[qa] Get rid of duplicate code #9221

pull MarcoFalke wants to merge 1 commits into bitcoin:master from MarcoFalke:Mf1611-qaMaxUploadDuplCode changing 3 files +19 −82

MarcoFalke commented at 5:12 PM on November 25, 2016: member

This removes some unused and duplicate code in maxuploadtarget.py
MarcoFalke added the label Tests on Nov 25, 2016
sdaftuar commented at 8:18 PM on November 28, 2016: member

Perhaps you could also do a similar replacement for pruning.py, which uses this same duplicated mine_full_block() function?

in qa/rpc-tests/maxuploadtarget.py:None in fa130b8f49 outdated

 133 | +        mine_large_block(self.nodes[0])
 134 |  
 135 |          # We'll be requesting this new block too
 136 |          big_new_block = self.nodes[0].getbestblockhash()
 137 | -        new_block_size = self.nodes[0].getblock(big_new_block)['size']
 138 | +        # print(self.nodes[0].getblock(big_new_block)['size'])

mrbandrews commented at 5:37 PM on December 1, 2016:

nit: think you left this in accidentally

in qa/rpc-tests/pruning.py:None in fa130b8f49 outdated

  84 | @@ -95,17 +85,17 @@ def create_chain_with_staleblocks(self):
  85 |              stop_node(self.nodes[0],0)
  86 |              self.nodes[0]=start_node(0, self.options.tmpdir, ["-debug","-maxreceivebuffer=20000","-blockmaxsize=999000", "-checkblocks=5"], timewait=900)
  87 |              # Mine 24 blocks in node 1
  88 | -            self.utxo = self.nodes[1].listunspent()
  89 | +            utxo = self.nodes[1].listunspent()

mrbandrews commented at 5:37 PM on December 1, 2016:

nit: can remove this line

in qa/rpc-tests/pruning.py:None in fa130b8f49 outdated

  94 |                  else:
  95 |                      self.nodes[1].generate(1) #tx's already in mempool from previous disconnects
  96 |  
  97 |              # Reorg back with 25 block chain from node 0
  98 | -            self.utxo = self.nodes[0].listunspent()
  99 | +            utxo = self.nodes[0].listunspent()

mrbandrews commented at 5:37 PM on December 1, 2016:

same

mrbandrews commented at 5:39 PM on December 1, 2016: contributor

A couple of nits, otherwise ACK
MarcoFalke renamed this:
~~[qa] maxuploadtarget: Get rid of duplicate code~~
[qa] Get rid of duplicate code
on Dec 2, 2016
[qa] Get rid of duplicate code facbfa5816
MarcoFalke force-pushed on Dec 2, 2016
MarcoFalke commented at 3:32 PM on December 2, 2016: member

Squashed and fixed nits: fa130b8..facbfa5
MarcoFalke merged this on Dec 2, 2016
MarcoFalke closed this on Dec 2, 2016
MarcoFalke referenced this in commit 9e4bb312e6 on Dec 2, 2016
MarcoFalke deleted the branch on Dec 2, 2016
morcos commented at 4:16 AM on December 4, 2016: member

@MarcoFalke @sdaftuar @mrbandrews Did you test this? It seems to have slowed pruning.py by 35%. I don't think its worth it at the expense of adding 10 mins to the pruning.py test.
MarcoFalke commented at 12:09 PM on December 4, 2016: member

Jup, I was running the tests after squash. Will take a look at the slowdown. Should be easy to fix if this commit caused it.
morcos commented at 1:55 PM on December 4, 2016: member

@marcofalke my guess is it's the repeated calling of listunspent instead of caching the utxo's. well I should say that's the first difference I saw and then was too tired to investigate further.
codablock referenced this in commit 2028e6dbdc on Jan 16, 2018
codablock referenced this in commit 3b0e3741c6 on Jan 16, 2018
codablock referenced this in commit 5d1f05afff on Jan 17, 2018
andvgal referenced this in commit 391bd4b168 on Jan 6, 2019
CryptoCentric referenced this in commit 20e609b0d2 on Feb 25, 2019
DrahtBot locked this on Sep 8, 2021

Contributors

Labels

Linked (view graph)

#9274 [qa] Use cached utxo set to fix performance regression