Optimize test_wallet_sync #11103
Optimize test_wallet_sync #11103
Conversation
| for i in range(20): | ||
| # Tests a reorg with the wallet | ||
| ph = both_phs[i % 2] |
There was a problem hiding this comment.
Not a big deal either way, but figured I'd offer this.
https://docs.python.org/3.10/library/itertools.html#itertools.cycle
| for i in range(20): | |
| # Tests a reorg with the wallet | |
| ph = both_phs[i % 2] | |
| for i, ph in zip(range(20), itertools.cycle(both_phs)): | |
| # Tests a reorg with the wallet |
There was a problem hiding this comment.
Thank you, but that is much more complicated imo
There was a problem hiding this comment.
It is true that modulus is a common programming idiom. How about this? We really don't care about the range nor the modulus of the iteration.
| for i in range(20): | |
| # Tests a reorg with the wallet | |
| ph = both_phs[i % 2] | |
| for ph in both_phs * 10: | |
| # Tests a reorg with the wallet |
I don't mean to badger about this but I realized this much simpler form than my last suggestion. Feel free to mark this resolved.
| ) | ||
| for i in range(1000, len(new_blocks)): | ||
| for i in range(base_num_blocks + 20, len(new_blocks)): |
There was a problem hiding this comment.
Am I reading this right?
| for i in range(base_num_blocks + 20, len(new_blocks)): | |
| for i in range(len(all_blocks), len(new_blocks)): |
|
|
||
| # Insert 400 blocks | ||
| for block in default_400_blocks: | ||
| await full_node_api.full_node.respond_block(full_node_protocol.RespondBlock(block)) | ||
|
|
||
| # Farm few more with reward | ||
| for i in range(0, num_blocks - 1): |
There was a problem hiding this comment.
The test was testing only one wallet, but farming rewards take 1 block to get confirmed. So here the first wallet needs less rewards because all of them will get confirmed, in the second one, the last one will not get confirmed
There was a problem hiding this comment.
Thanks for the explanation. I think I might generally prefer to make sure all the rewards get processed rather than finishing the test with unconfirmed transactions and off-by-one compensations.
But, you can go ahead and resolve this conversation once you've read my comment. No action requested.
Co-authored-by: Kyle Altendorf <sda@fstab.net>
* Optimize test_wallet_sync * Update tests/wallet/sync/test_wallet_sync.py Co-authored-by: Kyle Altendorf <sda@fstab.net> * Fix test Co-authored-by: Kyle Altendorf <sda@fstab.net>
Running every test twice with two modes is wasteful, we can share a lot of the slow stuff