Conversation
- Limit on-curve liquidity
Limit max on curve liquidity
✅ Deploy Preview for incandescent-kelpie-250f0e canceled.
|
|
This fixes the scenario_generator module for monte-carlo simulations. |
| if str(amount).endswith("%"): | ||
| return ( | ||
| get_user_balance(state, user_name, tkn_name) | ||
| * Decimal(str(amount)[:-1]) |
There was a problem hiding this comment.
Same with regards to str(amount)
There was a problem hiding this comment.
same response as previous
| state = self.get_state(copy_type="initial", timestamp=timestamp) | ||
| state, tkn_name, user_name = validate_input( | ||
| state, tkn_name, user_name, timestamp | ||
| tkn_amt = userAmount(state, tkn_name, user_name, tkn_amt) |
There was a problem hiding this comment.
You can/should revert this change.
Since tkn_amt is computed based on state, tkn_name, user_name, we should first have them validated (i.e., internally amended).
There was a problem hiding this comment.
How/why would tkn_amt be computed based on state, tkn_name, user_name? This is a BancorDapp method, meaning it is the intended API for users of this codebase - hence they should be the ones specifying what the tkn_amt is. The deleted validate_input(..) did not include tkn_amt per your recent changes, presumably because the validation would have failed for string amounts with % signs, however, it makes more sense to simply call validate_input(..) on the tkn_amt directly after calling userAmount(...) (i.e., line 195 in this case). In short, Im suggesting we retain the original validate_input(..) which includes tkn_amt, but simply handle this after calling userAmount(...).
| state, source_token, user_name, timestamp | ||
| tkn_amt = userAmount(state, source_token, user_name, tkn_amt) | ||
| state, source_token, tkn_amt, user_name = validate_input( | ||
| state, source_token, tkn_amt, user_name, timestamp |
There was a problem hiding this comment.
Same (you can/should revert this change).
| state, tkn_name, user_name, timestamp | ||
| tkn_amt = userAmount(state, tkn_name, user_name, tkn_amt) | ||
| state, tkn_name, tkn_amt, user_name = validate_input( | ||
| state, tkn_name, tkn_amt, user_name, timestamp |
There was a problem hiding this comment.
Same (you can/should revert this change).
…arch into quickstudies
- changed `tkn_amt: str` to `tkn_amt: Any` - changed camelcase functions to lowercase for simulator python style consistency
- fixed some inconsistent usage
2 participants
No description provided.