1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
[]
= "polyc-query-model"
# Published: polyc-query depends on it, and polyc-query publishes. A pure
# leaf with no in-workspace dependencies of its own, so this is a clean
# closure addition -- same argument as polyc-projection above.
= true
= true
= "DataFusion-free semantic request and result vocabulary shared by Query clients and the Query service."
= true
= true
= true
= true
= true
[]
= "component"
= true
= "shared"
[]
# Published library crate (crates.io, polyc-* namespace) with no `[[bin]]`
# dist should build archives/installers for. It carries no `publish = false`
# (the crates.io dependency closure needs it publishable), so without this
# explicit opt-out dist still tries to plan a Windows MSI for it, which then
# demands WiX GUIDs this crate has no reason to carry. `dist = false` opts it
# out explicitly instead — same fix as `crates/control-plane`.
= false
[]
= "src/lib.rs"
[]
# This crate is the query protocol's own vocabulary: no state plane, no
# engine, no transport, no columnar format. A client that links it acquires
# none of those through it.
= true
# The conversation-grant token codec (`src/grant.rs`, POLY-320): the signed
# claims a grant carries, and the one function that mints one. Moved here
# from `polyc-query`'s own `authority` module (which re-exports every item
# unchanged) so a caller holding only a local turn-read signing key — like
# `polychrome query verify`, dialing the standalone Query plane directly
# from native mode — can mint a grant without depending on `polyc-query`,
# which links `DataFusion`, Arrow, and Query's execution engine.
# `polyc-crypto` is itself a foundation Component (no state plane, no
# engine), so this is still an inward-only edge.
= { = "=2026.9.6", = "../crypto" }
= { = true, = ["derive"] }
= { = true }
= { = true }
[]
= true