Data Storage

TradeFlow uses IndexedDB via the Dexie.js library for all data persistence. This provides structured, queryable storage directly in the browser with no server required.

Why IndexedDB?

• Persistent — Data survives browser restarts (unlike in-memory state)
• Structured — Supports indexes, queries, and transactions (unlike localStorage)
• Large capacity — Can store megabytes of data including images (unlike localStorage’s ~5MB limit)
• No server needed — Everything runs client-side

Database Schema

The database is named TradeFlowDB and has evolved through 5 schema versions:

Version 1 — Core Tables

trades:     'id, symbol, direction, entryDate, exitDate, strategyId, accountId'
accounts:   'id, name'
strategies: 'id, name'

Version 2 — Financial Fields

Added fees, commissions, and tradeDuration to existing trades with migration logic:

• fees and commissions default to 0
• tradeDuration is computed from entryDate and exitDate (in seconds)

Version 3 — Settings Table

settings: 'key'

Used for storing the OpenAI API key and model selection.

Version 4 — Trade Images

Added images array to trades (default []). Existing trades are migrated to include an empty images array.

Version 5 — Mentor Mode Tables

students:           'id, name'
studentTrades:      'id, studentId, symbol, direction, entryDate, exitDate, strategyId'
studentStrategies:  'id, studentId, name'

Complete data isolation for mentor-student relationships.

Table Details

trades

The primary table storing all of the mentor’s own trades.

Field	Type	Description
id	string	UUID primary key
symbol	string	Ticker symbol
direction	string	“long” or “short”
entryPrice	number	Entry price
exitPrice	number	Exit price
entryDate	string	ISO date-time of entry
exitDate	string	ISO date-time of exit
positionSize	number	Contracts/shares
pnl	number	Profit or loss
strategyId	string	FK to strategies table
accountId	string	FK to accounts table
tags	array	Array of tag strings
notes	string	Free-text notes
takeProfits	array	Array of {price, quantity, date}
fees	number	Trading fees
commissions	number	Broker commissions
tradeDuration	number	Duration in seconds
images	array	Array of base64 JPEG strings

accounts

Field	Type	Description
id	string	UUID primary key
name	string	Account display name

strategies

Field	Type	Description
id	string	UUID primary key
name	string	Strategy display name

settings

Field	Type	Description
key	string	Setting identifier (primary key)
value	any	Setting value

Currently stores: openai_api_key, openai_model.

students

Field	Type	Description
id	string	UUID primary key
name	string	Student name
notes	string	Optional notes
createdAt	string	ISO date-time

studentTrades

Same schema as trades but with an additional studentId field. Indexed by studentId for efficient per-student queries.

studentStrategies

Field	Type	Description
id	string	UUID primary key
studentId	string	FK to students table
name	string	Strategy display name

Reactive Queries

TradeFlow uses Dexie’s useLiveQuery hook for reactive data binding:

import { useLiveQuery } from 'dexie-react-hooks';
import { db } from '@/db';

const trades = useLiveQuery(() => db.trades.toArray());

When any operation modifies the trades table, all components using a useLiveQuery on that table automatically re-render. This eliminates the need for manual state synchronization or Redux-style stores.

Versioning and Migrations

Dexie handles schema versioning automatically. Each db.version(n) call can include an .upgrade() function that runs once when a user’s database is at a lower version. This ensures:

• Existing users get new fields added to their data
• Default values are applied to existing records
• No data is lost during upgrades

Backup Format

See the Export & Backup Guide for details on the ZIP snapshot format used for data portability.