How to determine which parameters are required when using substrate.query() to query a storage_function from polkascan's py-substrate-interface?

Question

When querying storage functions using Polkascan's py-substrate-interface how do you figure out which parameters are required?

For example, I am trying to discern how much KSM is on Karura via storage functions in the tokens (ORML) pallet.

See below code for example. I tried to substrate.query() Tokens.TotalIssuance storage_function but it requires 6 parameters:

from substrateinterface import SubstrateInterface

substrate = SubstrateInterface(url="wss://karura-rpc-0.aca-api.network/") 

result = substrate.query(module='Tokens',storage_function='TotalIssuance')
        
**ValueError**: Storage function requires 6 parameters, 0 given

Note, I also tried to to use substrate.query_map() for Token.Accounts storage function but that requires 1 parameter which I can't discern.

result = substrate.query_map(
module='Tokens',
storage_function='Accounts',
page_size=10,
max_results = 100)

**ValueError**: Storage function map requires 1 parameters, 0 given

Does anyone have any advice on how to find out which parameters are required for substrate.query() & substrate.query_map()? I have been digging around in the library docs along with substrate storage documentation but I can't figure it out.

Other Note: Through trial and error and comparing with Polkadot.JS queries (see screenshot) I managed to figure out that substrate.query() Tokens.Accounts for a KSM balance of a single Karura account requires two paramaters: params = ['karura_address',{'Token':'KSM'}].

Thanks in advance!

Answer 1

Update: The storage_function.get_param_info() helper function is introduced, see: https://github.com/polkascan/py-substrate-interface#type-information-about-how-to-format-parameters

---

It is indeed a bit cumbersome to figure out a) what are the parameters for a storage function when this is not a primitive but a complex type and b) how to specify this that is correctly encoded.

I tried to figured this out purely using the library and this resulted in the following example:

# Retrieve storage function to get information about value and param types
storage_function = substrate.get_metadata_storage_function("Tokens", "TotalIssuance")

# Amount of params
print(len(storage_function.get_params_type_string())) # 1

# Retrieve a SCALE object of the param type
param_type = substrate.create_scale_object(storage_function.get_params_type_string()[0])

# Possible options of Enum type
print(param_type.type_mapping)
# [('Token', 'scale_info::42'), ('DexShare', '(scale_info::43, scale_info::43)'), ('Erc20', 'scale_info::44'), ('StableAssetPoolToken', 'scale_info::4'), ('LiquidCrowdloan', 'scale_info::4'), ('ForeignAsset', 'scale_info::35')]

# So if I want to use the 'Token' option: SCALE type is 'scale_info::42'
token_param = substrate.create_scale_object(param_type.type_mapping[0][1])

# Show possible options of Tokens
print(token_param.type_mapping)
# [('ACA', 'Null'), ('AUSD', 'Null'), ('DOT', 'Null'), ('LDOT', 'Null'), (None, 'Null'), (None, 'Null') ....

# So now if I want to show the TotalIssuance of the `Token` `KSM`:
result = substrate.query(
    module='Tokens',
    storage_function='TotalIssuance',
    params=[{'Token': 'KSM'}]
)

# Result is U128 object with value 81190424615873229
print(result.value)
# 81190424615873229

As said, this is not very intuitive.. 😅

It becomes more clear when using the query_map() function (get all results, omitting the param)

# Query the Mapped storage function
result_map = substrate.query_map(module='Tokens', storage_function='TotalIssuance')

print(list(result_map))
# [[<scale_info::41(value={'DexShare': ({'Token': 'KSM'}, {'Token': 'LKSM'})})>, <U128(value=19811150500020714)>], [<scale_info::41(value={'DexShare': ({'Token': 'KUSD'}, {'Token': 'BNC'})})>, <U128(value=2953340214648797607)>], [<scale_info::41(value={'DexShare': ({'Token': 'KSM'}, {'ForeignAsset': 0})})>, <U128(value=5454135724694137)>], ...

Here you will see the key to use immediately. There is clearly a need for an easier way to retrieve this information from the storage function, I will submit an issue for this feature request