Skip to main content

Query API Reference

All queries are read-only RPC calls using:

connection.getAccountInfo(PDA)

No instruction execution is required.

All accounts are Borsh-encoded and must be decoded using the exact schemas defined earlier.

Configuration Queries

These queries read protocol-level configuration from the 4 Config PDAs.

ConfigCore​

PDA​

["config"]

Purpose​

Global protocol configuration:

  • Owner
  • Authority
  • Fee parameters
  • Leverage limits
  • Margin settings
  • Order counters
  • System state (paused flag)

Query Example​

const [configCorePDA] = findConfigCoreAddress(PROGRAM_ID);
const accountInfo = await connection.getAccountInfo(configCorePDA);

if (!accountInfo) throw new Error("ConfigCore not found");

const config = ConfigCoreSchema.decode(accountInfo.data);

ConfigAssets​

PDA​

["config", "assets"]

Purpose​

Maps asset name → asset PDA.

Query Example​

const [configAssetsPDA] = findConfigAssetsAddress(PROGRAM_ID);
const accountInfo = await connection.getAccountInfo(configAssetsPDA);

const config = ConfigAssetsSchema.decode(accountInfo.data);

Returned structure:

{
  is_initialized: bool,
  asset_mappings: [
    { name: string, address: PublicKey }
  ],
  bump: u8
}

ConfigPairs​

PDA​

["config", "pairs"]

Purpose​

Stores all registered trading pairs.

Example:

BTC-USD
ETH-USD

Query Example​

const [configPairsPDA] = findConfigPairsAddress(PROGRAM_ID);
const accountInfo = await connection.getAccountInfo(configPairsPDA);

const config = ConfigPairsSchema.decode(accountInfo.data);

ConfigKeepers​

PDA​

["config", "keepers"]

Purpose​

Authorized keeper public keys.

Query Example​

const [configKeepersPDA] = findConfigKeepersAddress(PROGRAM_ID);
const accountInfo = await connection.getAccountInfo(configKeepersPDA);

const config = ConfigKeepersSchema.decode(accountInfo.data);

Trading Data Queries

Trading data is user-specific.

UserAccount

PDA​

["user_account", user, tokenMint]

Purpose​

Tracks:

  • Balances
  • Orders
  • Limit orders
  • Funding history
  • Trading stats

Query Example​

const [userAccountPDA] = findUserAccountAddress(
  PROGRAM_ID,
  userPubkey,
  tokenMint
);

const accountInfo = await connection.getAccountInfo(userAccountPDA);

const userAccount = UserAccountSchema.decode(accountInfo.data);

Positions

PDA​

["position", user, pair, direction]

direction:

0 = LONG
1 = SHORT

Query Logic​

You must check both directions for each pair.

for (const direction of [0, 1]) {
  const [positionPDA] = findPositionAddress(
    PROGRAM_ID,
    userPubkey,
    pair,
    direction
  );

  const accountInfo = await connection.getAccountInfo(positionPDA);

  if (accountInfo) {
    const position = PositionSchema.decode(accountInfo.data);
  }
}

Limit Orders

PDA​

["limit_order", orderId]

Order IDs are stored in:

userAccount.limit_orders

Query Example​

for (const orderId of userAccount.limit_orders) {
  const [pda] = findLimitOrderAddress(PROGRAM_ID, orderId);
  const accountInfo = await connection.getAccountInfo(pda);

  const order = LimitedOrderSchema.decode(accountInfo.data);
}

Deals

PDA​

["deal", orderId]

Order IDs are stored in:

userAccount.orders

Query Example​

for (const dealId of userAccount.orders) {
  const [pda] = findDealAddress(PROGRAM_ID, dealId);
  const accountInfo = await connection.getAccountInfo(pda);

  const deal = DealSchema.decode(accountInfo.data);
}

Query Execution Flow (User Trading Data)

  1. Fetch UserAccount
  2. Extract:
    • orders
    • limit_orders
  3. Fetch:
    • Positions (LONG + SHORT per pair)
    • Limit Orders
    • Deals
  4. Decode using matching Borsh schemas

Important Notes

  • All numeric values are u128, u64, or i64 and should be handled as bigint.
  • Always normalize BN → bigint after decoding.
  • If decoding fails:
    • Check schema ordering
    • Check account size
    • Confirm correct PDA derivation
  • Positions exist even if size = 0; check size > 0 to filter active ones.

All queries are deterministic and require no program execution.