ethereum/py-evm

Chris Calderon

@SerpentChris

On my big endian system, when I recently tried to install ethash via pip it complained about the issue that was fixed by this commit.

Chris Calderon

@SerpentChris

I think I found the issue? PyPI has an ehtash package and a pyethash package. The ethash package installs fine but py-evm tries to install the pyethash package which is broken.

Honestly I'm a bit confused here. These packages are all on PyPI: ethash, pyethash, and eth-hash.

Bryant Eisenbach

@fubuloubu

was there any change in how selfdestruct is being handled for Istanbul?

IIRC there was not, but we are experiencing some issues: ethereum/vyper#1657

Voith Mascarenhas

@voith

@fubuloubu I just had a look at the test failures and it seems like the transactions need more gas.

Bryant Eisenbach

@fubuloubu

The gas cost of SELFDESTRUCT wasn't changed though, was it?

Voith Mascarenhas

@voith

I just went through the Istanbul meta EIP and it doesn’t seem like the gas cost of SELFDESTRUCT was changed.

Bryant Eisenbach

@fubuloubu

yup

I forgot that this has nothing to do with Istanbul because the issue gets introduced when we use eth-tester v0.2b1+

(which uses py-evm 0.2.0a43+ under the hood)

Istanbul was only introduced in eth-tester v0.2b3 I believe

Voith Mascarenhas

@voith

Hmmm, yeah. But istanbul could break some tests, specially ethereum/EIPs#2200

Bryant Eisenbach

@fubuloubu

I replaced in our test cases SELFDESTRUCT with a simple send that replicates the same result, and it passes

Voith Mascarenhas

@voith

:+1:

Bryant Eisenbach

@fubuloubu

btw @carver the CHANGELOG for eth-tester still says v0.2.0b3 is unreleased

Noel Maersk

@veox

We should probably change 7_000_000 to 5_000_000 (same as Constantinople/Petersburg) on this line, if there is to be no difficulty bomb reset for Istanbul. (And looks like there won't be.) (EDIT: Or use the petersburg function altogether.)

I've done some slackjob calculations, and (given no changes in net-wide hashrate, a big given) the bomb won't make 10% of the difficulty for ~500 more days.

A tad too optimistic for my taste, since unexpected delays are always possible (as Constantinople/Petersburg showed), but I'll refrain from suggesting in ACD a "reset" should be done preemptively.

Christoph Burgdorf

@cburgdorf

Just merged Clique Consensus PR ethereum/py-evm#1855 Trinity side should follow soon.

Noel Maersk

@veox

QQ: Does newsfragments support multiple lines?..

Noel Maersk

@veox

As in, there are two noteworthy changes in PR #1878, that should both be listed...

OK, towncrier --draft is my friend. ^_^

Noel Maersk

@veox

Is there a "common repo" for the boilerplate newsfragments/README.md file? Turns out fragments from the misc category don't get their contents displayed, they just get the numbers listed in a "Miscellaneous internal changes" section.

Christoph Burgdorf

@cburgdorf

@veox if those are separate categories (e.g performance and feature) then you can create two individual files. If those are of the same category then I'm unaware of differentiating them apart from line breaks.

@veox eventually it will make it into the ethereum-python-project-template ethereum/ethereum-python-project-template#28 but we wanted to wait for the next Trinity release to see if there are any further tweaks to the release script needed.

Noel Maersk

@veox

Thanks, looks like just the thing.

Jason Carver

@carver

Turns out fragments from the misc category don't get their contents displayed, they just get the numbers listed in a "Miscellaneous internal changes" section.

Yeah, as far as I know, someone did this intentionally. I added an "internal" section to trinity that shows towncrier content here: https://github.com/ethereum/trinity/blob/62a85b1d3c89908c37fd62d9c4370dc8e3b139e2/pyproject.toml#L34-L37 -- the idea is that it's only of interest to contributors rather than users. Then misc would be for trivial changes, like maybe changing flake8 rules and updating the repo. @veox

Jason Carver

@carver

I merged in the #1858 changes (in #1871 ) and am starting in on a release...

Jason Carver

@carver

py-evm released: https://py-evm.readthedocs.io/en/latest/release_notes.html#py-evm-0-3-0-alpha-8-2019-11-05

@/all

Voith Mascarenhas

@voith

:tada:

Bryant Eisenbach

@fubuloubu

Thanks for the fix to ethereum/py-evm#1865

Jason Carver

@carver

:+1:

Jacques Wagener

@jacqueswww

:clap:

Chris Calderon

@SerpentChris

If I want to make a chain to use for testing a smart contract, can I use the Chain class instead of the MiningChain class?

Christoph Burgdorf

@cburgdorf

@SerpentChris You need the MiningChain class when you want to build blocks incrementally instead of importing blocks in one go.

Jason Carver

@carver

For testing, you might be interested in using web3.py with eth-tester (which uses py-evm under the hood)

Bryant Eisenbach

@fubuloubu

is Python 3.8 in the works for this project and eth-tester?

Jason Carver

@carver

I don't think there should be any problems w/ running it in py3.8. Do you know of anything? Obviously we'll want to add tests at some point, but it's not an active project that I know of @fubuloubu

Jason Carver

@carver

nm, I see in ethereum/vyper#1678 that there's some problem with the blake2b library

Bryant Eisenbach

@fubuloubu

We were trying to upgrade, but I think it's just been premature. Is it just me, or has 3.7/3.8 moved a lot faster than previous release?

Monir Shaker

@Moniroaf

i have some problems to compile the app.py i conudn't call the contract from the file sol ?!!

pinnaculum

@pinnaculum

congrats on py-evm this is huge

Noel Maersk

@veox

18 days later, ACD is noisy because of the difficulty bomb. :/

TL;DR: There might be an out-of-schedule call on Friday.

(I suggest you wonderful busy folk don't waste time on reading it all.)

Jason Carver

@carver

whoops, thanks for pointing it out (twice now!)

Noel Maersk

@veox

Heh, that's not what I meant by the comment, but thanks for fixing. :)

Jason Carver

@carver

Where communities thrive

People

Repo info

Activity