feat: embed genesis for well-known chains #2835

mojtaba-esk · 2026-02-11T07:20:03Z

Describe your changes and provide context

This PR proposes embedding genesis file for the well-known networks (pacific-1, arctic-1, atlantic-2) so users only need the seid binary.
Running seid init --chain-id pacific-1 (or arctic-1 / atlantic-2) writes the correct genesis from the binary; no separate genesis download. Adds an app/genesis package with go:embed and tests that assert SHA256 digests of the embedded files to avoid accidental changes.

Fixes PLT-63

github-actions · 2026-02-11T07:21:01Z

The latest Buf updates on your PR. Results from workflow Buf / buf (pull_request).

Build	Format	Lint	Breaking	Updated (UTC)
`✅ passed`	`✅ passed`	`✅ passed`	`✅ passed`	Feb 12, 2026, 5:49 AM

codecov · 2026-02-11T07:28:50Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 48.32%. Comparing base (a0b14c6) to head (e9942ac).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2835   +/-   ##
=======================================
  Coverage   48.31%   48.32%           
=======================================
  Files         671      671           
  Lines       50576    50576           
=======================================
+ Hits        24434    24439    +5     
+ Misses      24001    23998    -3     
+ Partials     2141     2139    -2

Flag	Coverage Δ
sei-cosmos	`48.13% <ø> (+<0.01%)`	⬆️
sei-db	`68.72% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.
see 1 file with indirect coverage changes

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

masih

Two minor blockers and a bunch of suggestions, otherwise LGTM 🚀

the genesis JSON files use inconsistent formatting; e.g. different indentation, extra spaces at the end, etc. Please format all three once in your IDE to at least have consistent formatting across the three.
the substitution logic needs to give a higher priority to explicitly present genesis file. See my comment for a deeper context.

Thank you for knocking this one out so quickly! 🙌

masih · 2026-02-11T11:25:17Z

cmd/seid/cmd/init.go

-					return err
+			var genDoc *types.GenesisDoc
+			var appState json.RawMessage
+			if genesis.IsWellKnown(chainID) {


I think we want the genesis substitution order to be:

genesis.json in config folder, if present.

if not --> is chain ID well known ?

yes: load from embedded genesis.

no: error out with genesis missing

Bonus logic (please feel free to do in a separate PR):

if explicit is present but for a well known chain ID:

Check that it is right ( note that the format is JSON and we need to sanitise formatting before checking hash, or... load as struct and compare types equality)

if good: log in debug that "hey I know my networks -- you don't have to have the genesis file"

if not good: panic out with appropriate messaging. Most likely accidentally edited genesis; invalid for the network in any case.

This would follow the general operational best practice of: "Explicit over implicit", and "Implicit over unspecified".

Does that make sense @mojtaba-esk? WDYT?

Thanks for the detailed comment.

genesis.json in config folder, if present.
This is checked a few lines above so it does not reach to this point when genesis.json exists.

Will apply the 2 as this makes sense.

Let's do the rest (Bonus logic) in a followup.

oh I see what you mean now. Created a followup for it: https://linear.app/seilabs/issue/PLT-116/validate-explicit-genesis-file-for-well-known-chains-in-seid-init

@mojtaba-esk I think we are still missing item No. 1 in my comment right?

We want to check if the genesis is explicitly set first, and only implicitly load for well known chains if it is absent.

masih · 2026-02-11T11:33:03Z

cmd/seid/cmd/init.go

-					return errors.Wrap(err, "Failed to read genesis doc from file")
+					return errors.Wrap(err, "Failed to marshall default genesis state")
+				}
+				genDoc = &types.GenesisDoc{}


Move closer to where it is first used?

we can't move this as it needs to have an empty gendoc in case the file does not exist and after the follwoing if statement, the given ChainID and the default values like appState is assigned to it.

masih · 2026-02-11T11:34:10Z

cmd/seid/cmd/init.go

+					return errors.Wrap(err, "Failed to marshall default genesis state")
+				}
+				genDoc = &types.GenesisDoc{}
+				if _, err := os.Stat(genFile); err != nil {


Also check that genFile is not a directory?

We could do that sure.

masih · 2026-02-11T11:39:52Z

app/genesis/genesis_test.go

+	require.Error(t, err)
+	require.Contains(t, err.Error(), "unknown chain-id")


Suggested change

require.Error(t, err)

require.Contains(t, err.Error(), "unknown chain-id")

require.ErrorContains(t, err, "unknown chain-id")

masih · 2026-02-11T11:42:07Z

app/genesis/genesis_test.go

+var expectedGenesisDigests = map[string]string{
+	"arctic-1":   "a8b60161e345d8afb0d9b0c524b6541e6135f5eda4092fae287d8496e14554d8",
+	"atlantic-2": "3c135db9177a428893353d7889149ca2ed9c075d6846be07af60354022b81318",
+	"pacific-1":  "4304cf1c7f46d153b79f1195b2d334f7f7cf02f26e02a3bb77c544a4987c1432",


These are the hashes of the indented JSON, and technically the same JSON values in different order/indentation is still valid.

But then we will/should never edit those JSON files anyway so hardcoding the hashes of pretty JSON is probably fine 👍

This is a critical point. I will see if I can hash it without considering whitespaces. Like the gendoc itself. if that works, the indents are fine as on some IDEs it is configured as one or two tabs one some it uses spaces... etc. Also the ordering should be fine as long as the gendoc structure is not touched.

masih · 2026-02-11T11:48:31Z

app/genesis/genesis.go

+
+var wellKnownChainIDs = []string{"arctic-1", "atlantic-2", "pacific-1"}
+
+func WellKnownChainIDs() []string {


This is used for testing only -- I would recommend hardcoding known chain IDs in tests instead and getting rid of this function. Alternatively, in tests you can get the list by lsing the chains/*.json files which has an added benfit that the code does include all json files put in there.

masih · 2026-02-11T11:49:18Z

app/genesis/genesis.go

+var wellKnownChainIDs = []string{"arctic-1", "atlantic-2", "pacific-1"}
+
+func WellKnownChainIDs() []string {
+	return append([]string(nil), wellKnownChainIDs...)


If you decide to keep this function, slices.Clone is more idiomatic.

Thanks for the hint

masih · 2026-02-11T11:51:12Z

app/genesis/genesis.go

+//go:embed chains/*.json
+var embeddedGenesisFS embed.FS
+
+var wellKnownChainIDs = []string{"arctic-1", "atlantic-2", "pacific-1"}


An edge-case can occure here where someone adds the JSON but forgets to update this.

We can avoid it by loading the known IDs via package Init, which would get rid of this struct entirely and makes the JSON files in chains directory the sole source of truth.

Good idea. will apply it

app/genesis/genesis.go

mojtaba-esk · 2026-02-11T16:21:15Z

app/genesis/genesis_test.go

+	"pacific-1":  "3a1f5d87df75f4fdb85eaf1b506e080cbcee3a748048e1e80721f68eb2193e43",
+}
+
+func genesisDocDigest(chainID string) ([]byte, error) {


Here the limitation is that we do not order the appState to have a canonical value of it as it would have added extra complexity to the code that seems not be required at this stage.

philipsu522

huge QoL improvement

@masih

## Describe your changes and provide context This PR is a followup for #2835 to handle a case where the user already has their own genesis file and does not want to remove them or overwrite them. Here is the slack comment by @masih that suggested this fix: https://sei-network-io.slack.com/archives/C06A3D4FWE5/p1770833867974669?thread_ts=1770803742.497719&cid=C06A3D4FWE5 ## Testing performed to validate your change --------- Co-authored-by: Mojtaba <mojtaba@celestia.org>

feat: embed genesis for well-known chains

ce64296

mojtaba-esk self-assigned this Feb 11, 2026

mojtaba-esk added the non-app-hash-breaking label Feb 11, 2026

mojtaba-esk requested review from a team and masih February 11, 2026 07:32

Merge branch 'main' into mojtaba/embed-genesis-for-well-known-chains

94ce211

mojtaba-esk requested review from philipsu522 and removed request for a team February 11, 2026 09:55

masih reviewed Feb 11, 2026

View reviewed changes

sei-will self-requested a review February 11, 2026 12:30

sei-will approved these changes Feb 11, 2026

View reviewed changes

Mojtaba and others added 4 commits February 11, 2026 16:08

fix: apply review feedback

06602ea

Merge branch 'main' into mojtaba/embed-genesis-for-well-known-chains

85eb6be

fix: import after mergin the main

f0d28b5

chore: cleanup

4949749

mojtaba-esk commented Feb 11, 2026

View reviewed changes

mojtaba-esk requested a review from masih February 11, 2026 16:41

philipsu522 approved these changes Feb 11, 2026

View reviewed changes

yzang2019 approved these changes Feb 11, 2026

View reviewed changes

Merge branch 'main' into mojtaba/embed-genesis-for-well-known-chains

e9942ac

masih approved these changes Feb 12, 2026

View reviewed changes

masih merged commit 9b10408 into main Feb 12, 2026
40 checks passed

masih deleted the mojtaba/embed-genesis-for-well-known-chains branch February 12, 2026 08:59

mojtaba-esk mentioned this pull request Feb 12, 2026

fix: respect existing genesis file #2868

Merged

		require.Error(t, err)
		require.Contains(t, err.Error(), "unknown chain-id")

	require.Error(t, err)
	require.Contains(t, err.Error(), "unknown chain-id")
	require.ErrorContains(t, err, "unknown chain-id")


		var wellKnownChainIDs = []string{"arctic-1", "atlantic-2", "pacific-1"}

		func WellKnownChainIDs() []string {

feat: embed genesis for well-known chains #2835

feat: embed genesis for well-known chains #2835

Conversation

mojtaba-esk commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Describe your changes and provide context

Uh oh!

github-actions bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

masih left a comment

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!

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!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

philipsu522 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

mojtaba-esk commented Feb 11, 2026 •

edited

Loading

github-actions bot commented Feb 11, 2026 •

edited

Loading

codecov bot commented Feb 11, 2026 •

edited

Loading