Iteration in fee approximation #565
Conversation
|
The PR needs the corresponding raiden PR merged first. I'll update it as soon as that PR is ready. |
| @@ -8,7 +8,7 @@ | |||
|
|
|||
| from pathfinding_service.constants import DEFAULT_REVEAL_TIMEOUT | |||
| from pathfinding_service.exceptions import InvalidPFSFeeUpdate | |||
| from raiden.exceptions import UndefinedMediationFee | |||
| from raiden.tests.utils.mediation_fees import fee_receiver, fee_sender | |||
There was a problem hiding this comment.
ok what I understand is that you take the fee calculation from the Raiden repo to do it in the pfs, now. Is that correct?
There was a problem hiding this comment.
but why do you take this crucial function from the Raiden tests?
There was a problem hiding this comment.
I am happy if that works, but we might need to refactor that anytime after the RC, right?
There was a problem hiding this comment.
Yes, we need to do that (see commit message).
| @@ -8,7 +8,7 @@ | |||
|
|
|||
| from pathfinding_service.constants import DEFAULT_REVEAL_TIMEOUT | |||
| from pathfinding_service.exceptions import InvalidPFSFeeUpdate | |||
| from raiden.exceptions import UndefinedMediationFee | |||
| from raiden.tests.utils.mediation_fees import fee_receiver, fee_sender | |||
There was a problem hiding this comment.
but why do you take this crucial function from the Raiden tests?
| @@ -8,7 +8,7 @@ | |||
|
|
|||
| from pathfinding_service.constants import DEFAULT_REVEAL_TIMEOUT | |||
| from pathfinding_service.exceptions import InvalidPFSFeeUpdate | |||
| from raiden.exceptions import UndefinedMediationFee | |||
| from raiden.tests.utils.mediation_fees import fee_receiver, fee_sender | |||
There was a problem hiding this comment.
I am happy if that works, but we might need to refactor that anytime after the RC, right?
| tn.set_fee(2, 1, imbalance_penalty=[(TA(0), FA(0)), (TA(200), FA(200))]) | ||
| assert tn.estimate_fee(1, 3) == 10 | ||
| tn.set_fee(2, 1, imbalance_penalty=[(TA(0), FA(0)), (TA(2000), FA(200))]) | ||
| assert tn.estimate_fee(1, 3) == 10 + 1 |
There was a problem hiding this comment.
can you add a comment that + 1 is due to the iteration?
There was a problem hiding this comment.
I wrote When approximation iterations matter, those are given as sums of the steps. in line 125. But apparently this is easily missed.
Codecov Report
@@ Coverage Diff @@
## master #565 +/- ##
========================================
Coverage ? 90.7%
========================================
Files ? 36
Lines ? 2346
Branches ? 297
========================================
Hits ? 2128
Misses ? 163
Partials ? 55
Continue to review full report at Codecov.
|
Use
fee_receiverandfee_senderfrom raiden to get the latest fee calculation. Those are currently in a test utils package and should be moved in a following PR.Closes #563.