Repository navigation
Add HiroPort (0x000C): HIRO's ogit/_type wire names onto the automation concepts - #333
Conversation
…on concepts HIRO puts the type of every node on the wire as a string such as "ogit/MARS/Machine". HiroPort maps those strings onto the nine 0x0CXX automation concepts that already exist, so a HIRO node dispatches on a u16 (docs/HIRO-IN-CLASSES.md §3.1). App prefix 0x000C is the next free one after HubSpo's 0x000B; 0x0006 stays held for q2. - ports.rs: HiroPort and HIRO_ALIASES, nine exact wire names. Nothing is minted. ogit/Automation/MAID has no alias on purpose: MAID has no concept and no workload needs one. - Tests: the port composes 0x0C04_000C for a machine; wire names are exact (no case folding, no OGIT CURIE spelling); MAID does not resolve; the port covers exactly the automation domain, so a tenth automation concept fails the test until it gets a wire name. Removing the Trigger alias turns both tests red. HiroPort joins the render-prefix collision list and the 0x1000 reservation list. - APP-CLASS-CODEBOOK-LAYOUT §2 row 0x000C; DISCOVERY-MAP §2.13 D-HIRO-PORT with the decision that Bardioc-rs meets spear through ports. ogar-vocab: 174 unit tests and 6 doc tests pass. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VysoWJ6vsyg3wEGc5v7T5v
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Warning Review limit reachedYour organization has reached its usage spending cap. Adjust your spending cap in the billing tab. Next included review available in 32 minutes. View limit details
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
📝 Walkthrough
Merge Risk: ⚪ Minimal · up to No actionable issue is identified; the change is mergeable after normal checks. Pre-merge checks |
|
cargo fmt --all --check failed on #333: the MARSNodeTemplate and ActionApplicability alias tuples exceed the line width. Formatting only; ogar-vocab's 174 unit and 6 doc tests pass unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VysoWJ6vsyg3wEGc5v7T5v
What
HIRO puts every node's type on the wire as a string such as
ogit/MARS/Machine.HiroPortmaps those strings onto the nine0x0CXXautomation concepts that already exist, so a HIRO node dispatches on au16. This is the dispatch tabledocs/HIRO-IN-CLASSES.md§3.1 describes. It also gives Bardioc-rs a way to meet spear on shared classids, since spear's only public contract isSpearPort.ogit/MARS/Application,/Resource,/Software,/Machine0x0C01–0x0C04ogit/Automation/KnowledgeItem,/MARSNodeTemplate,/ActionHandler,/ActionApplicability,/Trigger0x0C05–0x0C090x000C: the next free one after HubSpo's0x000B.0x0006stays held for q2.ogit/Automation/MAIDhas no alias on purpose: MAID has no concept, no workload needs one, and minting before a falsifier passes is enum explosion.0x000Crow inAPP-CLASS-CODEBOOK-LAYOUT.md§2, andD-HIRO-PORTinDISCOVERY-MAP.md§2.13 with the decision, its scope and when to revisit it.Tests
hiro_classview_composes_canon_high_custom_low0x0C04_000C. Wire names are exact: no case folding and no OGIT CURIE spelling. MAID does not resolve.hiro_covers_exactly_the_automation_domainHiroPortjoins the render-prefix collision list and the0x1000reservation listogar-vocab: 174 unit tests and 6 doc tests pass on Rust 1.99. With the Trigger alias removed, both new tests fail and nameautomation_trigger (0x0C09).A lance-graph PR follows that mirrors
0x000CasAppPrefix::Hiroinogar_codebook.rs, the same way #1438 mirrored HubSpo.🤖 Generated with Claude Code
https://claude.ai/code/session_01VysoWJ6vsyg3wEGc5v7T5v
Generated by Claude Code
Summary by CodeRabbit