fix(parser): bare 'it's a <type>' replaces card types per CR 205.1a (Arixmethes, #5213) by minion1227 · Pull Request #5231 · phase-rs/phase

minion1227 · 2026-07-07T06:58:36Z

Summary

Fixes Arixmethes, Slumbering Isle (issue #5213): while it has a slumber counter it should become just a Land, but the engine showed the impossible "Legendary Creature Land — Kraken" — the Creature type was never removed.

Its static reads As long as Arixmethes has a slumber counter on it, it's a land. (It's not a creature.). The parser emitted an additive AddType{Land}, so the Kraken stayed a creature. Per CR 205.1a, an effect that sets a card type replaces the object's existing card types unless it retains them (CR 205.1b: "in addition to its other types" / "still a"). A bare "it's a land" therefore replaces.

Fix

parse_pronoun_becomes_type_static now runs its computed modifications through maybe_replace_card_types, which converts them to a single replacing SetCardTypes only when all of:

every modification is an AddType (a pure card-type change — no P/T, subtype, keyword, color, or supertype),
none of the added types is Creature (creature forms retain via the CR 205.1b "artifact creature" rule and their P/T),
the body carries no CR 205.1b retention marker (in addition to / still a).

So it's a land → SetCardTypes{[Land]} (replaces, removing Creature), while creature animations (Gideon Blackblade, Midnight Mangler, Circle of the Moon Druid) and in addition to its other types forms stay additive. No new runtime: the existing SetCardTypes layer application (game/layers.rs) sets the core types and drops the correlated Kraken creature subtype, leaving a Legendary Land.

Files changed

crates/engine/src/parser/oracle_static/type_change.rs
crates/engine/src/parser/oracle_static/tests.rs
crates/engine/src/game/layers.rs

Anchored on

crates/engine/src/parser/oracle_static/type_change.rs parse_attached_isnt_and_is_type / the Imprisoned parse_enchanted_becomes_type_with_ability — existing handlers that already model type REPLACEMENT via SetCardTypes / RemoveType; this reuses SetCardTypes for the self "it's a [type]" form.
crates/engine/src/game/layers.rs:4886 ContinuousModification::SetCardTypes — the existing replacement runtime that sets core types and retains only correlated subtypes.

CR references

CR 205.1a — an effect that sets an object's card type replaces its existing card types; correlated subtypes of a removed card type are also removed.
CR 205.1b — retention only when the effect says "in addition to its other types", "still a [type]", or is an "artifact creature" form.

Track

Developer

LLM

Model: claude-opus-4-8
Thinking: high

Verification

cargo fmt --all — clean
./scripts/check-parser-combinators.sh — clean (exit 0)
./scripts/check-engine-authorities.sh — clean (exit 0)
cargo test -p engine --lib parser::oracle_static — 1014 pass / 0 fail
Parser test arixmethes_bare_is_a_land_replaces_card_types (asserts SetCardTypes{[Land]} + the slumber-counter condition, and that in addition to / artifact creature stay additive).
Runtime test set_card_types_land_strips_creature_and_creature_subtype (a Legendary Creature Kraken under SetCardTypes([Land]) becomes a Legendary Land: Creature removed, Kraken subtype dropped, Legendary retained).

Scope Expansion

None.

Validation Failures

None.

CI Failures

None.

🤖 Generated with Claude Code

…Arixmethes) Closes phase-rs#5213. A self type-change 'it's a <type>' with no CR 205.1b retention marker ('in addition to its other types' / 'still a') sets the object's card type, which REPLACES its existing card types (CR 205.1a). The parser emitted an additive AddType, so Arixmethes, Slumbering Isle stayed a 'Creature Land' while it had a slumber counter instead of becoming just a Land. parse_pronoun_becomes_type_static now converts a pure non-creature card-type change (only AddType mods, no Creature, no retention marker) to a replacing SetCardTypes via maybe_replace_card_types. Creature animations (Gideon Blackblade / Midnight Mangler / Circle of the Moon Druid) and 'in addition to its other types' forms keep the additive path. The existing SetCardTypes runtime drops the correlated creature subtype (Kraken), leaving a Legendary Land. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

gemini-code-assist

Code Review

This pull request implements card-type replacement logic for self type-change effects (CR 205.1a, e.g., Arixmethes, Slumbering Isle) by introducing maybe_replace_card_types and adding corresponding tests. Feedback highlights two style guide violations: a non-standard CR annotation format on line 1413 and the use of verbatim string scanning (scan_contains) instead of nom combinators on line 1433.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

gemini-code-assist · 2026-07-07T07:01:44Z

    Some(def)
 }

+/// CR 205.1a / CR 205.1b: Decide whether a self type-change is a REPLACEMENT or


[MEDIUM] Non-standard CR annotation format. Evidence: crates/engine/src/parser/oracle_static/type_change.rs:1413.

Why it matters: The CR annotation uses a combined slash format (CR 205.1a / CR 205.1b) which may fail automated regex validation requiring the strict CR <number>: <description> format.

Suggested fix: Split the CR annotations into separate, cleanly formatted lines.

Suggested change

/// CR 205.1a / CR 205.1b: Decide whether a self type-change is a REPLACEMENT or

/// CR 205.1a: An effect that sets an object's card type replaces its existing card types.

/// CR 205.1b: Some effects specify that the object retains one or more of its prior card types.

/// Decide whether a self type-change is a REPLACEMENT or

References

Every rules-touching line of engine code must carry a comment of the form CR : (regex CR \d{3}(.\d+[a-z]?)?). ^(link)

gemini-code-assist · 2026-07-07T07:01:44Z

+    if nom_primitives::scan_contains(body_lower, "in addition to")
+        || nom_primitives::scan_contains(body_lower, "still a")
+    {


[MEDIUM] Avoid verbatim string scanning for parsing Oracle phrases. Evidence: crates/engine/src/parser/oracle_static/type_change.rs:1433.

Why it matters: Using scan_contains for phrases like 'still a' or 'in addition to' bypasses the robust nom-based parser and creates fragile matches. Instead, decompose compound phrases into modular, reusable parsers for constituent parts and compose them using idiomatic combinator aggregates (like nested alt and tag sequences) to prevent combinatorial explosion and improve maintainability.

Suggested change

if nom_primitives::scan_contains(body_lower, "in addition to")

|| nom_primitives::scan_contains(body_lower, "still a")

{

let mut parser = alt((

tag("in addition to"),

tag("still a"),

tag("still an"),

));

References

Avoid verbatim string equality for parsing Oracle phrases as it bypasses the robust nom-based parser and creates fragile matches. Instead, decompose compound phrases into modular, reusable parsers for constituent parts and compose them using idiomatic combinator aggregates.

github-actions · 2026-07-07T07:16:04Z

Parse changes introduced by this PR · 1 card(s), 1 signature(s) (baseline: main `27d836864f61`)

1 card(s) · static/Continuous · field `mods`: `add type land` → `set card types land`

Examples: Arixmethes, Slumbering Isle

1 card(s) had Oracle-text changes (errata/reprint) — excluded as non-parser.

matthewevans

Maintainer review: right seam, CR 205.1a/205.1b checked, parse-diff is scoped to Arixmethes, and tests cover both parser shape and layer behavior.

…-land coverage phase-rs#5231 (merged) already covers Arixmethes land replacement, the slumber condition, additive retention, and artifact-creature retention. Swap the duplicate tests for a distinct untested sibling: the "it's a <type>" replacement and its "in addition to" retention exception are not land-specific — "~ is an artifact" REPLACES via SetCardTypes while "~ is an artifact in addition to its other types" stays additive.

…phase-rs#5218) * fix(parser): self-ref "it's a <type>" replaces card types (CR 205.1a) A self-referential type-change static of the form "it's a <core type>" (Arixmethes, Slumbering Isle: "As long as ~ has a slumber counter on it, it's a land.") went through the pronoun-animation path, which is unconditionally additive: it emitted `AddType(Land)` and left the permanent a "Legendary Creature Land — Kraken" — an impossible land creature. Per CR 205.1a, "is a <type>" WITHOUT "in addition to its other types" REPLACES the permanent's card types, so it must become a Land only (Creature removed → "Legendary Land — Kraken"). In `parse_pronoun_becomes_type_static`, when the body is a pure bare core-type set (no retained P/T, keywords, color, or subtype change) and carries no "in addition"/"still a <type>" retention clause, collapse the additive `AddType` list into one replacing `SetCardTypes` — mirroring the aura type-change path's `needs_set_card_types`, and reusing the existing `SetCardTypes` layer handler (which retains supertypes and creature subtypes, yielding "Legendary Land — Kraken"). The additive animations (Gideon Blackblade, manlands) are excluded because they carry a P/T, keywords, or an explicit retention clause. Closes phase-rs#5213 * Update crates/engine/src/parser/oracle_static/type_change.rs Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * Update crates/engine/src/parser/oracle_static/type_change.rs Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> * fix(parser): use scan_contains in type-set guard; robust Arixmethes test CI fixes for the self-ref "it's a <type>" replacement: - The retention-clause guard used raw `str::contains`, which the parser combinator gate bans. Route through the approved `nom_primitives::scan_contains` helper (mirrors the existing "in addition to" scan in this module). - Make the Arixmethes regression assert `contains(SetCardTypes[Land])` + no `AddType` rather than an exact `==` on the modification vector, so it is not brittle to incidental additional modifications, and fully-qualify `CoreType` to drop the local import. * fix(parser): preserve CR 205.1b artifact-creature retention in type-set flip The self-ref "it's a <type>" → SetCardTypes replacement wrongly collapsed the artifact-creature animation class (Grond, the Gatebreaker; Midnight Mangler; Phoenix Fleet Airship) from additive `AddType artifact/creature` to a replacing `SetCardTypes`, which removes prior card types. Per CR 205.1b, animating a permanent INTO a creature ("~ is an artifact creature") GAINS the creature type while retaining its other card types — it must stay additive. Restrict the replacement to type-sets that do NOT add Creature: only a non-creature type-set like Arixmethes' "it's a land" (CR 205.1a / CR 305.7) replaces the card types. Add a regression asserting "During your turn / During turns other than yours, ~ is an artifact creature" stays `AddType` (no `SetCardTypes`). * docs(parser): drop incorrect CR 305.7 from it's-a-land type-replacement comment CR 305.7 governs setting a land's SUBTYPE (basic land types) and explicitly does not add or remove card types; card-type replacement for "it's a land" is CR 205.1a alone. * test(parser): replace duplicate phase-rs#5231 tests with distinct non-land coverage phase-rs#5231 (merged) already covers Arixmethes land replacement, the slumber condition, additive retention, and artifact-creature retention. Swap the duplicate tests for a distinct untested sibling: the "it's a <type>" replacement and its "in addition to" retention exception are not land-specific — "~ is an artifact" REPLACES via SetCardTypes while "~ is an artifact in addition to its other types" stays additive. --------- Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> Co-authored-by: matthewevans <matthewevans@users.noreply.github.com>

minion1227 requested a review from matthewevans as a code owner July 7, 2026 06:58

github-actions Bot added the needs-maintainer AI-contribution PR requires human triage (Non-dev track or unresolved gaps) label Jul 7, 2026

gemini-code-assist Bot reviewed Jul 7, 2026

View reviewed changes

matthewevans self-assigned this Jul 7, 2026

matthewevans approved these changes Jul 7, 2026

View reviewed changes

matthewevans added the bug Bug fix label Jul 7, 2026

matthewevans added this pull request to the merge queue Jul 7, 2026

matthewevans removed their assignment Jul 7, 2026

matthewevans mentioned this pull request Jul 7, 2026

fix(parser): "it's a land" replaces card types so Arixmethes loses Creature (#5213) #5233

Closed

Merged via the queue into phase-rs:main with commit 7db13c5 Jul 7, 2026
12 checks passed

matthewevans mentioned this pull request Jul 7, 2026

fix(parser): self-ref "it's a <type>" replaces card types (CR 205.1a) #5218

Merged

minion1227 deleted the minion_arixmethes branch July 9, 2026 17:49

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

fix(parser): bare 'it's a <type>' replaces card types per CR 205.1a (Arixmethes, #5213)#5231

fix(parser): bare 'it's a <type>' replaces card types per CR 205.1a (Arixmethes, #5213)#5231
matthewevans merged 1 commit into
phase-rs:mainfrom
minion1227:minion_arixmethes

minion1227 commented Jul 7, 2026 •

edited

Loading

Uh oh!

gemini-code-assist Bot left a comment

Uh oh!

gemini-code-assist Bot Jul 7, 2026

Uh oh!

gemini-code-assist Bot Jul 7, 2026

Uh oh!

github-actions Bot commented Jul 7, 2026

Uh oh!

matthewevans left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

-/// CR 205.1a / CR 205.1b: Decide whether a self type-change is a REPLACEMENT or
+/// CR 205.1a: An effect that sets an object's card type replaces its existing card types.
+/// CR 205.1b: Some effects specify that the object retains one or more of its prior card types.
+/// Decide whether a self type-change is a REPLACEMENT or

-    if nom_primitives::scan_contains(body_lower, "in addition to")
-        || nom_primitives::scan_contains(body_lower, "still a")
-    {
+    let mut parser = alt((
+        tag("in addition to"),
+        tag("still a"),
+        tag("still an"),
+    ));

Uh oh!

Uh oh!

Conversation

minion1227 commented Jul 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Fix

Files changed

Anchored on

CR references

Track

LLM

Verification

Scope Expansion

Validation Failures

CI Failures

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

gemini-code-assist Bot Jul 7, 2026

Choose a reason for hiding this comment

Uh oh!

gemini-code-assist Bot Jul 7, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented Jul 7, 2026

Parse changes introduced by this PR · 1 card(s), 1 signature(s) (baseline: main 27d836864f61)

1 card(s) · static/Continuous · field mods: add type land → set card types land

Uh oh!

matthewevans left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

minion1227 commented Jul 7, 2026 •

edited

Loading

Parse changes introduced by this PR · 1 card(s), 1 signature(s) (baseline: main `27d836864f61`)

1 card(s) · static/Continuous · field `mods`: `add type land` → `set card types land`