Use cairo-lang-primitive-token from registry not path in unreleased cairo crates

[maciektr]

No description provided.

[reviewable-StarkWare]

This change is 

[mkaput]

Reviewed 3 of 3 files at r1, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @maciektr and @orizi)

---

a discussion (no related file):
what's the reasoning behind this change?

[maciektr]

Previously, mkaput (Marek Kaput) wrote…

what's the reasoning behind this change?

Without this change, when you depend on the parser through git dependency, transitive dependency to cairo-lang-primitive-token will be translated into local git dependency, not the remote registry one. In that case, Rustc is unable to determine that both of this crates (local git and registry remote) are identical and fails with an error:

   Compiling some v1.0.0 ([..]Scarb.toml)
error[E0271]: expected `SyntaxNodeWithDbIterator<'_, SimpleParserDatabase>` to be an iterator that yields `PrimitiveToken`, but it yields `PrimitiveToken`
   --> src/lib.rs:45:16
    |
45  |     let tokens = quote! {
    |  ________________^
46  | |     #syntax_node_with_db
47  | |     trait Circle {
48  | |       fn print() -> ();
...   |
54  | |     }
55  | |   };
    | |___^ expected `PrimitiveToken`, found a different `PrimitiveToken`
    |
    = note: `PrimitiveToken` and `PrimitiveToken` have similar names, but are actually distinct types

[mkaput]

Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @maciektr and @orizi)

---

a discussion (no related file):

Previously, maciektr (maciektr) wrote…

Without this change, when you depend on the parser through git dependency, transitive dependency to cairo-lang-primitive-token will be translated into local git dependency, not the remote registry one. In that case, Rustc is unable to determine that both of this crates (local git and registry remote) are identical and fails with an error:

   Compiling some v1.0.0 ([..]Scarb.toml)
error[E0271]: expected `SyntaxNodeWithDbIterator<'_, SimpleParserDatabase>` to be an iterator that yields `PrimitiveToken`, but it yields `PrimitiveToken`
   --> src/lib.rs:45:16
    |
45  |     let tokens = quote! {
    |  ________________^
46  | |     #syntax_node_with_db
47  | |     trait Circle {
48  | |       fn print() -> ();
...   |
54  | |     }
55  | |   };
    | |___^ expected `PrimitiveToken`, found a different `PrimitiveToken`
    |
    = note: `PrimitiveToken` and `PrimitiveToken` have similar names, but are actually distinct types

Can you provide this explanation as a comment here? It's surprising that local crate is not used via local path. 🙏

[maciektr]

Reviewable status: 1 of 3 files reviewed, 1 unresolved discussion (waiting on @mkaput and @orizi)

---

a discussion (no related file):

Previously, mkaput (Marek Kaput) wrote…

Can you provide this explanation as a comment here? It's surprising that local crate is not used via local path. 🙏

Done.

[orizi]

Reviewed all commit messages.
Reviewable status: 1 of 3 files reviewed, 1 unresolved discussion (waiting on @mkaput)

---

a discussion (no related file):

Previously, maciektr (maciektr) wrote…

Done.

this sounds like we should just move this crate out of the crate tree here completely - and ONLY depend on the registry one.

[mkaput]

Reviewed 2 of 2 files at r2, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @maciektr)

---

a discussion (no related file):

Previously, orizi wrote…

this sounds like we should just move this crate out of the crate tree here completely - and ONLY depend on the registry one.

Makes sense, although who should host it? I think the repo should be in starkware-libs, to minimize dependencies on SWM repos here.

[maciektr]

Previously, mkaput (Marek Kaput) wrote…

Makes sense, although who should host it? I think the repo should be in starkware-libs, to minimize dependencies on SWM repos here.

@orizi do you have an opinion here? 🤔

[orizi]

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

---

a discussion (no related file):

Previously, maciektr (maciektr) wrote…

@orizi do you have an opinion here? 🤔

we can have it in the repo - just not as a part of the workspace.

[maciektr]

Previously, orizi wrote…

we can have it in the repo - just not as a part of the workspace.

Removed from workspace

[orizi]

Reviewed 2 of 2 files at r2, 3 of 3 files at r3, all commit messages.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @maciektr)

---

crates/cairo-lang-parser/Cargo.toml line 14 at r3 (raw file):

# Registry dep is used instead of local path, so the same exact source can be used by crates
# depending on this crate, from both registry and git.
cairo-lang-primitive-token = "1"

move dep and doc to workspace.

Suggestion:

cairo-lang-primitive-token.worksspace = true

---

crates/cairo-lang-syntax/Cargo.toml line 15 at r3 (raw file):

# Registry dep is used instead of local path, so the same exact source can be used by crates
# depending on this crate, from both registry and git.
cairo-lang-primitive-token = "1"

move dep and doc to workspace.

Suggestion:

cairo-lang-primitive-token.worksspace = true

[maciektr]

Reviewable status: 2 of 5 files reviewed, 3 unresolved discussions (waiting on @mkaput and @orizi)

---

crates/cairo-lang-syntax/Cargo.toml line 15 at r3 (raw file):

Previously, orizi wrote…

move dep and doc to workspace.

Done.

[maciektr]

Reviewable status: 2 of 5 files reviewed, 3 unresolved discussions (waiting on @mkaput and @orizi)

---

crates/cairo-lang-parser/Cargo.toml line 14 at r3 (raw file):

Previously, orizi wrote…

move dep and doc to workspace.

Done.

[orizi]

Reviewed 3 of 3 files at r4, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @maciektr)

[orizi]

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