feat(blockifier): store block_context as Arc in TransactionContext

[liorgold2]

No description provided.

[reviewable-StarkWare]

This change is 

[github-actions]

Artifacts upload workflows:

• Started at 2025-02-14T15:04:36Z
• Started at 2025-02-14T15:14:23Z

[elintul]

Reviewed 5 of 5 files at r1, all commit messages.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @liorgold2)

---

-- commits line 2 at r1:
What's the context? What's the feature your working and your plan?

Code quote:

store block_context as Arc in TransactionContext

---

crates/blockifier/src/context.rs line 27 at r1 (raw file):

#[derive(Clone, Debug)]
pub struct TransactionContext {

Consider implementing new c-tor, to encapsulate this.

Code quote:

TransactionContext

[liorgold2]

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @elintul)

---

-- commits line 2 at r1:

Previously, elintul (Elin) wrote…

What's the context? What's the feature your working and your plan?

Adding a meta-tx syscall. This will require creating a TransactionContext with a different tx_info, but the same block_info. So I'm changing it to Arc to avoid cloning it.

[liorgold2]

Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @elintul)

---

crates/blockifier/src/context.rs line 27 at r1 (raw file):

Previously, elintul (Elin) wrote…

Consider implementing new c-tor, to encapsulate this.

Noted. I think BlockContext should be Arc in more places, so a constructor from BlockContext that wraps it in Arc is problematic.

[elintul]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @liorgold2)

---

crates/blockifier/src/context.rs line 27 at r1 (raw file):

Previously, liorgold2 wrote…

Noted. I think BlockContext should be Arc in more places, so a constructor from BlockContext that wraps it in Arc is problematic.

Not sure I understood.

[liorgold2]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @liorgold2)

---

crates/blockifier/src/context.rs line 27 at r1 (raw file):

Previously, elintul (Elin) wrote…

Not sure I understood.

The question is what the new constructor will get.

• If it gets Arc<BlockContext> then it doesn't encapsulates it.
• If it gets BlockContext then it needs to create an Arc, but I don't think this is what I want to encourage - if you have an Arc<BlockContext> and you want to create TransactionContext, then you won't be able to pass through this constructor.

[elintul]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @liorgold2)

---

crates/blockifier/src/context.rs line 27 at r1 (raw file):

Previously, liorgold2 wrote…

The question is what the new constructor will get.

• If it gets Arc<BlockContext> then it doesn't encapsulates it.
• If it gets BlockContext then it needs to create an Arc, but I don't think this is what I want to encourage - if you have an Arc<BlockContext> and you want to create TransactionContext, then you won't be able to pass through this constructor.

Okay, I see. Keep in mind as this feature goes, that if the second one is the more common case, new c-tor can be considered.

[elintul]

Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @liorgold2)