parity: run world_spawn in four shards

world_spawn starts a vanilla and a Horizon server for each of 60 seeds,
one at a time: 1,144-1,548 s on the CI runner, the parity job's longest
harness and its floor whenever a PR's changes reach it. With
PARITY_SHARDS = 4 each shard creates every fourth seed's worlds, in its
own build directory.

The shard helper datafix had moves to tools/parity_shard.py, which both
use. It stays out of lab.py, which every harness imports.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

tools/parity_shard.py

@@ -0,0 +1,21 @@
+"""A sharded harness's share of its cases.
+
+A harness that declares PARITY_SHARDS = n runs as n entries in tools/parity.py, which sets PARITY_SHARD=i/n for
+each. Outside the runner PARITY_SHARD is unset and a harness runs everything. Kept out of lab.py: every harness
+imports lab.py, so a change there reruns them all.
+"""
+import os
+
+
+def shard(items):
+    """Every n-th item from the i-th, or all of them outside the runner."""
+    spec = os.environ.get("PARITY_SHARD")
+    if not spec:
+        return items
+    index, count = map(int, spec.split("/"))
+    return items[index - 1::count]
+
+
+def label():
+    """A directory name for this share ("2-4", or "all"): shards run at once and must not share outputs."""
+    return os.environ.get("PARITY_SHARD", "all").replace("/", "-")

tools/test_datafix_parity.py

@@ -33,6 +33,7 @@ from pathlib import Path
 
 sys.path.insert(0, str(Path(__file__).resolve().parent))
 import anvil_dump  # noqa: E402
+import parity_shard  # noqa: E402
 
 ROOT = Path(__file__).resolve().parent.parent
 LOCAL = ROOT / '.local' / 'datafix'
@@ -63,7 +64,7 @@ def java_classpath():
     import test_model_baking_parity as baking
     lock, jars, _ = base.verify_artifacts()
     libraries = baking.client_libraries()
-    build = LOCAL / 'build' / os.environ.get('PARITY_SHARD', 'all').replace('/', '-')  # shards compile apart
+    build = LOCAL / 'build' / parity_shard.label()  # shards compile apart
     build.mkdir(parents=True, exist_ok=True)
     classes, dependencies, _ = harness.compile_oracle(build, 'VanillaDataFixOracle', jars, libraries,
                                                       lock['minecraft']['java_major'])
@@ -78,15 +79,6 @@ def build_rust():
     return Path(meta['target_directory']) / 'release' / 'examples' / 'datafix_world'
 
 
-def shard(items):
-    """This run's share of items when tools/parity.py runs the harness in parts (PARITY_SHARD=i/n)."""
-    spec = os.environ.get('PARITY_SHARD')
-    if not spec:
-        return items
-    index, count = map(int, spec.split('/'))
-    return items[index - 1::count]
-
-
 def run_vanilla(classpath, world, out, target=None):
     # The oracle replaces OUT itself: left in place, an unchanged run's outputs can be reused.
     log = out.parent / (out.name + '.log')
@@ -226,7 +218,7 @@ def main():
     band = args.band
     if band and not args.versions:  # worlds in the layout of before 1.16, saved before the target
         worlds = [w for w in worlds if json.loads((world_path(w) / 'datafix-source.json').read_text())['data_version'] < band]
-    versions = shard(args.versions or worlds)
+    versions = parity_shard.shard(args.versions or worlds)
     classpath = java_classpath()
     binary = None if args.vanilla_only else Path(args.binary) if args.binary else build_rust()
     total = 0

tools/test_world_spawn_parity.py

@@ -29,6 +29,10 @@ Seeds: the player report's seed (1891623539983467248: Horizon spawned at
 -48 -48, vanilla at 152 145), the chunk generation harness's seeds and
 --random seeds from a fixed generator.
 
+tools/parity.py runs it in PARITY_SHARDS parts, each creating the worlds of
+every PARITY_SHARDS-th seed: each seed is a vanilla and a Horizon server start,
+~15 s, and 60 seeds in one run were the longest harness (~25 min).
+
 usage: test_world_spawn_parity.py [--random N] [--jobs N] [--origins-only]
 """
 import argparse
@@ -43,16 +47,18 @@ from concurrent.futures import ThreadPoolExecutor
 from pathlib import Path
 
 try:
-    from . import anvil_dump, lab, oracle_harness as oh, test_rust_parity as base
+    from . import anvil_dump, lab, oracle_harness as oh, parity_shard, test_rust_parity as base
     from .test_dedicated_server_parity import Server, free_port
 except ImportError:
     import anvil_dump
     import lab
     import oracle_harness as oh
+    import parity_shard
     import test_rust_parity as base
     from test_dedicated_server_parity import Server, free_port
 
 NAME = 'world-spawn-parity'
+PARITY_SHARDS = 4
 REPORT = 1891623539983467248
 SEEDS = [REPORT, 12345, -7046029254386353131, 987654321987, 4242, -1, 0x5DEECE66D, 20260607, 777, 0]
 RESOURCES = lab.RESOURCES
@@ -107,7 +113,7 @@ def main():
     parser.add_argument('--seed', type=int, action='append', help='only these seeds (repeatable)')
     args = parser.parse_args()
     lock, jars, libraries = base.verify_artifacts()
-    build = lab.LOCAL / NAME
+    build = lab.LOCAL / NAME / parity_shard.label()  # shards run at once
     build.mkdir(parents=True, exist_ok=True)
     dependencies = os.pathsep.join(map(str, [jars['server'], *libraries]))
     classes = build / 'classes'
@@ -116,7 +122,7 @@ def main():
                    '-d', classes, lab.ROOT / 'mods/horizon/src/test/java/dev/horizonlab/parity/VanillaWorldSpawnOracle.java'],
                   build, 'compile-java')
     oh.run_logged(['cargo', 'build', '--release', '-p', 'horizon-world', '--example', 'spawn_origin'], build, 'compile-probe')
-    all_seeds = args.seed or seeds(args.random)
+    all_seeds = parity_shard.shard(args.seed or seeds(args.random))
     started = time.time()
 
     vanilla, rust = origins(build, classes, dependencies, all_seeds)