Skip to content
  1. Mar 01, 2024
  3. Feb 29, 2024
  5. Feb 28, 2024
    • Oliver Tale-Yazdi's avatar
      Multi-Block-Migrations, `poll` hook and new System callbacks (#1781) · eefd5fe4
      Oliver Tale-Yazdi authored 
      This MR is the merge of
https://github.com/paritytech/substrate/pull/14414 and
https://github.com/paritytech/substrate/pull/14275. It implements
[RFC#13](https://github.com/polkadot-fellows/RFCs/pull/13), closes
https://github.com/paritytech/polkadot-sdk/issues/198.

----- 

This Merge request introduces three major topicals:

1. Multi-Block-Migrations
1. New pallet `poll` hook for periodic service work
1. Replacement hooks for `on_initialize` and `on_finalize` in cases
where `poll` cannot be used

and some more general changes to FRAME.  
The changes for each topical span over multiple crates. They are listed
in topical order below.

# 1.) Multi-Block-Migrations

Multi-Block-Migrations are facilitated by creating `pallet_migrations`
and configuring `System::Config::MultiBlockMigrator` to point to it.
Executive picks this up and triggers one step of the migrations pallet
per block.
The chain is in lockdown mode for as long as an MBM is ongoing.
Executive does this by polling `MultiBlockMigrator::ongoing` and not
allowing any transaction in a block, if true.

A MBM is defined through trait `SteppedMigration`. A condensed version
looks like this:
```rust
/// A migration that can proceed in multiple steps.
pub trait SteppedMigration {
	type Cursor: FullCodec + MaxEncodedLen;
	type Identifier: FullCodec + MaxEncodedLen;

	fn id() -> Self::Identifier;

	fn max_steps() -> Option<u32>;

	fn step(
		cursor: Option<Self::Cursor>,
		meter: &mut WeightMeter,
	) -> Result<Option<Self::Cursor>, SteppedMigrationError>;
}
```

`pallet_migrations` can be configured with an aggregated tuple of these
migrations. It then starts to migrate them one-by-one on the next
runtime upgrade.
Two things are important here:
- 1. Doing another runtime upgrade while MBMs are ongoing is not a good
idea and can lead to messed up state.
- 2. **Pallet Migrations MUST BE CONFIGURED IN `System::Config`,
otherwise it is not used.**

The pallet supports an `UpgradeStatusHandler` that can be used to notify
external logic of upgrade start/finish (for example to pause XCM
dispatch).

Error recovery is very limited in the case that a migration errors or
times out (exceeds its `max_steps`). Currently the runtime dev can
decide in `FailedMigrationHandler::failed` how to handle this. One
follow-up would be to pair this with the `SafeMode` pallet and enact
safe mode when an upgrade fails, to allow governance to rescue the
chain. This is currently not possible, since governance is not
`Mandatory`.

## Runtime API

- `Core`: `initialize_block` now returns `ExtrinsicInclusionMode` to
inform the Block Author whether they can push transactions.

### Integration

Add it to your runtime implementation of `Core` and `BlockBuilder`:
```patch
diff --git a/runtime/src/lib.rs b/runtime/src/lib.rs
@@ impl_runtime_apis! {
	impl sp_block_builder::Core<Block> for Runtime {
-		fn initialize_block(header: &<Block as BlockT>::Header) {
+		fn initialize_block(header: &<Block as BlockT>::Header) -> RuntimeExecutiveMode {
			Executive::initialize_block(header)
		}

		...
	}
```

# 2.) `poll` hook

A new pallet hook is introduced: `poll`. `Poll` is intended to replace
mostly all usage of `on_initialize`.
The reason for this is that any code that can be called from
`on_initialize` cannot be migrated through an MBM. Currently there is no
way to statically check this; the implication is to use `on_initialize`
as rarely as possible.
Failing to do so can result in broken storage invariants.

The implementation of the poll hook depends on the `Runtime API` changes
that are explained above.

# 3.) Hard-Deadline callbacks

Three new callbacks are introduced and configured on `System::Config`:
`PreInherents`, `PostInherents` and `PostTransactions`.
These hooks are meant as replacement for `on_initialize` and
`on_finalize` in cases where the code that runs cannot be moved to
`poll`.
The reason for this is to make the usage of HD-code (hard deadline) more
explicit - again to prevent broken invariants by MBMs.

# 4.) FRAME (general changes)

## `frame_system` pallet

A new memorize storage item `InherentsApplied` is added. It is used by
executive to track whether inherents have already been applied.
Executive and can then execute the MBMs directly between inherents and
transactions.

The `Config` gets five new items:
- `SingleBlockMigrations` this is the new way of configuring migrations
that run in a single block. Previously they were defined as last generic
argument of `Executive`. This shift is brings all central configuration
about migrations closer into view of the developer (migrations that are
configured in `Executive` will still work for now but is deprecated).
- `MultiBlockMigrator` this can be configured to an engine that drives
MBMs. One example would be the `pallet_migrations`. Note that this is
only the engine; the exact MBMs are injected into the engine.
- `PreInherents` a callback that executes after `on_initialize` but
before inherents.
- `PostInherents` a callback that executes after all inherents ran
(including MBMs and `poll`).
- `PostTransactions` in symmetry to `PreInherents`, this one is called
before `on_finalize` but after all transactions.

A sane default is to set all of these to `()`. Example diff suitable for
any chain:
```patch
@@ impl frame_system::Config for Test {
 	type MaxConsumers = ConstU32<16>;
+	type SingleBlockMigrations = ();
+	type MultiBlockMigrator = ();
+	type PreInherents = ();
+	type PostInherents = ();
+	type PostTransactions = ();
 }
```

An overview of how the block execution now looks like is here. The same
graph is also in the rust doc.

<details><summary>Block Execution Flow</summary>
<p>

![Screenshot 2023-12-04 at 19 11
29](https://github.com/paritytech/polkadot-sdk/assets/10380170/e88a80c4-ef11-4faa-8df5-8b33a724c054)

</p>
</details> 

## Inherent Order

Moved to https://github.com/paritytech/polkadot-sdk/pull/2154



---------------


## TODO

- [ ] Check that `try-runtime` still works
- [ ] Ensure backwards compatibility with old Runtime APIs
- [x] Consume weight correctly
- [x] Cleanup

---------

Signed-off-by: default avatarOliver Tale-Yazdi <[email protected]>
Co-authored-by: default avatarLiam Aharon <[email protected]>
Co-authored-by: default avatarJuan Girini <[email protected]>
Co-authored-by: command-bot <>
Co-authored-by: default avatarFrancisco Aguirre <[email protected]>
Co-authored-by: default avatarGavin Wood <[email protected]>
Co-authored-by: default avatarBastian Köcher <[email protected]>
      eefd5fe4
  7. Feb 26, 2024
  9. Feb 20, 2024
    • Oliver Tale-Yazdi's avatar
      Lift dependencies to the workspace (Part 2/x) (#3366) · e89d0fca
      Oliver Tale-Yazdi authored 
      

Lifting some more dependencies to the workspace. Just using the
most-often updated ones for now.
It can be reproduced locally.

```sh
# First you can check if there would be semver incompatible bumps (looks good in this case):
$ zepter transpose dependency lift-to-workspace --ignore-errors syn quote thiserror "regex:^serde.*"

# Then apply the changes:
$ zepter transpose dependency lift-to-workspace --version-resolver=highest syn quote thiserror "regex:^serde.*" --fix

# And format the changes:
$ taplo format --config .config/taplo.toml
```

---------

Signed-off-by: default avatarOliver Tale-Yazdi <[email protected]>
      e89d0fca
  11. Feb 19, 2024
  13. Feb 18, 2024
  15. Feb 16, 2024
  17. Feb 13, 2024
  19. Feb 12, 2024
  21. Feb 09, 2024
  23. Feb 06, 2024
  25. Jan 31, 2024
    • Oliver Tale-Yazdi's avatar
      [FRAME] Make `core-fellowship` ans `salary` work for swapped members (#3156) · 07e55006
      Oliver Tale-Yazdi authored 
      Fixup for https://github.com/paritytech/polkadot-sdk/pull/2587 to make
the `core-fellowship` crate work with swapped members.

Adds a `MemberSwappedHandler` to the `ranked-collective` pallet that are
implemented by `core-fellowship+salary`.
There is are exhaustive tests
[here](https://github.com/paritytech/polkadot-sdk/blob/72aa7ac17a0e5b16faab5d2992aa2db2e01b05d0/substrate/frame/core-fellowship/src/tests/integration.rs#L338)
and
[here](https://github.com/paritytech/polkadot-sdk/blob/ab3cdb05a5ebc1ff841f8dda67edef0ea40bbba5/substrate/frame/salary/src/tests/integration.rs#L224

)
to check that adding member `1` is equivalent to adding member `0` and
then swapping.

---------

Signed-off-by: default avatarOliver Tale-Yazdi <[email protected]>
      07e55006
    • Branislav Kontur's avatar
      [frame] `#[pallet::composite_enum]` improved variant count handling + removed... · bb8ddc46
      Branislav Kontur authored 
      [frame] `#[pallet::composite_enum]` improved variant count handling + removed `pallet_balances`'s `MaxHolds` config (#2657)

I started this investigation/issue based on @liamaharon question
[here](https://github.com/paritytech/polkadot-sdk/pull/1801#discussion_r1410452499).

## Problem

The `pallet_balances` integrity test should correctly detect that the
runtime has correct distinct `HoldReasons` variant count. I assume the
same situation exists for RuntimeFreezeReason.

It is not a critical problem, if we set `MaxHolds` with a sufficiently
large value, everything should be ok. However, in this case, the
integrity_test check becomes less useful.

**Situation for "any" runtime:**
- `HoldReason` enums from different pallets:
```rust
        /// from pallet_nis
        #[pallet::composite_enum]
	pub enum HoldReason {
		NftReceipt,
	}

        /// from pallet_preimage
        #[pallet::composite_enum]
	pub enum HoldReason {
		Preimage,
	}

        // from pallet_state-trie-migration
        #[pallet::composite_enum]
	pub enum HoldReason {
		SlashForContinueMigrate,
		SlashForMigrateCustomTop,
		SlashForMigrateCustomChild,
	}
```

- generated `RuntimeHoldReason` enum looks like:
```rust
pub enum RuntimeHoldReason {

    #[codec(index = 32u8)]
    Preimage(pallet_preimage::HoldReason),

    #[codec(index = 38u8)]
    Nis(pallet_nis::HoldReason),

    #[codec(index = 42u8)]
    StateTrieMigration(pallet_state_trie_migration::HoldReason),
}
```

- composite enum `RuntimeHoldReason` variant count is detected as `3`
- we set `type MaxHolds = ConstU32<3>`
- `pallet_balances::integrity_test` is ok with `3`(at least 3)

However, the real problem can occur in a live runtime where some
functionality might stop working. This is due to a total of 5 distinct
hold reasons (for pallets with multi-instance support, it is even more),
and not all of them can be used because of an incorrect `MaxHolds`,
which is deemed acceptable according to the `integrity_test`:
  ```
  // pseudo-code - if we try to call all of these:

T::Currency::hold(&pallet_nis::HoldReason::NftReceipt.into(),
&nft_owner, deposit)?;
T::Currency::hold(&pallet_preimage::HoldReason::Preimage.into(),
&nft_owner, deposit)?;

T::Currency::hold(&pallet_state_trie_migration::HoldReason::SlashForContinueMigrate.into(),
&nft_owner, deposit)?;

  // With `type MaxHolds = ConstU32<3>` these two will fail

T::Currency::hold(&pallet_state_trie_migration::HoldReason::SlashForMigrateCustomTop.into(),
&nft_owner, deposit)?;

T::Currency::hold(&pallet_state_trie_migration::HoldReason::SlashForMigrateCustomChild.into(),
&nft_owner, deposit)?;
  ```  


## Solutions

A macro `#[pallet::*]` expansion is extended of `VariantCount`
implementation for the `#[pallet::composite_enum]` enum type. This
expansion generates the `VariantCount` implementation for pallets'
`HoldReason`, `FreezeReason`, `LockId`, and `SlashReason`. Enum variants
must be plain enum values without fields to ensure a deterministic
count.

The composite runtime enum, `RuntimeHoldReason` and
`RuntimeFreezeReason`, now sets `VariantCount::VARIANT_COUNT` as the sum
of pallets' enum `VariantCount::VARIANT_COUNT`:
```rust
#[frame_support::pallet(dev_mode)]
mod module_single_instance {

	#[pallet::composite_enum]
	pub enum HoldReason {
		ModuleSingleInstanceReason1,
		ModuleSingleInstanceReason2,
	}
...
}

#[frame_support::pallet(dev_mode)]
mod module_multi_instance {

	#[pallet::composite_enum]
	pub enum HoldReason<I: 'static = ()> {
		ModuleMultiInstanceReason1,
		ModuleMultiInstanceReason2,
		ModuleMultiInstanceReason3,
	}
...
}


impl self::sp_api_hidden_includes_construct_runtime::hidden_include::traits::VariantCount
    for RuntimeHoldReason
{
    const VARIANT_COUNT: u32 = 0
        + module_single_instance::HoldReason::VARIANT_COUNT
        + module_multi_instance::HoldReason::<module_multi_instance::Instance1>::VARIANT_COUNT
        + module_multi_instance::HoldReason::<module_multi_instance::Instance2>::VARIANT_COUNT
        + module_multi_instance::HoldReason::<module_multi_instance::Instance3>::VARIANT_COUNT;
}
```

In addition, `MaxHolds` is removed (as suggested
[here](https://github.com/paritytech/polkadot-sdk/pull/2657#discussion_r1443324573))
from `pallet_balances`, and its `Holds` are now bounded to
`RuntimeHoldReason::VARIANT_COUNT`. Therefore, there is no need to let
the runtime specify `MaxHolds`.


## For reviewers

Relevant changes can be found here:
- `substrate/frame/support/procedural/src/lib.rs` 
-  `substrate/frame/support/procedural/src/pallet/parse/composite.rs`
-  `substrate/frame/support/procedural/src/pallet/expand/composite.rs`
-
`substrate/frame/support/procedural/src/construct_runtime/expand/composite_helper.rs`
-
`substrate/frame/support/procedural/src/construct_runtime/expand/hold_reason.rs`
-
`substrate/frame/support/procedural/src/construct_runtime/expand/freeze_reason.rs`
- `substrate/frame/support/src/traits/misc.rs`

And the rest of the files is just about removed `MaxHolds` from
`pallet_balances`

## Next steps

Do the same for `MaxFreezes`
https://github.com/paritytech/polkadot-sdk/issues/2997

.

---------

Co-authored-by: command-bot <>
Co-authored-by: default avatarBastian Köcher <[email protected]>
Co-authored-by: default avatarDónal Murray <[email protected]>
Co-authored-by: default avatargupnik <[email protected]>
      bb8ddc46
  27. Jan 30, 2024
  29. Jan 26, 2024
  31. Jan 24, 2024
  33. Jan 23, 2024
  35. Jan 22, 2024
  37. Jan 18, 2024
  39. Jan 16, 2024
    • Francisco Aguirre's avatar
      XCMv4 (#1230) · 8428f678
      Francisco Aguirre authored 
      

# Note for reviewer

Most changes are just syntax changes necessary for the new version.
Most important files should be the ones under the `xcm` folder.

# Description 

Added XCMv4.

## Removed `Multi` prefix
The following types have been renamed:
- MultiLocation -> Location
- MultiAsset -> Asset
- MultiAssets -> Assets
- InteriorMultiLocation -> InteriorLocation
- MultiAssetFilter -> AssetFilter
- VersionedMultiAsset -> VersionedAsset
- WildMultiAsset -> WildAsset
- VersionedMultiLocation -> VersionedLocation

In order to fix a name conflict, the `Assets` in `xcm-executor` were
renamed to `HoldingAssets`, as they represent assets in holding.

## Removed `Abstract` asset id

It was not being used anywhere and this simplifies the code.

Now assets are just constructed as follows:

```rust
let asset: Asset = (AssetId(Location::new(1, Here)), 100u128).into();
```

No need for specifying `Concrete` anymore.

## Outcome is now a named fields struct

Instead of

```rust
pub enum Outcome {
  Complete(Weight),
  Incomplete(Weight, Error),
  Error(Error),
}
```

we now have

```rust
pub enum Outcome {
  Complete { used: Weight },
  Incomplete { used: Weight, error: Error },
  Error { error: Error },
}
```

## Added Reanchorable trait

Now both locations and assets implement this trait, making it easier to
reanchor both.

## New syntax for building locations and junctions

Now junctions are built using the following methods:

```rust
let location = Location {
    parents: 1,
    interior: [Parachain(1000), PalletInstance(50), GeneralIndex(1984)].into()
};
```

or

```rust
let location = Location::new(1, [Parachain(1000), PalletInstance(50), GeneralIndex(1984)]);
```

And they are matched like so:

```rust
match location.unpack() {
  (1, [Parachain(id)]) => ...
  (0, Here) => ...,
  (1, [_]) => ...,
}
```

This syntax is mandatory in v4, and has been also implemented for v2 and
v3 for easier migration.

This was needed to make all sizes smaller.

# TODO
- [x] Scaffold v4
- [x] Port github.com/paritytech/polkadot/pull/7236
- [x] Remove `Multi` prefix
- [x] Remove `Abstract` asset id

---------

Co-authored-by: command-bot <>
Co-authored-by: default avatarKeith Yeung <[email protected]>
      8428f678
    • Dónal Murray's avatar
      Bump rococo relay and coretime-rococo to 1.6 (#2913) · a42a47f8
      Dónal Murray authored 
      Co-authored-by: command-bot <>
      a42a47f8
  41. Jan 10, 2024
    • joe petrowski's avatar
      Unique Usernames in Identity Pallet (#2651) · d1f678c0
      joe petrowski authored 
      

This PR allows _username authorities_ to issue unique usernames that
correspond with an account. It also provides two-way lookup, that is
from `AccountId` to a single, "primary" `Username` (alongside
`Registration`) and multiple unique `Username`s to an `AccountId`.

Key features:

- Username Authorities added (and removed) via privileged origin.
- Authorities have a `suffix` and an `allocation`. They can grant up to
`allocation` usernames. Their `suffix` will be appended to the usernames
that they issue. A suffix may be up to 7 characters long.
- Users can ask an authority to grant them a username. This will take
the form `myusername.suffix`. The entire name (including suffix) must be
less than or equal to 32 alphanumeric characters.
- Users can approve a username for themselves in one of two ways (that
is, authorities cannot grant them arbitrarily):
- Pre-sign the entire username (including suffix) with a secret key that
corresponds to their `AccountId` (for keyed accounts, obviously); or
- Accept the username after it has been granted by an authority (it will
be queued until accepted) (for non-keyed accounts like pure proxies or
multisigs).
- The system does not require any funds or deposits. Users without an
identity will be given a default one (presumably all fields set to
`None`). If they update this info, they will need to place the normal
storage deposit.
- If a user does not have any username, their first one will be set as
`Primary`, and their `AccountId` will map to that one. If they get
subsequent usernames, they can choose which one to be their primary via
`set_primary_username`.
- There are some state cleanup functions to remove expired usernames
that have not been accepted and dangling usernames whose owners have
called `clear_identity`.

TODO:

- [x] Add migration to runtimes
- [x] Probably do off-chain migration into People Chain genesis
- [x] Address a few TODO questions in code (please review)

---------

Co-authored-by: default avatarLiam Aharon <[email protected]>
Co-authored-by: default avatarGonçalo Pestana <[email protected]>
Co-authored-by: default avatarOliver Tale-Yazdi <[email protected]>
Co-authored-by: default avatarDónal Murray <[email protected]>
      d1f678c0
  43. Jan 07, 2024
  45. Jan 06, 2024
  47. Jan 04, 2024
  49. Dec 27, 2023
  51. Dec 22, 2023
  53. Dec 21, 2023
  55. Dec 20, 2023
    • joe petrowski's avatar
      Add Authorize Upgrade Pattern to Frame System (#2682) · 280aa0b5
      joe petrowski authored 
      Adds the `authorize_upgrade` -> `enact_authorized_upgrade` pattern to
`frame-system`. This will be useful for upgrading bridged chains that
are under the governance of Polkadot without passing entire runtime Wasm
blobs over a bridge.

Notes:

- Changed `enact_authorized_upgrade` to `apply_authorized_upgrade`.
Personal opinion, "apply" more accurately expresses what it's doing. Can
change back if outvoted.
- Remove `check_version` in favor of two extrinsics, so as to make
_checked_ the default.
- Left calls in `parachain-system` and marked as deprecated to prevent
breaking the API. They just call into the `frame-system` functions.
- Updated `frame-system` benchmarks to v2 syntax.

---------

Co-authored-by: command-bot <>
      280aa0b5
  57. Dec 14, 2023
    • Svyatoslav Nikolsky's avatar
      Add Rococo People <> Rococo Bulletin bridge support to Rococo Bridge Hub (#2540) · 097308e3
      Svyatoslav Nikolsky authored 
      This PR adds [Rococo
People](https://github.com/paritytech/polkadot-sdk/pull/2281) <> [Rococo
Bulletin](https://github.com/zdave-parity/polkadot-bulletin-chain) to
the Rococo Bridge Hub code. There's a couple of things left to do here:
- [x] add remaining tests - it'd need some refactoring in the
`bridge-hub-test-utils` - will do in a separate PR;
- [x] actually run benchmarks for new messaging pallet (do we have bot
nowadays?).

The reason why I'm opening it before this ^^^ is ready, is that I'd like
to hear others opinion on how to deal with hacks with that bridge.
Initially I was assuming that Rococo Bulletin will be the 1:1 copy of
the Polkadot Bulletin (to avoid maintaining multiple
runtimes/releases/...), so you can see many `PolkadotBulletin` mentions
in this PR, even though we are going to bridge with the parallel chain
(`RococoBulletin`). That's because e.g. pallet names from
`construct_runtime` are affecting runtime storage keys and bridges are
using runtime storage proofs => it is important to use names that the
Bulletin chain expects.

But in the end, this hack won't work - we can't use Polkadot Bulletin
runtime to bridge with Rococo Bridge Hub, because Polkadot Bulletin
expects Polkadot Bridge hub to use `1002` parachain id and Rococo Bridge
Hub seats on the `1013`. This also affects storage keys using in
bridging, so I had to add the [`rococo`
feature](https://github.com/svyatonik/polkadot-bulletin-chain/blob/add-bridge-pallets/runtime/Cargo.toml#L198)
to the Bulletin chain. So now we can actually alter its runtime and
adapt it for Rococo.

So the question here is - what's better for us here
- to leave everything as is (seems hacky and non-trivial);
- change Bulletin chain runtime when `rococo` feature is used - e.g. use
proper names there (`WithPolkadotGrandpa` -> `WithRococoGrandpa`, ...)
- add another set of pallets to the Bulletin chain runtime to bridge
with Rococo and never use them in production. Similar to hack that we
had in Rococo/Wococo

cc @acatangiu @bkontur @serban300 

also cc @joepetrowski

 as the main "client" of this bridge

---

A couple words on how this bridge is different from the Rococo <>
Westend bridge:
- it is a bridge with a chain that uses GRANDPA finality, not the
parachain finality (hence the tests needs to be changed);
- it is a fee-free bridge. So
`AllowExplicitUnpaidExecutionFrom<Equals<SiblingPeople>>` + we are not
paying any rewards to relayers (apart from compensating transaction
costs).

---------

Signed-off-by: default avatardependabot[bot] <[email protected]>
Signed-off-by: default avatarAndrei Sandu <[email protected]>
Co-authored-by: default avatarAdrian Catangiu <[email protected]>
Co-authored-by: default avatardependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: default avatarAndrei Sandu <[email protected]>
Co-authored-by: default avatarEgor_P <[email protected]>
Co-authored-by: command-bot <>
      097308e3
    • Francisco Aguirre's avatar
      Add FungibleAdapter (#2684) · 10a91f82
      Francisco Aguirre authored 
      In the move from the old `Currency` traits to the new `fungible/s`
family of traits, we already had the `FungiblesAdapter` and
`NonFungiblesAdapter` for multiple fungible and non fungible assets
respectively. However, for handling only one fungible asset, we were
missing a `FungibleAdapter`, and so used the old `CurrencyAdapter`
instead. This PR aims to fill in that gap, and provide the new adapter
for more updated examples.

I marked the old `CurrencyAdapter` as deprecated as part of this PR, and
I'll change it to the new `FungibleAdapter` in a following PR.
The two stages are separated so as to not bloat this PR with some name
fixes in tests.

---------

Co-authored-by: command-bot <>
      10a91f82
    • Egor_P's avatar
      [Backport] txn version bump from 1.5.0 (#2709) · 3e4e8c0b
      Egor_P authored 
      This PR backports `transaction_version` bump from `1.5.0` release back
to `master`
      3e4e8c0b