chore: remove hardcoded fork reference osaka and fetch latest fork #1337

Dyslex7c · 2025-07-19T10:04:21Z

What was wrong?

Related to Issue #1330
Hardcoded fork reference osaka was being used in tox.ini

How was it fixed?

Implemented a CLI command ethereum-spec-forks to fetch hardfork information. Used it in tox.ini to fetch the latest fork.

Closes #1330

Cute Animal Picture

* Update to latest EIP-7918 * EIP-7918 t8n tool update --------- Co-authored-by: Sam Wilson <[email protected]> Co-authored-by: Guruprasad Kamath <[email protected]>

* feat(vm): add eip-7939 count leading zeros opcode. * Update src/ethereum/osaka/vm/instructions/bitwise.py Co-authored-by: Guruprasad Kamath <[email protected]> * code formatting --------- Co-authored-by: Guruprasad Kamath <[email protected]> Co-authored-by: Guruprasad Kamath <[email protected]>

…state management and interpreter. Updated comments for clarity on the implications of SELFDESTRUCT.

…MAX_CODE_SIZE across multiple modules

update exceptions for other forks fix tests use fork module rename to NonceOverFlowError revert load test catching specific exc style: indent comment line

- ignore all pytest functions - remove unused variables, functions, methods, and classes - create vulture whitelist and mark some code that either implicitly gets called or only gets called during test runs for vulture to ignore

Signed-off-by: bytetigers <[email protected]>

* update dependencies * add fill to the entry points * add eest tests * run eest tests in py3 * create EELS TransitionTool subclass * post review fix and latest eest * remove pytest-eest.ini * minor fix: fix tox config * minor fix: add vulture exception * run eest tests in pypy * update optimized tox command

SamWilsn · 2025-07-22T17:24:13Z

pyproject.toml

@@ -34,6 +34,7 @@ dependencies = [
 package-dir = {"" = "src"}
 packages = [
    "ethereum_spec_tools",
+    "ethereum_spec_tools.cli",


The other CLI tools exist directly under ethereum_spec_tools. For consistency, this tool should as well.

SamWilsn · 2025-07-22T17:26:09Z

src/ethereum_spec_tools/cli/forks.py

+    from hardfork import Hardfork
+
+
+def get_fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:


I'm partial to omitting get_.

Suggested change

def get_fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:

def fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:

SamWilsn · 2025-07-22T17:35:10Z

src/ethereum_spec_tools/cli/forks.py

+    if index < 0:
+        # Negative indexing: -1 is last, -2 is second last, etc.
+        if abs(index) <= len(forks):
+            return forks[index]
+        else:
+            return None
+    else:
+        # Positive indexing: 0 is first (Frontier), 1 is second, etc.
+        if index < len(forks):
+            return forks[index]
+        else:
+            return None


Python lists natively support negative indexing. Is there something that makes return forks[index] infeasible?

yes if the index is out if bounds, it looks more verbose than a try-catch block which is slower in case an exception occurs

I don't think we care that much about performance in this tool 🤣

SamWilsn · 2025-07-22T17:44:01Z

tox.ini

 commands =
    fill \
        -m "not slow and not zkevm and not benchmark" \
        -n auto --maxprocesses 6 \
        --basetemp="{temp_dir}/pytest" \
        --clean \
        eest_tests/execution-spec-tests/tests
-    fill \
+    bash -c 'fill \


This'll break our CI on Windows... There's no other way to do a variable substitution from a command?

If not, I don't think this approach will work after all. We might have to do it as a pytest plugin (or something in conftest.py) that takes a --fork-number argument instead of a separate command line tool.

We can create platform specific commands separately for windows, although I'm not sure if that will work

I would like to avoid anything platform specific if we can.

I couldn't think of any other way to implement variable substitution that is platform agnostic. The only way that comes to my mind is probably running it as a python script

Dyslex7c · 2025-07-23T11:04:18Z

@SamWilsn I've added all the changes you've mentioned, please do take a look

SamWilsn · 2025-07-25T20:24:50Z

I've rebased forks/osaka to pick up the last straggling commits from master. If you do something like this, should be easy enough to resolve the conflicts:

git checkout forks/osaka
git pull
git checkout tox/remove-hardcoded-fork-osaka
git reset --hard forks/osaka
git cherry-pick 6e993fca90bea968771bf6decebb2cc5e6cfed4d
git cherry-pick 52403548c40109a32b66316224ebb6141c580ea1

Dyslex7c · 2025-07-26T14:18:50Z

I've rebased forks/osaka to pick up the last straggling commits from master. If you do something like this, should be easy enough to resolve the conflicts:
git checkout forks/osaka
git pull
git checkout tox/remove-hardcoded-fork-osaka
git reset --hard forks/osaka
git cherry-pick 6e993fca90bea968771bf6decebb2cc5e6cfed4d
git cherry-pick 52403548c40109a32b66316224ebb6141c580ea1

For some reason, it's showing my forks/osaka branch is 36 commits ahead of, 45 commits behind ethereum/execution-specs:forks/osaka. What to do now? I can't sync the latest changes

SamWilsn and others added 30 commits June 25, 2025 20:02

Initial Osaka commit

1f5b621

Implement EIP-7823

268f67f

Implement EIP-7825

7f4fc84

Implement EIP-7883

0af9e4e

Implement EIP-7918

30b1cd2

Eip 7918 changes (ethereum#1273)

ba4c566

* Update to latest EIP-7918 * EIP-7918 t8n tool update --------- Co-authored-by: Sam Wilson <[email protected]> Co-authored-by: Guruprasad Kamath <[email protected]>

run osaka tests (ethereum#1290)

9f9bdb7

chore(osaka): update modexp eip-7883.

0a1eebf

Implement EIP-7934 (ethereum#1289)

96fed55

Implement EIP-7907

f069e08

introduce warm_code_addresses

4708376

feat: implement eip-7951

275d66a

Code review feedback

54aa418

Bump version to 2.18.0rc2.dev1

1d290e8

Grab latest EEST

0aa92a0

Bump version to 2.18.0rc2

f947ab1

Enhance account creation handling to respect EIP-6780 constraints in …

afdd3b7

…state management and interpreter. Updated comments for clarity on the implications of SELFDESTRUCT.

Update docs to mention EIP-6780 (prague,osaka)

9a5a6a8

Make workflows agnostic to default branch name

5614686

Changed to_address to to_address_masked

bcee9d9

improve readability of calculations

bfa4b14

fix: update EIP-7610 link to official EIP page

d12896a

Refactor transaction validation to use MAX_INIT_CODE_SIZE instead of …

e6108ed

…MAX_CODE_SIZE across multiple modules

add specific exceptions on tests

7b1694e

update exceptions for other forks fix tests use fork module rename to NonceOverFlowError revert load test catching specific exc style: indent comment line

Turn vulture fully on, remove dead code, whitelist

16762ec

- ignore all pytest functions - remove unused variables, functions, methods, and classes - create vulture whitelist and mark some code that either implicitly gets called or only gets called during test runs for vulture to ignore

remove setup.cfg

7631b97

Switch to newer license metadata format

e9e94ed

chore: fix some minor issues

5bab4da

Signed-off-by: bytetigers <[email protected]>

update eip7951 gas

a91a505

souradeep-das and others added 7 commits July 15, 2025 14:22

update eip7918

2171f03

update clz gas cost

07bd1f1

Update EIP-7883: Change modexp gas calculation (ethereum#1318)

2eb1ba4

eip-7825: update tx gas limit (ethereum#1317)

58d6dca

Remove redundant set_account call in json_to_state

20cdaf9

chore: remove hardcoded fork reference osaka and fetch latest fork

6e993fc

SamWilsn reviewed Jul 22, 2025

View reviewed changes

fix(tox): add windows command for EEST tests

5240354

Dyslex7c requested a review from SamWilsn July 23, 2025 11:04

SamWilsn force-pushed the forks/osaka branch 2 times, most recently from 301eb89 to 5a49b2f Compare July 25, 2025 19:55

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

chore: remove hardcoded fork reference osaka and fetch latest fork #1337

chore: remove hardcoded fork reference osaka and fetch latest fork #1337

Uh oh!

Dyslex7c commented Jul 19, 2025

Uh oh!

SamWilsn Jul 22, 2025

Uh oh!

SamWilsn Jul 22, 2025

Uh oh!

SamWilsn Jul 22, 2025

Uh oh!

Dyslex7c Jul 23, 2025

Uh oh!

SamWilsn Jul 25, 2025

Uh oh!

SamWilsn Jul 22, 2025

Uh oh!

Dyslex7c Jul 23, 2025

Uh oh!

SamWilsn Jul 25, 2025

Uh oh!

Dyslex7c Jul 26, 2025

Uh oh!

Dyslex7c commented Jul 23, 2025

Uh oh!

SamWilsn commented Jul 25, 2025

Uh oh!

Dyslex7c commented Jul 26, 2025

Uh oh!

Uh oh!

		from hardfork import Hardfork


		def get_fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:

	def get_fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:
	def fork_by_index(forks: List[Hardfork], index: int) -> Optional[Hardfork]:

chore: remove hardcoded fork reference osaka and fetch latest fork #1337

Are you sure you want to change the base?

chore: remove hardcoded fork reference osaka and fetch latest fork #1337

Uh oh!

Conversation

Dyslex7c commented Jul 19, 2025

What was wrong?

How was it fixed?

Cute Animal Picture

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Dyslex7c commented Jul 23, 2025

Uh oh!

SamWilsn commented Jul 25, 2025

Uh oh!

Dyslex7c commented Jul 26, 2025

Uh oh!

Uh oh!