Error getting available models: {str(e)}") - raise HTTPException(status_code=500, detail=f"Error getting available models: {str(e)}") \ No newline at end of file + raise HTTPException(status_code=500, detail=f"Error getting available models: {str(e)}") + + +@router.get("/usage-logs") +async def get_usage_logs_endpoint( + page: int = 0, + items_per_page: int = 1000, + current_user_id: str = Depends(get_current_user_id_from_jwt) +): + """Get detailed usage logs for a user with pagination.""" + try: + # Get Supabase client + db = DBConnection() + client = await db.client + + # Check if we're in local development mode + if config.ENV_MODE == EnvMode.LOCAL: + logger.info("Running in local development mode - usage logs are not available") + return { + "logs": [], + "has_more": False, + "message": "Usage logs are not available in local development mode" + } + + # Validate pagination parameters + if page < 0: + raise HTTPException(status_code=400, detail="Page must be non-negative") + if items_per_page < 1 or items_per_page > 1000: + raise HTTPException(status_code=400, detail="Items per page must be between 1 and 1000") + + # Get usage logs + result = await get_usage_logs(client, current_user_id, page, items_per_page) + + return result + + except HTTPException: + raise + except Exception as e: + logger.error(f"Error getting usage logs: {str(e)}") + raise HTTPException(status_code=500, detail=f"Error getting usage logs: {str(e)}") \ No newline at end of file diff --git a/frontend/src/components/billing/usage-logs.tsx b/frontend/src/components/billing/usage-logs.tsx index 3ddd6da2..ff3bdd21 100644 --- a/frontend/src/components/billing/usage-logs.tsx +++ b/frontend/src/components/billing/usage-logs.tsx @@ -1,7 +1,6 @@ 'use client'; -import { useEffect, useState, useMemo } from 'react'; -import { createClient } from '@/lib/supabase/client'; +import { useState, useEffect } from 'react'; import { Card, CardContent, @@ -27,26 +26,12 @@ import { Badge } from '@/components/ui/badge'; import { Skeleton } from '@/components/ui/skeleton'; import { Button } from '@/components/ui/button'; import { ExternalLink, Loader2 } from 'lucide-react'; -import { isLocalMode } from '@/lib/config'; -import { useAvailableModels } from '@/hooks/react-query/subscriptions/use-model'; import Link from 'next/link'; import { OpenInNewWindowIcon } from '@radix-ui/react-icons'; +import { useUsageLogs } from '@/hooks/react-query/subscriptions/use-billing'; +import { UsageLogEntry } from '@/lib/api'; + -interface UsageLogEntry { - message_id: string; - thread_id: string; - created_at: string; - content: { - usage: { - prompt_tokens: number; - completion_tokens: number; - }; - model: string; - }; - total_tokens: number; - estimated_cost: number | string; - project_id: string; -} interface DailyUsage { date: string; @@ -62,259 +47,32 @@ interface Props { } export default function UsageLogs({ accountId }: Props) { - const [usageLogs, setUsageLogs] = useState([]); - const [loading, setLoading] = useState(true); - const [error, setError] = useState(null); const [page, setPage] = useState(0); + const [allLogs, setAllLogs] = useState([]); const [hasMore, setHasMore] = useState(true); - const [loadingMore, setLoadingMore] = useState(false); - - // Use React Query hook instead of manual fetching - const { - data: modelsData, - isLoading: isLoadingModels, - error: modelsError, - } = useAvailableModels(); - + const ITEMS_PER_PAGE = 1000; - // Helper function to normalize model names for better matching - const normalizeModelName = (name: string): string => { - return name - .toLowerCase() - .replace(/[-_.]/g, '') // Remove hyphens, underscores, dots - .replace(/\s+/g, '') // Remove spaces - .replace(/latest$/, '') // Remove 'latest' suffix - .replace(/preview$/, '') // Remove 'preview' suffix - .replace(/\d{8}$/, ''); // Remove date suffixes like 20250514 - }; - - // Helper function to find matching pricing for a model - const findModelPricing = ( - modelName: string, - pricingData: Record, - ) => { - // Direct match first - if (pricingData[modelName]) { - return pricingData[modelName]; - } - - // Try normalized matching - const normalizedTarget = normalizeModelName(modelName); - - for (const [pricingKey, pricingValue] of Object.entries(pricingData)) { - const normalizedKey = normalizeModelName(pricingKey); - - // Exact normalized match - if (normalizedKey === normalizedTarget) { - return pricingValue; - } - - // Partial matches - check if one contains the other - if ( - normalizedKey.includes(normalizedTarget) || - normalizedTarget.includes(normalizedKey) - ) { - return pricingValue; - } - - // Try matching without provider prefix from pricing key - const keyWithoutProvider = pricingKey.replace(/^[^\/]+\//, ''); - const normalizedKeyWithoutProvider = - normalizeModelName(keyWithoutProvider); - - if ( - normalizedKeyWithoutProvider === normalizedTarget || - normalizedKeyWithoutProvider.includes(normalizedTarget) || - normalizedTarget.includes(normalizedKeyWithoutProvider) - ) { - return pricingValue; - } - - // Try matching the end part of the pricing key with the model name - const pricingKeyParts = pricingKey.split('/'); - const lastPart = pricingKeyParts[pricingKeyParts.length - 1]; - const normalizedLastPart = normalizeModelName(lastPart); - - if ( - normalizedLastPart === normalizedTarget || - normalizedLastPart.includes(normalizedTarget) || - normalizedTarget.includes(normalizedLastPart) - ) { - return pricingValue; - } - } - - console.log(`No pricing match found for: "${modelName}"`); - return null; - }; - - // Create pricing lookup from models data - const modelPricing = useMemo(() => { - if (!modelsData?.models) { - return {}; - } - - const pricing: Record = {}; - modelsData.models.forEach((model) => { - if ( - model.input_cost_per_million_tokens && - model.output_cost_per_million_tokens - ) { - // Use the model.id as the key, which should match the model names in usage logs - pricing[model.id] = { - input: model.input_cost_per_million_tokens, - output: model.output_cost_per_million_tokens, - }; - - // Also try to match by display_name and short_name if they exist - if (model.display_name && model.display_name !== model.id) { - pricing[model.display_name] = { - input: model.input_cost_per_million_tokens, - output: model.output_cost_per_million_tokens, - }; - } - - if (model.short_name && model.short_name !== model.id) { - pricing[model.short_name] = { - input: model.input_cost_per_million_tokens, - output: model.output_cost_per_million_tokens, - }; - } - } - }); - - console.log( - 'Pricing lookup ready with', - Object.keys(pricing).length, - 'entries', - ); - return pricing; - }, [modelsData, isLoadingModels, modelsError]); - - const calculateTokenCost = ( - promptTokens: number, - completionTokens: number, - model: string, - ): number | string => { - // Use the more lenient matching function - const costs = findModelPricing(model, modelPricing); - - if (costs) { - // Convert from per-million to per-token costs - return ( - (promptTokens / 1000000) * costs.input + - (completionTokens / 1000000) * costs.output - ); - } - - // Return "unknown" instead of fallback cost - return 'unknown'; - }; - - const fetchUsageLogs = async ( - pageNum: number = 0, - append: boolean = false, - ) => { - try { - if (!append) setLoading(true); - else setLoadingMore(true); - - const supabase = createClient(); - - // First, get all thread IDs for this user - const { data: threads, error: threadsError } = await supabase - .from('threads') - .select('thread_id') - .eq('account_id', accountId); - - if (threadsError) throw threadsError; - - if (!threads || threads.length === 0) { - setUsageLogs([]); - setLoading(false); - setLoadingMore(false); - setHasMore(false); - return; - } - - const threadIds = threads.map((t) => t.thread_id); - - // Then fetch usage messages with pagination, including thread project info - const { data: messages, error: messagesError } = await supabase - .from('messages') - .select( - ` - message_id, - thread_id, - created_at, - content, - threads!inner(project_id) - `, - ) - .in('thread_id', threadIds) - .eq('type', 'assistant_response_end') - .order('created_at', { ascending: false }) - .range(pageNum * ITEMS_PER_PAGE, (pageNum + 1) * ITEMS_PER_PAGE - 1); - - if (messagesError) throw messagesError; - - const processedLogs: UsageLogEntry[] = (messages || []).map((message) => { - const usage = message.content?.usage || { - prompt_tokens: 0, - completion_tokens: 0, - }; - const model = message.content?.model || 'unknown'; - const totalTokens = usage.prompt_tokens + usage.completion_tokens; - const estimatedCost = calculateTokenCost( - usage.prompt_tokens, - usage.completion_tokens, - model, - ); - - return { - message_id: message.message_id, - thread_id: message.thread_id, - created_at: message.created_at, - content: { - usage, - model, - }, - total_tokens: totalTokens, - estimated_cost: estimatedCost, - project_id: message.threads?.[0]?.project_id || 'unknown', - }; - }); - - if (append) { - setUsageLogs((prev) => [...prev, ...processedLogs]); - } else { - setUsageLogs(processedLogs); - } - - setHasMore(processedLogs.length === ITEMS_PER_PAGE); - } catch (err) { - console.error('Error fetching usage logs:', err); - setError( - err instanceof Error ? err.message : 'Failed to fetch usage logs', - ); - } finally { - setLoading(false); - setLoadingMore(false); - } - }; + // Use React Query hook for the current page + const { data: currentPageData, isLoading, error, refetch } = useUsageLogs(page, ITEMS_PER_PAGE); + // Update accumulated logs when new data arrives useEffect(() => { - // Only fetch usage logs after models data is loaded - if (!isLoadingModels && modelsData) { - fetchUsageLogs(0, false); + if (currentPageData) { + if (page === 0) { + // First page - replace all logs + setAllLogs(currentPageData.logs || []); + } else { + // Subsequent pages - append to existing logs + setAllLogs(prev => [...prev, ...(currentPageData.logs || [])]); + } + setHasMore(currentPageData.has_more || false); } - }, [accountId, isLoadingModels, modelsData]); + }, [currentPageData, page]); const loadMore = () => { const nextPage = page + 1; setPage(nextPage); - fetchUsageLogs(nextPage, true); }; const formatDate = (dateString: string) => { @@ -322,8 +80,8 @@ export default function UsageLogs({ accountId }: Props) { }; const formatCost = (cost: number | string) => { - if (typeof cost === 'string') { - return cost; + if (typeof cost === 'string' || cost === 0) { + return typeof cost === 'string' ? cost : '$0.0000'; } return `$${cost.toFixed(4)}`; }; @@ -380,24 +138,9 @@ export default function UsageLogs({ accountId }: Props) { ); }; - if (isLocalMode()) { - return ( - - - Usage Logs - - -
-

- Usage logs are not available in local development mode -

-
- - - ); - } - if (loading || isLoadingModels) { + + if (isLoading && page === 0) { return ( @@ -415,7 +158,7 @@ export default function UsageLogs({ accountId }: Props) { ); } - if (error || modelsError) { + if (error) { return ( @@ -424,11 +167,7 @@ export default function UsageLogs({ accountId }: Props) {

- Error:{' '} - {error || - (modelsError instanceof Error - ? modelsError.message - : 'Failed to load data')} + Error: {error.message || 'Failed to load usage logs'}

@@ -436,8 +175,26 @@ export default function UsageLogs({ accountId }: Props) { ); } - const dailyUsage = groupLogsByDate(usageLogs); - const totalUsage = usageLogs.reduce( + // Handle local development mode message + if (currentPageData?.message) { + return ( + + + Usage Logs + + +
+

+ {currentPageData.message} +

+
+ + + ); + } + + const dailyUsage = groupLogsByDate(allLogs); + const totalUsage = allLogs.reduce( (sum, log) => sum + (typeof log.estimated_cost === 'number' ? log.estimated_cost : 0), 0, @@ -558,10 +315,10 @@ export default function UsageLogs({ accountId }: Props) {