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
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
|
#!/usr/bin/env python3
"""P45 character-animation analysis.
Reads the 4x `p45-char-{a,b,c}.png` crops (real LV_P45_SHOT screenshots, cropped
around the avatar cell) + the shot log and writes docs/evidence/p45-analysis.txt:
1. character source cell + on-screen dest rect + facing/frame per frame;
2. pairwise inter-frame difference: Δchanged_px (max-channel > 16) and mean |Δ|;
3. the deterministic frame sequence table (frame index -> source cell) per facing;
4. a palette check that the original pixel-art character really is in each crop.
Exit 0 when the differences are significant, 1 otherwise, 2 on usage error.
"""
import re
import sys
from pathlib import Path
import numpy as np
from PIL import Image
CELL_W = 32
CELL_H = 48
VARIANT_STRAW_HAT = 4
DIR_SOUTH = 1
DIR_EAST = 3
CHANGED_MIN = 400
LINE_RE = re.compile(r"p45-shot=(\S+) tick=(\d+) facing=(\S+) moving=(\w+) dest=(\d+),(\d+),(\d+),(\d+) frame=(.+)")
SKIN = np.array([234, 194, 152])
ROBE = np.array([146, 96, 70])
STRAW = np.array([216, 190, 120])
def source_x(variant, direction, frame):
return ((variant * 4 + direction) * 6 + frame) * CELL_W
def load(path):
return np.asarray(Image.open(path).convert("RGB"), dtype=int)
def palette_hits(img):
def near(ref):
return (np.abs(img - ref).max(axis=2) <= 18).sum()
return int(near(SKIN) + near(ROBE) + near(STRAW))
def main(argv):
if len(argv) != 4:
sys.stderr.write(__doc__)
return 2
record_dir = Path(argv[1])
log_text = Path(argv[2]).read_text(encoding="utf-8")
out_txt = Path(argv[3])
shots = []
for line in log_text.splitlines():
m = LINE_RE.search(line)
if m:
shots.append(
{
"name": m.group(1),
"tick": int(m.group(2)),
"facing": m.group(3),
"moving": m.group(4) == "true",
"dest": tuple(int(m.group(i)) for i in range(5, 9)),
"frame": m.group(9),
}
)
if len(shots) < 3:
sys.stderr.write("need three p45-shot frames\n")
return 1
images = {s["name"]: load(record_dir / f"{s['name']}.png") for s in shots}
lines = ["P45 character animation analysis (LV_P45_SHOT=1, 4x NEAREST crops)", ""]
failures = []
for s in shots:
x, y, w, h = s["dest"]
hits = palette_hits(images[s["name"]])
lines.append(f"[{s['name']}] facing={s['facing']} moving={s['moving']} frame={s['frame']} "
f"dest=({x},{y},{w}x{h}) palette_hits={hits}")
if w != CELL_W or h != CELL_H:
failures.append(f"{s['name']} dest cell is not 32x48")
if hits < 50:
failures.append(f"{s['name']} character palette not found ({hits})")
lines.append("")
lines.append("inter-frame delta (4x crop, aligned on the same avatar cell):")
names = [s["name"] for s in shots]
for i in range(len(names)):
for j in range(i + 1, len(names)):
a, b = images[names[i]], images[names[j]]
d = np.abs(a - b)
changed = int((d.max(axis=2) > 16).sum())
mean = float(d.mean())
lines.append(f" {names[i]} vs {names[j]}: changed_px(>16)={changed} mean_abs_diff={mean:.2f}")
if changed < CHANGED_MIN:
failures.append(f"{names[i]}-{names[j]} changed_px {changed} < {CHANGED_MIN}")
lines.append("")
lines.append("deterministic frame sequence (source cell x offsets, StrawHat variant):")
for label, direction in (("SouthFacing", DIR_SOUTH), ("EastFacing", DIR_EAST)):
offsets = [source_x(VARIANT_STRAW_HAT, direction, f) for f in range(6)]
lines.append(f" {label}: walk1..4+idle1..2 -> {offsets}")
lines.append("")
lines.append("verdict = " + ("PASS" if not failures else "FAIL"))
out_txt.parent.mkdir(parents=True, exist_ok=True)
out_txt.write_text("\n".join(lines) + "\n", encoding="utf-8")
sys.stdout.write("\n".join(lines) + "\n")
if failures:
for f in failures:
sys.stderr.write(f"FAIL {f}\n")
return 1
return 0
if __name__ == "__main__":
raise SystemExit(main(sys.argv))
|
