codec-re: 00 NN is RLE; full Tran segment-0 decode (4 of 5 events)
User uploaded a Vert-heavy event (JQ0) and a Mic-heavy event (V70). Those two were exactly what was needed to crack the next piece: - 00 NN block = run-length-encoded zero deltas in the current channel. Append NN copies of the current cumulative value (no change). - find_data_start now recognizes 00 NN as a valid first tag (some events begin with a leading 00 NN RLE block). - decode_tran_initial now decodes the FULL segment 0 (not just the first data block). Results across 5 fixture events: - M529LL1A.SP0 (loud-all-channels) : 510 / 510 ✓ - M529LL1L.JQ0 (Vert-heavy) : 510 / 510 ✓ - M529LL1L.V70 (Mic-heavy) : 510 / 510 ✓ - M529LL1A.SV0 (loud-from-start) : 58 / 58 ✓ - M529LL1A.SS0 (loud-from-start) : 42 / 502 (stops at first 30 04) The 30 04 block (only seen in loud-from-start events) hasn't been decoded yet — likely a channel-switch marker for the high-amplitude regime. Also discovered: segment header (40 02) payload bytes [0:2] = T_delta at first sample of new segment, [6:8] = byte length to next segment. Multi-segment Tran decoding still diverges after sample 512 because the per-segment channel ordering after the header is unknown. Tests: 40 pass (up from 36). Files: - minimateplus/waveform_codec.py: find_data_start fix, RLE handling, full segment-0 decode in decode_tran_initial - tests/test_waveform_codec.py: synthetic RLE test, full segment 0 tests for JQ0 and V70 - tests/fixtures/5-11-26/: M529LL1L.JQ0, M529LL1L.V70 + TXT exports - docs/instantel_protocol_reference.md §7.6.1: RLE + segment-header docs
This commit is contained in:
Claude
@@ -78,25 +78,23 @@ def test_find_data_start_locates_first_block(event_name):
|
||||
body = _bw_body(path)
|
||||
start = find_data_start(body)
|
||||
assert 0 <= start < 20, f"expected start in [0, 20), got {start}"
|
||||
assert body[start] == 0x10
|
||||
assert body[start + 1] % 4 == 0
|
||||
assert 0 < body[start + 1] <= 0xFC
|
||||
assert body[start] in (0x00, 0x10, 0x20, 0x30, 0x40), (
|
||||
f"first tag byte 0x{body[start]:02x} not a recognized block type"
|
||||
)
|
||||
assert body[start + 1] % 4 == 0 or (body[start] == 0x40 and body[start + 1] == 0x02)
|
||||
|
||||
|
||||
def test_find_data_start_preamble_lengths():
|
||||
"""All 4 events have either a 7-byte (single-shot) or 9-byte (continuous) preamble."""
|
||||
starts = {}
|
||||
def test_find_data_start_canonical_offset_7():
|
||||
"""All events have a 7-byte preamble (3-byte magic + 4-byte Tran anchors)."""
|
||||
for name in FIXTURES_INFO:
|
||||
path = _fixture_path(name)
|
||||
if not os.path.exists(path):
|
||||
pytest.skip(f"fixture missing: {path}")
|
||||
body = _bw_body(path)
|
||||
starts[name] = find_data_start(body)
|
||||
# Empirically: events a, b have 9-byte preamble; events c, d have 7-byte.
|
||||
assert starts["event-a"] == 9
|
||||
assert starts["event-b"] == 9
|
||||
assert starts["event-c"] == 7
|
||||
assert starts["event-d"] == 7
|
||||
# Sanity: magic
|
||||
assert body[0:3] == b"\x00\x02\x00", f"{name}: bad magic"
|
||||
# First tag at offset 7
|
||||
assert find_data_start(body) == 7, f"{name}: expected start=7"
|
||||
|
||||
|
||||
# ── Block walker ─────────────────────────────────────────────────────────────
|
||||
@@ -274,9 +272,46 @@ TRAN_INITIAL_FIXTURES = [
|
||||
[-745, -762, -771, -774, -779, -794, -808, -811, -811, -819],
|
||||
46,
|
||||
),
|
||||
# Vert-heavy event (T near zero) — segment 0 = 510 samples, all decode correctly.
|
||||
(
|
||||
os.path.join(os.path.dirname(__file__), "fixtures", "5-11-26", "M529LL1L.JQ0"),
|
||||
[0] * 4 + [-1, 0, 0, -1, -1, 0],
|
||||
38,
|
||||
),
|
||||
# Mic-heavy event (geos all near zero) — segment 0 = 482 samples.
|
||||
(
|
||||
os.path.join(os.path.dirname(__file__), "fixtures", "5-11-26", "M529LL1L.V70"),
|
||||
[0] * 10,
|
||||
6,
|
||||
),
|
||||
]
|
||||
|
||||
|
||||
def _full_truth(path):
|
||||
"""Load the BW ASCII truth for an event."""
|
||||
import re
|
||||
with open(path + ".TXT", "r", encoding="utf-8", errors="replace") as f:
|
||||
lines = f.read().splitlines()
|
||||
# Find columns header.
|
||||
header_idx = None
|
||||
for i, line in enumerate(lines):
|
||||
if "Tran" in line and "Vert" in line and "Long" in line and "MicL" in line:
|
||||
header_idx = i
|
||||
break
|
||||
if header_idx is None:
|
||||
return None
|
||||
out = []
|
||||
for line in lines[header_idx + 1:]:
|
||||
parts = re.split(r"\s+", line.strip())
|
||||
if len(parts) < 4:
|
||||
continue
|
||||
try:
|
||||
out.append(round(float(parts[0]) * 200))
|
||||
except ValueError:
|
||||
continue
|
||||
return out
|
||||
|
||||
|
||||
@pytest.mark.parametrize("path,expected,n_required", TRAN_INITIAL_FIXTURES)
|
||||
def test_decode_tran_initial_matches_ground_truth(path, expected, n_required):
|
||||
"""The Tran initial decoder produces values matching the BW ASCII export exactly."""
|
||||
@@ -312,3 +347,32 @@ def test_decode_tran_initial_synthetic_body():
|
||||
decoded = decode_tran_initial(body)
|
||||
# T[0]=10, T[1]=20, then deltas (+1, -1, +2, -2) from T[1]=20
|
||||
assert decoded == [10, 20, 21, 20, 22, 20]
|
||||
|
||||
|
||||
def test_decode_tran_initial_with_rle():
|
||||
"""A synthetic body with 00 NN RLE block runs the current Tran value forward."""
|
||||
# T[0]=5, T[1]=5, then 00 08 RLE block = 8 zero deltas → T[2..9] = 5
|
||||
body = b"\x00\x02\x00\x00\x05\x00\x05" + b"\x00\x08"
|
||||
decoded = decode_tran_initial(body)
|
||||
assert decoded == [5, 5, 5, 5, 5, 5, 5, 5, 5, 5]
|
||||
|
||||
|
||||
def test_decode_tran_initial_full_segment_silent_events():
|
||||
"""For events with near-silent Tran, segment 0 (~482-510 samples) decodes fully."""
|
||||
for path, _, _ in TRAN_INITIAL_FIXTURES[3:]: # JQ0 (Vert-heavy) and V70 (Mic-heavy)
|
||||
if not os.path.exists(path):
|
||||
pytest.skip(f"fixture missing: {path}")
|
||||
with open(path, "rb") as f:
|
||||
body = f.read()[43:-26]
|
||||
truth = _full_truth(path)
|
||||
decoded = decode_tran_initial(body)
|
||||
assert decoded is not None
|
||||
# The decoder should produce a clean run of samples; check ALL of them
|
||||
# match truth (segment 0 is fully solved for events where T is near zero).
|
||||
n = len(decoded)
|
||||
for i in range(n):
|
||||
assert decoded[i] == truth[i], (
|
||||
f"{os.path.basename(path)}: sample {i}: decoded={decoded[i]} truth={truth[i]}"
|
||||
)
|
||||
# And we should have decoded at least 400 samples (= segment 0 worth).
|
||||
assert n >= 400, f"only {n} samples decoded for {path}"
|
||||
|
||||
Reference in New Issue
Block a user