xarray-spatial 0.10.11__tar.gz → 0.10.13__tar.gz

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-accuracy-state.csv

@@ -11,7 +11,7 @@ diffusion,2026-05-01,,LOW,1;2;5,"LOW: no Kahan summation across long iterations
 edge_detection,2026-05-01,,,,Thin wrappers around convolve_2d with fixed Sobel/Prewitt/Laplacian kernels; no issues found
 emerging_hotspots,2026-04-30,,MEDIUM,2;3,MEDIUM: threshold_90 uses int() (truncation) instead of ceil() so n_times=11 requires only 9/11 (81.8%) instead of 90%. MEDIUM: NaN time steps produce gi_bin=0 which classifier counts as 'non-significant' rather than missing; threshold_90 uses full n_times not valid count. LOW: 'global_std == 0' check does not catch NaN std for fully/mostly NaN inputs.
 fire,2026-06-19,3394,MEDIUM,5,"Cat5: dask+numpy map_blocks declared float64 (meta default) while ngjit kernels return float32; numpy/cupy/dask+cupy all float32. Fixed 6 wrappers with dtype=np.float32 (PR #3396); bsc already dtype=int8. Cats 1-4 clean: per-pixel ops, no stencil/accumulation/projected-distance; NaN via x!=x; CUDA bounds strict <; rdnbr/ros divisions guarded. cupy+dask+cupy tests run on GPU host."
-flood,2026-04-30,,MEDIUM,2;5,"MEDIUM (not fixed): dask backend preserves float32 input dtype while numpy promotes to float64 in flood_depth and curve_number_runoff; DataArray inputs for curve_number, mannings_n bypass scalar > 0 (and CN <= 100) range validation, silently producing NaN/garbage."
+flood,2026-06-25,3499,MEDIUM,5,"Cat5 backend dtype divergence (#3499/this PR): flood_depth and curve_number_runoff document float64 output; numpy/cupy cast to float64 but _flood_depth_dask/_cn_runoff_dask skipped the cast, so float32 input leaked float32 on dask + dask+cupy (numpy/cupy=float64, dask/dask+cupy=float32). Confirms the 2026-04-30 note. Fix: cast hand/p to float64 at the top of both dask helpers; dask+cupy wrappers reuse them so all 4 backends now return float64. Other flood fns unaffected: travel_time/flood_depth_vegetation upcast via float64 _TAN_MIN clamp, inundation via 1.0/0.0 literals, vegetation_roughness via np.interp. 8 new dtype tests across all 4 backends. Cats 1-4 clean; pure vectorized numpy/cupy/dask, no numba/cuda kernels, no neighborhood stencil, no geodesic math. CUDA available; cupy + dask+cupy verified (96 tests pass). LOW (not fixed, documented): curve_number/mannings_n DataArray inputs bypass scalar range validation (CN in (0,100], n>0)."
 focal,2026-06-10,3214,MEDIUM,1;5,"mean() dtype divergence: numpy/dask+numpy cast to float64 (astype(float)) while cupy/dask+cupy forced float32, so output dtype was backend-dependent and float64 rasters lost precision on GPU (offset 1e7: GPU error 0.58 > true spread 0.42, same class as fixed #2831). mean() was left out of the #2769 _promote_float contract that apply/focal_stats follow. Fix #3214: _promote_float in mean(), drop hardcoded cupy.float32 in _mean_cupy/_mean_dask_cupy, excludes cast to working dtype for cross-backend match parity. CUDA available; all 4 backends executed (245 focal tests pass incl new 3214 dtype tests). Cats 2-4 clean: GPU kernels two-pass std/var (#2831 fix verified), NaN checks via v!=v, map_overlap depths == kernel radius, Gi* validated against reference test. LOW (documented, not fixed): mean() excludes mask only the center pixel; excluded sentinel values (e.g. -9999) still contribute to neighboring cells' means on all backends -- docstring says 'left unchanged rather than averaged', backend-consistent."
 geotiff,2026-06-14,3331,MEDIUM,2,"Pass 28 (2026-06-14, deep-sweep accuracy): MEDIUM fixed -- issue #3331 / PR #3332. Direct GeoTIFF read accepted a zero or non-finite ModelPixelScale (33550) / ModelTransformation (34264) diagonal: _geotags._extract_transform read sx=scale[0]/sy=scale[1] (and m[0]/m[5]) with no finite-nonzero check, then coords_from_pixel_geometry built arange(N)*pixel_width+origin, so a zero scale collapsed the whole axis onto the origin (constant, non-georeferenced coords) and a NaN/Inf scale produced an all-NaN/all-Inf axis -- a degenerate raster returned with no error on all four backends. Asymmetric: the VRT read path already rejected zero res_x/res_y (VRTUnsupportedError, _vrt_validation.py:202, comment notes the eager read 'would currently surface this as an opaque coord error' -- but it actually surfaced nothing) and the writer rejected zero-step coords (NonUniformCoordsError, _validation.py:1361). Fix adds DegeneratePixelSizeError (GeoTIFFAmbiguousMetadataError subclass, exported + docs table) and a _check_finite_nonzero_pixel_size guard in _extract_transform covering the scale-only, tiepoint+scale, and ModelTransformation paths; the unit-scale tiepoint fallback (literal 1.0) and the allow_rotated no-georef path are untouched. Guard sits in the shared extract path so numpy/dask/gpu/dask+gpu all reject identically -- verified all four locally on a degenerate fixture (CUDA available). NaN ModelTransformation diagonal still rejected: the rotation-tol check short-circuits on NaN (x>NaN is False) and falls through to the new guard. 29 tests in tests/unit/test_degenerate_pixel_size_3331.py (zero/NaN/+-Inf x 3 paths, subclass contract, positive controls, end-to-end open_geotiff). Existing geotiff unit+read suites: 2271 passed, 5 skipped; flake8+isort clean. Scope this pass: numerical core re-audit (overview kernels, coords transform math, decode predictor/nodata, dtypes) via 3 parallel readers -- overview float32-vs-float64 mean accumulator 'divergence' DISMISSED (the ngjit kernel float path and the numpy nanmean float path are mutually exclusive: float mean/min/max/median route to the kernel and return early, _overview.py:340-353); coords single-neighbor pixel-size recovery LOW (standard, negligible, bypassed by transform attr); decode/predictor/nodata clean (#3098 int-sentinel-before-float64-promote pattern not repeated). cuda-available. | Pass 27 (2026-06-12, deep-sweep): HIGH fixed -- issue #3260. to_geotiff(pack=True) cast the packed values to the integer dtype from attrs['mask_and_scale_dtype'] with no range or finiteness check: finite values outside the dtype range wrapped in the astype (4000.0 with SCALE=0.1 on int16 packs to 40000 and landed on disk as -25536, unpacking to -2553.6) and +/-Inf cast to a platform-defined integer (0 on linux/x86), all silently on all four backends; dask deferred the cast into the write's compute so there was no warning at all. Internally inconsistent: the adjacent NaN-no-sentinel guard exists precisely because 'the astype below would silently wrap'. Fix adds _pack_guard_int_range after the round, before the cast (eager raises at call time, dask via map_blocks from the write's single compute, mirroring #3235); exclusive upper bound iinfo.max+1 stays exact in float64 so int64/uint64 reject exactly-2**63 instead of wrapping. Also fixed en route: eager cupy no-sentinel integer pack crashed with TypeError in bool(out.isnull().any()) (implicit cupy->numpy); now routed through the cupy-safe _pack_guard_no_nan. 15 new tests in tests/write/test_pack_range_guard_3260.py covering all four backends (CUDA available, gpu legs executed), boundary iinfo.min/max round trips, round-back-into-range, uint underflow, and the 2**63 float64 bound. Scope this pass: post-2026-06-09 commits only (pack/unpack #3174/#3175/#3239/#3240/#3241, VRT offsets #3135, GPU streaming writer) since Pass 26 covered the rest 3 days earlier; overview kernels, _coords transform math, _decode predictor/orientation/LERC fill, and _nodata lifecycle re-read with no new findings; GPU streaming writer reviewed (per-band NaN rewrite and tile-row alignment mirror the full-array path). cuda-available. | Pass 26 (2026-06-09, deep-sweep): MEDIUM fixed -- issue #3098. _apply_eager_nodata_mask in _attrs.py compared the integer nodata sentinel AFTER promoting the buffer to float64, so int64/uint64 sentinels above 2**53 (INT64_MAX/UINT64_MAX) swallowed up to 512/1024 nearby valid values into NaN on the numpy-eager and cupy-eager backends, while the dask per-chunk mask (_delayed_read_window), the GPU GDS chunk path (_apply_nodata_mask_gpu), and the VRT path all compare at native integer width and masked only exact hits. Reproduced end-to-end: int64 file with nodata=INT64_MAX and values INT64_MAX-1..-513 gave 4 NaN eager vs 1 NaN dask. Same function was internally inconsistent: the mask_nodata=False pixels_present scan already compared at native width. Fix computes the mask at source dtype width before the float64 promotion (promotion itself stays unconditional per #2990); one site fixes both numpy and cupy eager since GPU routes through the helper via duck typing. 5 regression tests in tests/read/test_nodata.py (int64 exact-hit, eager-vs-dask parity, uint64, near-sentinel-no-hit pixels_present, gpu eager); verified on all 4 backends with CUDA. Also audited this pass with no findings: overview reduce kernels CPU vs GPU (empirical parity run incl. float32 median midpoint analysis: RN(a+b)/2 == RN((a+b)/2) so no divergence), unpack/pack scale-offset paths (#3075/#3065, mask-before-scale ordering consistent eager/dask, dask+gpu reuses CPU dask graph), bbox-to-window floor/ceil (GDAL touched semantics), VRT nearest mapping floor((out+0.5)*src/out) and Int64 nodata native-width round-trip, predictor 2/3 GPU kernels (lossless), writer NaN-to-sentinel gates. cuda-available; GPU paths executed, not just reviewed. | Pass 25 (2026-05-15): HIGH fixed -- issue #1975. _block_reduce_2d's cubic branch in xrspatial/geotiff/_writer.py gated the sentinel-to-NaN mask on arr2d.dtype.kind=='f', so to_geotiff(cog=True, overview_resampling='cubic', nodata=<finite>) on an integer raster fell through to an unmasked zoom(arr2d, 0.5, order=3). The bicubic spline blended the sentinel (e.g. -9999) into neighbouring valid cells; cast back to the source integer dtype, the boundary pixels surfaced as silent garbage. Reproduction (1024x1024 int16 + 256x256 nodata corner + nodata=-9999): lvl1 boundary [128, 124:132] showed [1082, 1082, 1085, 1134, 5, 93, 100, 100] instead of [-9999/NaN, ..., 100, 100, 100, 100]; max poisoned value 1134 (11x the actual data value of 100) and min -11104 (below the sentinel -9999). Same root cause as #1623 (float cubic + nodata) but for the integer dtype branch. Both CPU and GPU writers affected because _block_reduce_2d_gpu's cubic path falls back to _block_reduce_2d on CPU. Fix mirrors the float branch: promote the cropped block to float64, mask sentinel to NaN via the integer-range guard (mirrors _int_nodata_in_range), run scipy.ndimage.zoom(prefilter=False), rewrite NaN back to the sentinel, then np.round(...).astype(source_int_dtype) so the integer cast is well-defined. 12 regression tests in test_cog_cubic_int_overview_nodata_1975.py: helper-level cubic per int dtype (int16, uint16, int32), no-nodata regression, out-of-range sentinel no-op, fractional sentinel no-op, all-sentinel block fallback, float cubic regression guard, end-to-end 1024x1024 round-trip, non-constant int regression, cubic-vs-mean sentinel-mask parity, and GPU/CPU byte parity. All 3186 non-stale geotiff tests still pass (2 pre-existing failures unrelated: test_predictor2_big_endian_gpu references the hidden read_to_array symbol, and test_size_param_validation_gpu_vrt_1776 asserts pre-#1767 tile_size=4 behaviour). Categories: Cat 1 (precision loss from cubic spline blending sentinel into valid cells) + Cat 2 (NaN-equivalent corruption: the read-side int-to-NaN mask only catches exact sentinel hits, so the poisoned values survive as legitimate measurements) + Cat 5 (backend parity: CPU and GPU writers shared the same wrong cubic path). | Pass 23 (2026-05-14): HIGH fixed -- issue #1847. extract_geo_info parsed GDAL_NODATA via float() unconditionally, which loses 1 ULP on uint64 max (2**64-1) and int64 max (2**63-1). The downstream integer-mask gate info.min <= int(nodata) <= info.max then rejects the cast because float-rounded sentinel is one above the dtype max; the sentinel pixel survives as a literal valid integer instead of NaN. Same float-only parse in _reader._resolve_masked_fill (LERC fill) and _reader._sparse_fill_value (SPARSE_OK fill). VRT _vrt._parse_band_nodata had already fixed this for the XML parse path (PR #1833) but TIFF source-of-truth was never updated, so write_vrt([uint64.tif]) stringified the float-parsed nodata as '1.8446744073709552e+19' into XML where the VRT reader then rejected it for being out of range. Fix: lift the int-first parse into shared helper _parse_nodata_str in _geotags.py and reuse across the three TIFF-side sites. The helper tries int(text) first to preserve full precision, falls back to float(text) for NaN/Inf/scientific/fractional. Downstream gates already handle int values transparently because np.isfinite(int) works and int(int) is a no-op. 25 regression tests in test_nodata_int64_precision_1847.py: unit-level _parse_nodata_str matrix (int vs float branches, edge cases), eager open_geotiff (uint64 max / int64 max / int64 min / uint16 / int32 / float regression guards), read_geotiff_dask (uint64 max, int64 max), write_vrt + read_vrt round-trip with XML literal assertion, and a GPU parity test. All 2434 non-stale geotiff tests still pass (1 pre-existing test_size_param_validation_gpu_vrt_1776 failure unrelated -- test asserts pre-#1767 tile_size=4 behaviour). Categories: Cat 2 (NaN propagation: sentinel pixel survived as literal valid number on all 4 backends) + Cat 5 (backend inconsistency: VRT XML parse path handled 64-bit sentinels via _parse_band_nodata but TIFF parse path did not, even though write_vrt fed the latter into the former). Audited but did not file: LOW silent kwarg drop -- to_geotiff(da, 'out.vrt', photometric='miniswhite') drops the photometric arg at _write_vrt_tiled call (per-tile files written as MinIsBlack). Data round-trips correctly because no inversion happens on either side; only the tile photometric tag disagrees with the user's request. Niche path + no data corruption + metadata-only drift = LOW, not filed. | Pass 22 (2026-05-13): HIGH fixed -- issue #1809. MinIsWhite (photometric=0) inversion ran before the sentinel-to-NaN nodata mask on all four backends (eager numpy in open_geotiff, dask chunk reader, eager GPU in read_geotiff_gpu, GPU stripped fallback). Because the inversion rewrites the original sentinel value (e.g. uint8 nodata=0 becomes 255, float32 nodata=-9999 becomes 9999), the post-inversion mask matched the wrong pixels: cells whose stored value happened to equal iinfo.max - sentinel were flagged NaN while real sentinel cells survived as inverted values. PR #1804 (a5d78e4) had refactored the helper but kept the original ordering. Fix: introduce _miniswhite_inverted_nodata in _reader.py and stash the inverted sentinel on geo_info._mask_nodata; route every backend mask through that field, keeping geo_info.nodata + attrs[nodata] at the original value for write-side round-trip. Dask path also re-inverts the closure nodata at graph-build time, picking up _ifd_photometric / _ifd_samples_per_pixel stashed in _read_geo_info. 9 regression tests in test_miniswhite_nodata_1809.py cover uint8 nodata=0, uint16 nodata=65535, float32 nodata=-9999 across numpy, dask, and GPU backends plus no-collision and no-nodata controls. All 2424 non-stale geotiff tests pass (4 pre-existing failures unrelated to this fix). Categories: Cat 2 (NaN propagation: real data became NaN while sentinel survived as inverted value) + Cat 5 (backend inconsistency: all four backends share the identical wrong result, so they agreed on the wrong answer rather than diverged). | Pass 21 (2026-05-13): MEDIUM fixed -- issue #1774. open_geotiff / read_geotiff_dask / _apply_nodata_mask_gpu crashed with ValueError: cannot convert float NaN to integer when reading an integer TIFF whose GDAL_NODATA tag was the string ""nan"" / ""inf"" / ""-inf"". Three sites in xrspatial/geotiff/__init__.py called int(nodata) on the integer-dtype branch without first checking np.isfinite. _geotags.py:extract_geo_info parses the GDAL_NODATA tag through float(nodata_str) so a ""nan"" tag surfaces as Python NaN; the integer mask code then explodes. Sibling helpers _resolve_masked_fill and _sparse_fill_value in _reader.py already gate on not math.isnan(v) and not math.isinf(v) (the unfinished pass of #1581). Fix: gate each int(nodata) cast on np.isfinite(nodata). A non-finite sentinel on an integer file cannot match any pixel, so the mask is a no-op and the file dtype is preserved; attrs['nodata'] still carries the raw NaN/Inf sentinel so a write round-trip keeps the original GDAL_NODATA tag. The read_geotiff_dask effective_dtype branch already used try/except and was safe in practice, but tightened with the same isfinite gate for readability. 15 regression tests in test_nodata_nan_int_1774.py covering eager numpy (3 NaN variants + 6 Inf variants), in-range finite still masks regression guard, dask (NaN + Inf), and GPU (NaN + Inf + finite). All pass; 2023 existing geotiff tests still pass (7 pre-existing test_predictor2_big_endian_gpu failures unrelated: they reference xrspatial.geotiff.read_to_array which was hidden from the public namespace in #1708, 3 pre-existing matplotlib palette failures in test_features.py unrelated). Categories: Cat 2 (NaN propagation: NaN nodata produced a crash instead of being treated as missing) + Cat 5 (backend inconsistency: _resolve_masked_fill / _sparse_fill_value already guarded; the three __init__.py sites did not). | Pass 20 (2026-05-12): HIGH fixed -- PR #1691 (no issue created; agent harness blocked gh issue create). Integer COG overview pyramid mixed sentinel into reduced pixels. _block_reduce_2d (_writer.py:258-264) and _block_reduce_2d_gpu (_gpu_decode.py:3027-3028) promoted integer blocks to float64 but never masked the sentinel to NaN before nanmean / nanmin / nanmax / nanmedian. The reduction averaged the sentinel into surrounding valid cells (e.g. (-9999 + 100 + 100 + 100)/4 = -2425 cast back to int16), producing overview pixels that the read-side int-to-NaN mask in open_geotiff couldn't recover because they didn't equal the sentinel. Silent garbage at every zoom above level 0 for to_geotiff(int_data, cog=True, nodata=N). Methods affected: mean, min, max, median; nearest/mode safe (no averaging). Fix: gate the sentinel-to-NaN mask on representability in the source integer dtype (mirrors _int_nodata_in_range in _reader.py) so uint16+GDAL_NODATA=""-9999"" stays a no-op; rewrite all-sentinel-block NaN back to sentinel before the integer dtype cast so the cast is well-defined (the caller's post-overview loop in write() only runs for floats). GPU mirror gets the same path with cupy.where + cupy.isnan for byte parity with CPU. 38 regression tests in test_cog_int_overview_nodata_2026_05_12.py: _block_reduce_2d per-dtype/per-method matrix (uint8/uint16/int16/int32 x mean/min/max/median), all-sentinel-block, no-nodata regression, out-of-range sentinel no-op, end-to-end uint16 + int16 round-trip, 3-band integer COG, GPU per-dtype/per-method matrix, CPU/GPU byte-match parity. All 1606 existing geotiff tests still pass. Categories: Cat 1 (precision/representation loss in nan-aware reduction) + Cat 2 (silent NaN-equivalent corruption from sentinel poisoning) + Cat 5 (backend parity between float and integer code paths within the same writer). Deferred LOW: HTTP COG path (_read_cog_http at _reader.py:1638) skips the band-range validation that local/dask/GPU added in #1673; band=-1 silently selects the last channel on HTTP while local raises IndexError. Cat 5, MEDIUM-leaning but separate concern from the overview fix; one-finding-per-PR per project policy. | Pass 19 (2026-05-12): MEDIUM fixed -- issue #1655. read_vrt silently dropped <NODATA>0</NODATA> on a SimpleSource because of src.nodata or nodata at _vrt.py:370. Python treats 0.0 as falsy, so the per-source sentinel fell through to the band-level <NoDataValue> (or None when missing) and pixels equal to 0.0 in the source file survived as valid data. The in-code comment acknowledged the quirk as backward compat, but the resulting behaviour silently biased every NaN-aware aggregation on VRT mosaics whose sources used 0 as a sentinel (a common convention for unsigned remote-sensing imagery). Fix: src_nodata = src.nodata if src.nodata is not None else nodata. Five regression tests in test_vrt_source_nodata_zero_1655.py covering source NODATA=0, integer XML literal, non-zero unchanged, band-level NoDataValue=0 still honoured, and source-overrides-band precedence. All 100 vrt-related geotiff tests still pass; 3 pre-existing test_features.py matplotlib palette failures unrelated. Categories: Cat 2 (NaN propagation) + Cat 5 (backend inconsistency: read_geotiff masks 0 correctly when GDAL_NODATA tag is set; only VRT path was broken). | Pass 18 (2026-05-11): MEDIUM fixed -- issue #1642. PR #1641 (issue #1640) inherited level-0 georef on overview reads but kept the level-0 origin_x/origin_y unchanged. That is correct for PixelIsArea (origin = upper-left corner of pixel (0,0)) but wrong for PixelIsPoint (origin = center of pixel (0,0), GeoKey 1025 = 2). For a 1024x1024 PixelIsPoint COG with 10 m pixels and origin (0, 0), open_geotiff(overview_level=1) returned x[:3]=[0,20,40] instead of [5,25,45] (level-1 pixel 0 covers level-0 pixels 0-1 whose centers are 0 and 10, centroid 5); same for y. Downstream sel/interp/reproject silently snaps to the wrong pixel for any DEM-style PixelIsPoint COG (USGS, OpenTopography, Copernicus DEM). Categories: Cat 3 (off-by-one / boundary handling) + Cat 5 (raster_type-dependent backend convention). Fix: in extract_geo_info_with_overview_inheritance (_geotags.py), pick the effective raster_type first (overview-declared if non-default, otherwise inherited from parent), then when it is PixelIsPoint apply origin_shift = (scale - 1) * 0.5 * pixel_size_lvl0 along each axis before building the new GeoTransform. PixelIsArea path is byte-equivalent. 13 regression tests in test_overview_pixel_is_point_1642.py: centroid identity across all 4 backends, transform tuple across all 4 backends, uniform grid step, unit-level helper tests for both raster_types via stubbed extract_geo_info, own-geokeys-not-clobbered path on PixelIsPoint, and a PixelIsArea regression check. All 1397 existing non-network geotiff tests still pass (3 pre-existing matplotlib palette failures unrelated). Deferred LOW: non-power-of-two overview dimensions cause scale = base_w/ov_w to diverge from the true 2^level reduction (writer drops the right/bottom strip via h2=(h//2)*2; for h=1023 a level-1 overview has 511 rows so scale=2.0019 not 2.0). Fix would need to either (a) emit explicit geo tags on overview IFDs from the writer or (b) pass the level number into the inheritance helper; neither is a one-line change and the resulting coord error is sub-pixel of level 0. | Pass 17 (2026-05-11): MEDIUM fixed -- issue #1634. open_geotiff eager path windowed read produced confusing CoordinateValidationError when window extended past source extent. read_to_array clamped the window internally and returned a smaller array, but the eager code path used unclamped window indices for y/x coord generation (xrspatial/geotiff/__init__.py lines 562-572), so the coord array length differed from the data and xarray refused to construct the DataArray. Same bug affected the windowed transform shift in _populate_attrs_from_geo_info. The dask path (read_geotiff_dask) already validated up front since #1561, raising a clear ValueError with the format 'window=... is outside the source extent (HxW) or has non-positive size.' so the two backends diverged on the contract. Fix: validate the window up front in open_geotiff's eager branch via _read_geo_info (metadata-only read, no extra pixel cost) using the exact same condition the dask path uses, raising the same ValueError message format. Reproduction: 10x10 raster + window=(5,5,15,15) on eager raised CoordinateValidationError('conflicting sizes ... length 5 ... length 10'); now raises ValueError('window=(5, 5, 15, 15) is outside the source extent (10x10) or has non-positive size.'). Categories: Cat 3 (off-by-one / boundary handling) + Cat 5 (backend inconsistency). 12 regression tests in test_window_out_of_bounds_1634.py: negative start, past-right-edge, past-bottom-edge, past-both-edges, zero-size, inverted window, full-extent ok, interior subset, edge-aligned, eager-vs-dask parity, message-format parity, issue reproducer. All 1286 existing non-network geotiff tests still pass. | Pass 16 (2026-05-11): HIGH fixed -- issue #1623. to_geotiff(cog=True, overview_resampling='cubic', nodata=<finite>) on a float raster with NaN regions produced overview pixels with severe ringing artefacts near nodata borders. Same class of bug as #1613 but for the cubic branch: writer rewrites NaN to the sentinel upstream, then _block_reduce_2d(method=cubic) handed the sentinel-poisoned array straight to scipy.ndimage.zoom(order=3). The cubic spline blended the sentinel (e.g. -9999) into neighbouring cells, producing values like 1133.44, -10290.08 where the data was a constant 100. Repro on 16x16 float32 with a 4x4 NaN corner showed 18 polluted pixels in the 8x8 overview. Fix: when nodata is supplied on a float dtype and the sentinel is found, mask sentinel to NaN, run cubic with prefilter=False so a single NaN cannot poison the entire row/column (default B-spline prefilter is global), then rewrite any NaN in the result back to the sentinel. prefilter=False only fires when a sentinel is present so the non-nodata cubic semantics are unchanged. GPU side: _block_reduce_2d_gpu previously raised on method='cubic'; added a CPU fallback (same pattern as 'mode') so GPU writer produces byte-equivalent overviews. GPU_OVERVIEW_METHODS now includes 'cubic'. 12 regression tests in test_cog_cubic_overview_nodata_1623.py (helper no-ringing, poisoning repro, no-nodata unchanged, end-to-end round-trip, GPU fallback, CPU/GPU byte-match, +/-inf nodata mask, NaN-sentinel no-op, GPU_OVERVIEW_METHODS contract). All 1256 existing geotiff tests still pass (3 pre-existing matplotlib failures unrelated). | Pass 15 (2026-05-11): HIGH fixed -- issue #1613. to_geotiff(cog=True, nodata=<finite>) on a float raster with NaN produced a corrupted overview pyramid. The NaN-to-sentinel rewrite in __init__.py:1202 (CPU) and :2852 (GPU write_geotiff_gpu) ran BEFORE _make_overview / make_overview_gpu, so the nan-aware aggregations (np.nanmean/min/max/median, cupy.nanmean/min/max/median) saw the sentinel as a real number and biased every overview pixel. Reproduction with -9999 sentinel produced [[-4998.75,-4997.75],..] where np.nanmean gives [[1.5,3.5],..]. Both CPU and GPU paths affected; backend results matched each other but were both wrong (CAT 2 NaN propagation + CAT 5 documents the parity). Fix: _block_reduce_2d / _block_reduce_2d_gpu accept a nodata kwarg that masks the sentinel back to NaN for float dtypes before the reduction; the writer's overview loop passes nodata in, then rewrites all-sentinel reductions (which surface as NaN from the reducer) back to the sentinel for the on-disk pyramid. 11 regression tests in test_cog_overview_nodata_1613.py (CPU mean / partial-block / min/max/median / no-nodata passthrough / helper kwarg / all-sentinel block / GPU mean / GPU helper / CPU-GPU agreement). All 235 nodata/overview/cog tests still pass. | Pass 14 (2026-05-11): HIGH fixed -- issue #1611. read_vrt(band=None) on a multi-band integer VRT with per-band <NoDataValue> tags only masks band 0's sentinel. __init__.py lines 2795-2809 in read_vrt apply vrt.bands[0].nodata to the full ndim==3 array; bands 1+ keep their integer sentinels as literal finite values (e.g. 65000 surfaces as 65000.0 after the dtype=float64 cast, not NaN). Float-VRT path masks per-band correctly in _vrt._read_data lines 296-297 + 347-351. PR #1602 fixed the single-band band=N case for issue #1598; the band=None multi-band case is the same class of bug. Repro: 2-band uint16 VRT with NoDataValue 65535 / 65000 returns r.values[1,1,1] == 65000.0 instead of NaN; r.values[1,1,0] is NaN (band 0 sentinel masked). Fix scope: in read_vrt, when band is None, iterate over vrt.bands and mask each arr[..., i] slice against its own <NoDataValue> (gated by the same _int_nodata_in_range guard PR #1583 introduced). Severity HIGH (Cat 2 NaN propagation + Cat 5 backend inconsistency: identical input semantics produce different masking outcomes based on dtype, with finite garbage values where NaN expected). Fix in PR #1612: walks vrt.bands when band is None and ndim==3, masks each arr[..., i] slice against its own <NoDataValue> via the refactored _sentinel_for_dtype helper (reuses PR #1583's range guard so out-of-range/non-finite/fractional sentinels are a no-op). attrs['nodata'] still carries band 0's sentinel for band=None reads (documented contract). 7 regression tests in test_vrt_multiband_int_nodata_1611.py: uint16 per-band, int32 negative, mixed presence, dtype preservation when no sentinel hit, out-of-range gating, band=N non-regression, attrs contract. 135 existing vrt/nodata geotiff tests still pass. | Pass 13 (2026-05-11): HIGH fixed -- issue #1599. write_geotiff_gpu (and to_geotiff gpu=True) emitted raw NaN bytes for missing pixels even when nodata=<finite> was supplied, while the CPU writer substituted NaN with the sentinel before encoding. xrspatial-only round-trips were unaffected (the reader masks both NaN and the sentinel), but external readers (rasterio/GDAL/QGIS) that mask only on the GDAL_NODATA tag saw NaN pixels as valid data -- rasterio reported 100% valid pixels on a 25-NaN file vs CPU's 25-invalid report. Root cause: __init__.py lines 2579-2587 jumped from shape/dtype resolution straight to compression, missing the equivalent of the CPU writer's NaN-to-sentinel rewrite at to_geotiff line ~1156. Fix: cupy.isnan + masked write on a defensive copy of arr, gated on np_dtype.kind=='f' and not np.isnan(float(nodata)). Caller's CuPy buffer preserved (copy before mutate). 7 regression tests in test_gpu_writer_nan_sentinel_1599.py: substitution lands as sentinel, CPU/GPU byte-equivalent, caller buffer not mutated, no-NaN no-op, NaN sentinel skips substitution, rasterio sees identical invalid count on CPU/GPU, multiband 3D path. All other GPU writer tests still pass (50 passed across band-first, attrs, nodata, dask+cupy, writer, nodata aliases). | Pass 12 (2026-05-11): HIGH fixed -- issue #1581. Reading a uint TIFF with a negative GDAL_NODATA sentinel (e.g. uint16 + -9999) raised OverflowError on every backend because the nodata-mask code did arr.dtype.type(int(nodata)) with no range check. Three identical cast sites in __init__.py (numpy eager, _apply_nodata_mask_gpu, _delayed_read_window) plus _resolve_masked_fill and _sparse_fill_value in _reader.py. Fix: _int_nodata_in_range helper gates the cast; out-of-range sentinels are a no-op for value matching (the file can never contain that value), file dtype is preserved, attrs['nodata'] still surfaces the original sentinel so write round-trips keep the GDAL_NODATA tag intact. Matches rasterio behavior. 8 regression tests in test_nodata_out_of_range_1581.py cover the helper, both eager and dask read paths, in-range sentinel non-regression, and GPU helper (cupy-gated). | Pass 11 (2026-05-10): CLEAN. Audited the one additional commit since pass 10 -- #1559 (PR 1548, Centralise GeoTIFF attrs population across all read backends). Refactor extracts _populate_attrs_from_geo_info helper and routes eager numpy, dask, GPU stripped, GPU tiled read paths through it; before the fix dask only emitted crs/transform/raster_type/nodata while numpy emitted the full attrs set including x/y_resolution, resolution_unit, image_description, extra_samples, GDAL metadata, and the CRS-description fields. No data-path arithmetic touched; only attrs dict population. Windowed origin math (origin_x + c0*pixel_width, origin_y + r0*pixel_height) verified to produce -98.0 / 48.75 origin for window=(10,20,50,70) on a (0.1,-0.125) pixel-size raster, with PixelIsArea half-pixel offset preserved on coord lookups (-97.95, 48.6875). Cross-backend attrs parity re-verified: numpy/dask/cupy all emit identical key set on deflate+predictor3+nodata round-trip (crs, crs_wkt, nodata, transform, x_resolution, y_resolution). Data bit-parity re-verified across numpy/dask/cupy on same payload (np.array_equal with equal_nan=True). test_attrs_parity_1548.py (5 tests), test_reader.py/test_writer.py/test_dask_cupy_combined.py (25 tests), GPU orientation/predictor2-BE/LERC-mask/nodata/byteswap suites (65 tests) all green. No accuracy or backend-divergence findings. | Pass 10 (2026-05-10): CLEAN. Audited 5 recent commits: #1558 drop-defensive-copies (frombuffer path still .copy()s before in-place predictor decode at _reader.py:778), #1556 fp-predictor ngjit (writer pre-ravels so 1-D slice arg is correct, float32/64 LE+BE bit-exact), #1552 batched D2H (OOM guard fires before cupy.concatenate, host_buf offsets correct), #1551 parallel-decode gate (>= vs > sends 256x256 default to parallel path, no value diff confirmed via partial-tile parity), #1549 nvjpeg constants (gray + RGB GPU JPEG decode pixel-identical to Pillow CPU, max diff = 0). Cross-backend parity re-verified clean: numpy/dask+numpy/cupy/dask+cupy equal .data/.dtype/.coords/nodata/NaN-mask on deflate+predictor3+nodata; orientations 1-8 numpy==GPU; partial edge tiles 100x150, 257x383, 512x257 numpy==GPU==dask; predictor2 LE/BE round-trip uint8/int16/uint16/int32/uint32 pass; predictor3 LE/BE float32/64 pass. Deferred LOW (pre-existing, not opened): float16 (bps=16, SampleFormat=3) absent from tiff_dtype_to_numpy map - writer never emits, asymmetric but unreachable. | Pass 9 (2026-05-09): TWO HIGH fixed -- (a) PR #1539 closes #1537: TIFF Orientation tag 2/3/4 (mirror flips) on georeferenced files left y/x coords computed from the un-flipped transform, so xarray label lookups returned the wrong pixel even though _apply_orientation flipped the buffer. PR #1521 only updated the transform for the 5-8 axis-swap branch. Fix updates origin and pixel-scale signs along whichever axes were flipped, for both PixelIsArea (origin shifts by N*step) and PixelIsPoint (shifts by (N-1)*step). 10 new tests in test_orientation.py. (b) PR #1546 closes #1540: read_geotiff_gpu ignored Orientation tag completely; CPU correctly applied 2-8 (PR #1521) but GPU returned the raw stored buffer. Cross-backend disagreement on every non-default orientation. Fix adds _apply_orientation_gpu (cupy slicing mirror of the CPU helper) and _apply_orientation_geo_info, threads them into the tiled GPU pipeline, reuses CPU-fallback geo_info for the stripped path to avoid double-applying. 28 new tests in test_orientation_gpu.py (every orientation, single-band tiled, single-band stripped, 3-band tiled, mirror-flip sel-fidelity, default no-tag passthrough). Re-confirmed clean: HTTP coalesce_ranges with overlapping ranges and zero-length ranges, parallel streaming write thread-safety (each tile gets independent buffer via copy or padded zeros), planar=2 + chunky GPU LERC mask propagation matches CPU, IFD chain cap MAX_IFDS=256, max_z_error round-trip on tiled write, _resolve_masked_fill float vs integer dtype semantics. Deferred LOW: per-sample LERC mask (3D mask (h,w,samples)) collapsed to per-pixel ""any sample invalid"" on GPU while CPU honours per-sample; LERC implementations rarely emit 3D masks (verified: lerc.encode with 2D mask on 3-band returns 2D mask). Documented planar=2 + LERC + GPU silently drops mask (rare in practice, source comment acknowledges). | Pass 8 (2026-05-07): HIGH fixed in fix-jpeg-tiff-disable -- to_geotiff(compression='jpeg') wrote files that no external reader can decode. The writer tags compression=7 (new-style JPEG) but emits a self-contained JFIF stream per tile/strip and never writes the JPEGTables tag (347) that the TIFF spec requires for that codec. libtiff/GDAL/rasterio all reject the file with TIFFReadEncodedStrip() failed; our reader round-trips because Pillow decodes the standalone JFIF, hiding the break. Pass-4 notes flagged the read side of the same JPEGTables gap and deferred it; pass-8 covers the write side. Fix: reject compression='jpeg' at the to_geotiff entry with a clear ValueError pointing at deflate/zstd/lzw. The internal _writer.write is untouched so the existing self-decoding tests still cover the codec; re-enabling the public path needs a JPEGTables-aware encoder. PR diffs reviewed but not merged: #1512 (BytesIO source) and #1513 (LERC max_z_error) -- both look correct; #1512 file-like read path goes through read_all() once so the per-call BytesIOSource lock is theoretical, and #1513 forwards max_z_error through every overview/tile/strip/streaming path including _write_vrt_tiled and _compress_block. No regressions found in either open PR. Other surfaces audited clean: predictor=3 with float16 (writer auto-promotes to float32 on both eager and streaming paths, value-exact round-trip); planar=2 multi-tile read uses band_idx*tiles_per_band offset so no cross-contamination between planes; _header.py multi-byte tag parsing uses bo (byte_order) consistently; Pillow YCbCr-vs-tagged-RGB photometric mismatch becomes moot once JPEG is disabled. Deferred (LOW/MEDIUM, not filed): JPEG2000 writer accepts arbitrary dtype with no validation (rare codec, narrow risk); float16 dtype not in tiff_dtype_to_numpy decode map (writer never emits it - asymmetric but unreachable); Orientation tag (274) still ignored on read (pass-4 deferral). | Pass 7 (2026-05-07): HIGH fixed in fix-mmap-cache-refcount-after-replace -- _MmapCache.release() looked up the cache entry by realpath, so a holder that acquired the OLD mmap before an os.replace and released it AFTER another caller had acquired the post-replace entry would decrement the new holder's refcount. Subsequent eviction (cache full, or another acquire) closed the still-in-use mmap, breaking reads with 'mmap closed or invalid'. Real exposure: any concurrent reader/writer pattern where to_geotiff replaces a file that another reader had just opened via open_geotiff with chunks= or via _FileSource. PR #1506 added stale-replacement detection but did not fix the refcount confusion across the pop. Fix: acquire returns an opaque entry token; release takes the token and decrements that exact entry, regardless of cache state. Orphaned (popped) entries close their fh+mmap when their own refcount hits zero. _FileSource updated to pass the token. Regression test test_release_after_path_replacement_does_not_clobber_new_holder added. All 665 geotiff tests pass; GPU path verified. | Pass 6 (2026-05-07) PR #1507: BE pred2 numba TypingError. | Pass 5 (2026-05-06) PR #1506: mmap cache stale after file replace. | Pass 4 (2026-05-06) PR #1501: sparse COG tiles. | Pass 3 (2026-05-06) PR #1500: predictor=3 byte order. | Pass 2 (2026-05-05) PR #1498: predictor=2 sample-wise. | Pass 1 (2026-04-23) PR #1247. Re-confirmed clean over passes 2-7: items 2 (writer always emits LE TIFFs - hardcoded b'II'), 3 (RowsPerStrip default = height when missing), 4 (StripByteCounts missing raises clear ValueError), 5 (TileWidth without TileLength caught by 'tw <= 0 or th <= 0' check at _reader.py:688), 9 (read determinism on compressed+tiled+multiband), 11 (predictor=2 with awkward sample stride round-trips), 18 (compression_level=99 raises ValueError 'out of range for deflate (valid: 1-9)'), 21 (concurrent writes serialize correctly via mkstemp+os.replace), 24 (uint16 dtype preserved on numpy backend, dask honors chunks param), 26 (chunks rounds correctly with remainder chunk for non-tile-aligned). Deferred: item 8 (BytesIO/file-like sources are not supported, source.lower() error) - documented as 'str' parameter, not a bug; item 19 (LERC max_z_error not user-exposed by to_geotiff) - missing feature, not a bug."
 glcm,2026-05-01,1408,HIGH,2,"angle=None averaged NaN as 0, masking no-valid-pairs as zero texture; fixed via nanmean-style averaging"
@@ -27,7 +27,7 @@ normalize,2026-05-01,,,,rescale and standardize across all 4 backends. NaN/inf f
 perlin,2026-04-10T12:00:00Z,,,,Improved Perlin noise implementation correct. Fade/gradient functions verified. Backend-consistent. Continuous at cell boundaries.
 polygon_clip,2026-06-10,3186,HIGH,5,"Cat5 backend inconsistency: dask+cupy clip_polygon rasterizes the mask with a uniform chunk size from the raster's first chunk, then feeds raster+mask to da.map_blocks (positional block pairing). Non-uniform raster chunks gave the mask a different block layout -> IndexError/ValueError (or silent mis-stamp). Repro (8,6) rechunk ((3,5),(6,)) on dask+cupy raised ValueError Shapes do not align; dask+numpy was fine via xarray.where rechunk. Fix #3186/PR: rechunk cond to raster.data.chunks[-2:] before map_blocks; added non-uniform regression tests for dask+numpy and dask+cupy. use_cuda->gpu migration in that branch was already landed by #3089/#3122. CUDA available; cupy+dask+cupy verified, 25 tests pass. Cats 1-4 clean: numpy path uses raster.where, cupy path operates on raw arrays, NaN inputs preserved, no neighborhood ops/curvature. Prior fix #1197/#1200 (crop+all_touched) merged and unrelated."
 polygonize,2026-05-29,2606,HIGH,5,"Cat 5 HIGH: dask connectivity=8 cross-chunk merge filled diagonal notch where same-value regions meet only at a corner across a chunk boundary; total area exceeded raster. Hole ring was dropped because containment tested hole[0] (on exterior at pinch). Fixed via _ring_interior_point in PR for #2606. numpy, dask+numpy, dask+cupy area parity now holds; 4-conn was already correct. cupy + dask+cupy paths validated on GPU host. Other cats clean: NaN masked on numpy/cupy float paths (tested), _is_close handles +/-inf via exact-equality short-circuit, atol/rtol/simplify_tolerance reject NaN/inf, integer GPU CCL matches numpy."
-proximity,2026-06-18,3389,MEDIUM,1;5,"Cat1/Cat5 (MEDIUM, fixed #3389): CPU brute-force _distance casts to float32 before best_dist<=max_distance, but the CUDA kernel compared the float64 distance; a max_distance in a target's float32 ulp gap qualified it on numpy/dask+numpy yet rejected it on cupy/dask+cupy (GREAT_CIRCLE proximity + EUCLIDEAN/MANHATTAN allocation/direction). Fixed by rounding best_dist to float32 in the kernel before the range test. CUDA host: 518 tests pass incl cupy + dask+cupy. Out of scope (separate pre-existing edge, predates fix): bounded dask EUCLIDEAN allocation can still NaN a target exactly at a float32-ulp max_distance via the halo-sizing path. Prior #3108 fix (antimeridian halo) still verified. LOW (unfixed): great_circle_distance uses WGS84 equatorial radius 6378137 as sphere radius (~0.1% vs mean-radius), but documented and param-exposed."
+proximity,2026-06-22,3442,MEDIUM,5,"Cat5 backend divergence: EUCLIDEAN proximity/allocation/direction on numpy and dask+numpy use cKDTree, which dropped a target sitting exactly at max_distance to NaN (clearest at max_distance=0: a target pixel is distance 0 from itself), while cupy/dask+cupy brute-force kept it (0.0 / target value / 0). Root cause: cKDTree squares its exclusive distance_upper_bound for p=2, and the old np.nextafter(0,inf) widen underflows when squared, collapsing the bound back to max_distance; _kdtree_query_lowest_index passed the raw exclusive bound with no widen at all. Fix #3442/PR #3443: route all 3 cKDTree call sites through one _inclusive_upper_bound helper (relative eps + absolute floor, robust for p=1 and p=2). MANHATTAN (p=1) was unaffected. Verified all 4 backends agree; 543 proximity tests pass incl. 25 new cross-backend regression cases. CUDA available; cupy + dask+cupy executed. Cats 1-4 clean: float32/float64 boundary already handled (#3389), antimeridian halo (#3108), irregular-coord halo (#2908), non-finite target_values rejected (#2868), non-monotonic coords rejected (#2875), tie-break row-major across backends (#3090/#2881)."
 rasterize,2026-06-18,3384,HIGH,1;5,dask all_touched polygon-boundary supercover walk re-extracted per tile in tile-local float pixel coords; floor() tie on an on-grid boundary segment diverged from eager (Cat1 precision -> Cat5 backend split). Fixed by extracting boundary float segments in the global grid frame and shifting by integer tile offset (#3384). Verified numpy/cupy/dask+numpy/dask+cupy.
 reproject,2026-06-12,3274,HIGH,1;4,"3 confirmed bugs, all kernel-vs-PROJ parity: #3274 HIGH LAEA inverse spurious /rq (2.6 km err for 3035) + _authalic_apa inverse series wrong (4.8 m in AEA/CEA inverses; PROJ 3-term = 1.6 mm), CPU+CUDA kernels both; #3275 HIGH _is_wgs84_compatible_ellipsoid passes R-defined spheres (MODIS sinusoidal 18.9 km err) and _aea_params/_cea_params lack the guard entirely (23.8 km on spherical aea/cea); #3276 MEDIUM itrf helmert scale 1e-9 but PROJ +s is ppm (1e-6), ~23 mm err. Verified clean: merc/emerc/UTM/tmerc/LCC/polar stere (incl lat_ts akm1) forward+inverse <=1e-5 m vs pyproj; resampling kernels NaN handling and GDAL renorm match across numpy/cupy/dask (CUDA run, gpu-vs-cpu 1.3e-7); dask footprint chunk-skip bbox is a superset in all probed cases (no holes). LOW (documented only): _source_footprint_in_target probe array typo uses x-midpoint mx as a latitude in last 3 ys entries (bbox superset, correctness unaffected)."
 resample,2026-05-29,2610,HIGH,3;5,"dask interp (nearest/bilinear) overlap depth=1 too small on downsample; block-centered source coord landed past chunk, map_coordinates clamped to edge -> wrong seam rows. Fixed PR #2627 via per-axis _downsample_radius. cupy+dask+cupy verified."

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-api-consistency-state.csv

@@ -7,6 +7,7 @@ interpolate,2026-06-12,3285,MEDIUM,2,"Sweep 2026-06-12 (deep-sweep-api-consisten
 mcda,2026-06-10,3148,HIGH,1;2;3;5,"Sweep 2026-06-10 (deep-sweep-api-consistency-mcda-2026-06-10). Fixed in this branch (#3148): (HIGH Cat 1) owa() named its criterion-weight dict criterion_weights while wlc/wpm/sensitivity use weights (same semantics, same _validate_weights); renamed to weights with keyword-only criterion_weights deprecation shim (DeprecationWarning; both names -> TypeError; positional callers untouched). (MEDIUM Cat 2) boolean_overlay annotated criteria as dict-only while every sibling combiner takes xr.Dataset; Dataset already worked via the Mapping interface -- now annotated/documented as xr.Dataset | dict. (MEDIUM Cat 3) ahp_weights docstring Raises claimed ValueError on incomplete comparisons but code warns (UserWarning) and defaults missing pairs to 1 -- docstring now documents Warns behaviour. (MEDIUM Cat 5) ConsistencyResult returned by public ahp_weights but absent from xrspatial/mcda __all__ and docs/source/reference/mcda.rst -- exported and documented. Documented, NOT fixed here: (MEDIUM Cat 2, deferred to parallel sweep-metadata sibling to avoid duplicate PR) constrain() drops attrs via xr.where while the other nine public functions preserve them. (LOW Cat 2) ahp_weights returns (weights, ConsistencyResult) tuple vs rank_weights bare dict -- intentional, documented in both docstrings, no fix. (LOW Cat 4) name=None inherit-input-name (standardize/constrain) vs literal-name defaults (combiners) -- defensible split, document only. Pre-existing backend bugs surfaced by the mandated cupy smoke (accuracy/test-coverage lane, recorded in #3148 body): owa fails on cupy (numpy order-weights array mixed into cupy multiply, combine.py ~336-340) and on ANY dask backend at graph construction (da.sort does not exist, combine.py:356, despite the owa MemoryError message recommending dask); sensitivity(method=monte_carlo) fails on cupy (template.values implicit-conversion guard). constrain on cupy blocked by the known library-wide cupy 13.6 + xarray xr.where astype incompat (dependency-pin issue), not mcda-specific. cuda-validated: CUDA_AVAILABLE=True; all 10 public functions smoke-tested on cupy DataArrays; owa weights=/criterion_weights= shim verified on numpy AND cupy entry points (cupy execution stops at the pre-existing mixed-array bug, signature acceptance confirmed)."
 morphology,2026-06-20,3399,MEDIUM,5,"Sweep 2026-06-20 (deep-sweep-api-consistency-morphology-2026-06-20). 1 MEDIUM Cat 5 finding filed as #3399, fixed in this branch (PR #3409). Cat 5 accessor-parity gap: all 7 public morphology functions are exported in xrspatial/__init__.py, documented, and tested, but the .xrs accessor exposed only morph_erode/dilate/opening/closing on both XrsSpatialDataArrayAccessor and XrsSpatialDatasetAccessor; morph_gradient/white_tophat/black_tophat were missing (da.xrs.morph_gradient -> AttributeError while da.xrs.morph_erode works). Root cause: base 4 ops landed #949 + accessors #1042; derived 3 ops landed later #1026 and the accessor was never updated. Fix adds the 3 forwarding methods to both accessor classes (mechanical, matches existing pattern) plus accessor tests (none existed for morph before) and guards all 7 method names in the expected-methods checks. Clean elsewhere: Cat 1 no in-module naming drift -- all 7 publics share the exact signature (agg, kernel=None, boundary='nan', name='<op>'), verified programmatically; matches kernel-op siblings convolution_2d/focal.apply/hotspots which also use agg/kernel/boundary/name. Cat 2 no return drift (all 7 return xr.DataArray with coords/dims/attrs preserved via _dispatch / @supports_dataset). Cat 3 docstring/signature parity exact on all 7 (every param documented, Returns sections all DataArray). Cat 4 no default drift (kernel=None->3x3 ones, boundary='nan' uniform across all 7). Cross-cutting, notes only per template: convolution_2d orders name before boundary while morphology orders boundary before name (both keyword-defaulted, cosmetic); focal.apply/hotspots accept a keyword-only raster= alias for agg but that alias is focal-only (not a library convention) so morphology lacking it is not drift; library-wide first-arg agg vs raster drift spans 20+ modules, out of per-module scope. cuda-validated: CUDA_AVAILABLE=True on this host; all 7 publics smoke-tested with identical kwargs on numpy AND cupy DataArrays (shape parity, no signature drift between numpy/cupy entry points). PR reviewed (COMMENTED), no findings; branch merged with origin/main (clean), left BLOCKED on REVIEW_REQUIRED for user merge."
 multispectral,2026-06-20,3433,MEDIUM,3,"Sweep 2026-06-20 (deep-sweep-api-consistency-multispectral). 18 public funcs, all single 2D DataArray returns except true_color (3D composite, inherently different). (#3433 MEDIUM Cat 3) nbr() docstring documented swir_agg but signature param is swir2_agg; copying the name from docs raises TypeError. Docs-only rename to swir2_agg + guard test test_docstring_params_match_signature over the 17 index funcs. No deprecation needed. LOW, documented not fixed: (Cat 5) gci/nbr2/ndmi/true_color/ebbi are NOT re-exported in xrspatial/__init__.py while 13 module siblings are; not an orphan API since the canonical documented path is xrspatial.multispectral.X (all 18 in multispectral.rst) and tests import from the submodule -- convenience-surface gap only. (Cat 3 LOW) only savi types name: str; true_color lacks band type hints and uses r/g/b (conventional for an RGB composite). Cross-module note (not filed per template): multispectral has no -> xr.DataArray return annotations while sibling fire.py annotates all 7 -- library-wide convention drift. No Cat 1 in-module (all band inputs are <band>_agg), no Cat 2 (return shapes consistent), no Cat 4 (no mutable defaults; soil_factor=1.0 consistent evi/savi). CUDA available: numpy+cupy smoke-tested, signatures parity-clean, full suite 171 passed."
+perlin,2026-06-23,3465,MEDIUM,3,"Sweep 2026-06-23 (deep-sweep-api-consistency-perlin-2026-06-23). 1 MEDIUM Cat 3 finding filed as #3465, fixed in PR #3468. perlin() signature has name: str = 'perlin' (perlin.py:288) but the docstring Parameters section documented only agg/freq/seed, omitting name; sibling generate_terrain() documents its identical name param (terrain.py:607). Fix adds 'name : str, default=""perlin"" / Name for the output DataArray.' matching the terrain style. Pure docstring fix, no signature/behavior change, no deprecation shim needed. Tests: test_perlin_name_param (default+custom name honored), test_perlin_docstring_documents_name (regression). 14/14 test_perlin.py pass incl GPU paths. No HIGH/CRITICAL. Cat 1 clean: agg/seed/name consistent with generate_terrain. Cat 2 clean: all generators return xr.DataArray. Cat 4: seed default 5 (perlin) vs 10 (terrain) is a per-generator domain default, not surprise drift. Cat 5 clean: perlin in __init__.py. cuda-validated: CUDA_AVAILABLE=True on this host, numpy and cupy entry points both accept/propagate name. Cross-cutting, notes only: bump() exposes no name param while perlin/generate_terrain both do -- adding name to bump is a separate bump-module change, out of per-module scope."
 polygonize,2026-06-12,3306;3307,MEDIUM,1;3,"Re-sweep 2026-06-12 (deep-sweep-api-consistency-polygonize-2026-06-12); prior pass 2026-05-19 (#2148). 2 MEDIUM findings filed and fixed on branches -01/-02 off this one. (#3306, MEDIUM Cat 3, branch -01) column_name docstring says 'Only used if return_type is geopandas or spatialpandas' but _to_geojson also consumes it as the per-feature property key (verified: properties={'myval': 1}); docs-only fix + test pinning geojson property naming. (#3307, MEDIUM sibling-behavior drift, branch -02) return_type is the only polygonize parameter validated AFTER the computation: invalid value runs the full backend (spy-verified 1 invocation before raise) while sibling contours() validates up front and lists allowed values; fix hoists the check into the top validation block with an allowed-values message (existing test matches on prefix, unaffected). Re-confirmed prior dispositions, still documented-only per cross-module rule: (HIGH Cat 1 cross-module) connectivity (polygonize, matches GDAL/rasterio/skimage) vs neighborhood (sieve.py, zonal.regions) for the identical 4|8 rook/queen concept -- rename shim belongs in sieve/zonal, out of polygonize scope; (LOW Cat 1 cross-cutting) raster (polygonize/sieve/clip_polygon) vs agg (contours/terrain family) first-arg drift, library-wide, not filed per-module. No new Cat 2 (return_type dispatch shapes match docstring Returns section exactly); no Cat 4 (atol/rtol mirror numpy.isclose, connectivity=4 == sieve neighborhood=4); Cat 5 LOW documented-only: module has no __all__ and the non-underscore internals generated_jit + Turn leak via import-star; polygonize re-exported in __init__.py and accessor, no orphan API. Docstring/signature parity otherwise exact (all 10 params documented, all annotated). Open polygonize issues #3292/#3293 checked -- no overlap with these findings. cuda-validated: CUDA_AVAILABLE=True on this host; polygonize smoke-tested with identical full kwargs on numpy, cupy (int + float atol/rtol=0), and dask+cupy; no backend signature drift."
 proximity,2026-06-09,3090;3091,HIGH,2;3,"Sweep 2026-06-09 (deep-sweep-api-consistency-proximity-2026-06-09). 1 HIGH Cat 2 finding (#3090): dask+numpy (and unbounded dask+cupy, which converts to it) KDTree path violates the documented lowest-flat-index tie-break in allocation()/direction() whenever the raster has >1 chunk column. _collect_region_targets concatenates targets chunk-major (iy outer, ix inner) so the tree's target order is not global row-major; _kdtree_query_lowest_index then ties to the wrong target. Existing tie-break tests put both targets in the same raster row where chunk order coincides with row-major, so they pass. Repro: 5x5, targets 2@(1,3) and 3@(2,2), chunks (5,3), pixel (2,3) tied at d=1 -> numpy gives 2, dask gives 3. Bounded map_overlap paths are fine (local row-major order is offset-invariant). 1 MEDIUM Cat 3 finding (#3091): all 3 public docstrings claim numpy + dask+numpy support only while cupy/dask+cupy backends exist, are dispatched, and are tested (the tie-break paragraphs in the same docstrings name all 4 backends); direction() opens with a stray copy-pasted slope line ('downward slope direction') plus a doubled 'the the'; allocation example output reads as float64 but the function returns float32; stale '# convert to have same type as of input @raster' comment. Within-module Cat 1/4/5 clean: proximity/allocation/direction share an identical signature (raster, x='x', y='y', target_values=None, max_distance=np.inf, distance_metric='EUCLIDEAN'); consistent with surface_distance siblings (raster/x/y/target_values/max_distance); all 6 public symbols (incl. euclidean/manhattan/great_circle_distance) re-exported in __init__.py, no orphan API. Cross-cutting, documented not filed: sibling distance modules (surface_distance, cost_distance, balanced_allocation) use mutable default target_values: list = [] while proximity uses the None sentinel - the mutable-default fix belongs to those modules; proximity's target_values: list = None hint would be more precise as Optional[list] (LOW, matches library style). cuda-validated: CUDA_AVAILABLE=True on this host; proximity/allocation/direction smoke-tested with identical kwargs on numpy, cupy, dask+numpy, dask+cupy (proximity parity passed; allocation/direction parity failure is finding #3090)."
 rasterize,2026-06-09,3089,HIGH,1,"Sweep 2026-06-09 (deep-sweep-api-consistency-rasterize-2026-06-09). 1 HIGH Cat 1 fixed in this branch (#3089): rasterize(use_cuda=) vs open_geotiff(gpu=) named the identical GPU-backend opt-in differently; these are the only two public entry points with an explicit GPU boolean (no input array to dispatch on; both pair it with chunks= for dask) and both names were live in the public API at once. Fix renames the positional param to gpu (same slot, positional callers unaffected) and appends use_cuda=None as a deprecated alias: DeprecationWarning on use, TypeError when combined with gpu=True. Docstring, GPU merge warning text, CuPy ImportError text, and polygon_clip.py's internal dask+cupy caller updated (guarded so a legacy use_cuda in rasterize_kw does not collide with the new default); all rasterize test call sites migrated to gpu=; regression tests in test_rasterize_gpu_alias_3089.py pin slot position, warning, TypeError, backend parity, and the warning-free clip_polygon path. Re-inspection after the 2026-05-21 pass (#2250); prior cross-module notes (clip_polygon nodata vs fill, name default drift, polygonize column_name vs column) still documented-only. Docstring/signature parity verified programmatically (17/17 params, order matches). New params since last pass (check_crs, max_pixels) consistent with geotiff naming (max_pixels matches geotiff's). No Cat 2/4/5 findings. LOW noted, not fixed (other module's docs): docs/source/user_guide/focal.ipynb claims convolve_2d takes use_cuda, which it does not. cuda-validated: CUDA_AVAILABLE=True; numpy/cupy/dask+numpy/dask+cupy smoke-tested with identical kwargs, values equal."
@@ -14,4 +15,4 @@ reproject,2026-06-09,3095;3097,HIGH,1;2;3,"Sweep 2026-06-09 (deep-sweep-api-cons
 resample,2026-05-27,2544,MEDIUM,3,"Sweep 2026-05-27 (deep-sweep-api-consistency-resample-2026-05-27). 1 MEDIUM Cat 3 finding fixed in this branch (#2544): resample() was the only public symbol in xrspatial.resample without type annotations on any parameter or return; siblings slope/aspect/hillshade/curvature all annotate `agg: xr.DataArray` and `-> xr.DataArray`. Fix adds annotations matching the docstring (agg: xr.DataArray; scale_factor / target_resolution: float | tuple[float, float] | None; method: str; nodata: float | None; name: str) and a `-> xr.DataArray` return type, plus a docstring note that the @supports_dataset decorator accepts Dataset too. Regression test test_resample_signature_annot_2544.py pins every param and the return annotation. Other findings documented but not filed per template: (MEDIUM Cat 1 cross-module) `method` (resample) vs `resampling` (reproject/merge) -- same conceptual parameter, different name, cross-cutting rename, needs design issue. (LOW Cat 1 cross-cutting) first-arg `agg` (resample/slope/aspect/...) vs `raster` (reproject/rasterize/polygonize/sieve) -- library-wide drift, not per-module. (LOW Cat 5) ALL_METHODS imported by tests but not in __all__ (module has no __all__); borderline orphan but used for test parametrisation only. No Cat 2 (returns xr.DataArray as documented). No Cat 4 mutable defaults. resample is exported in xrspatial/__init__.py. cuda-validated: cupy backend smoke-tested with nearest, bilinear, and average on host with CUDA_AVAILABLE=True."
 slope,2026-05-29,2681,MEDIUM,3,"Sweep 2026-05-29 (deep-sweep-api-consistency-slope-2026-05-29). 1 MEDIUM Cat 3 finding fixed in this branch (#2681, PR #2687): slope() annotated name as `str` while every terrain-family sibling (aspect/northness/eastness in aspect.py, curvature in curvature.py) uses Optional[str]. name flows into xr.DataArray(name=name) which accepts None, so slope(agg, name=None) already worked at runtime -- the annotation was just wrong and inconsistent. Fix widens to Optional[str] and imports Optional (module previously imported only Union). Non-breaking (type-hint widening), no deprecation shim. Added test_name_annotation_matches_terrain_family (pins parity vs the 4 siblings via get_type_hints, unwrapping @supports_dataset) and test_name_none_accepted (slope(agg, name=None).name is None). Full test_slope.py passes (43). No backend logic touched -- numpy/cupy/dask+numpy/dask+cupy paths unchanged; public signature is shared across backends via ArrayTypeFunctionMapping. Other categories: no Cat 1 in-module rename (slope/aspect share identical public param names agg/name/method/z_unit/boundary); no Cat 2 return drift (returns xr.DataArray/Dataset via @supports_dataset, same coords/dims/attrs convention as siblings); no Cat 4 default drift (name/method='planar'/z_unit='meter'/boundary='nan' match across the family); no Cat 5 orphan API (slope re-exported in __init__.py, documented, no __all__ but consistent with module convention). Cross-cutting (documented, not filed per template): first-arg `agg` (slope/aspect/curvature) vs `raster` (reproject/rasterize/polygonize) is library-wide drift. cuda-validated: CUDA_AVAILABLE=True on this host; cupy slope smoke-tested (planar) and signature parity confirmed between numpy and cupy entry points."
 visibility,2026-06-10,3183,MEDIUM,3;5,"Sweep 2026-06-10 (deep-sweep-api-consistency-visibility-2026-06-10). 2 MEDIUM findings filed as issue #3183, fixed in this branch. (MEDIUM Cat 5) output-name convention drift: viewshed() sets a fixed output name and exposes name=, but cumulative_viewshed (visibility.py:289) and visibility_frequency built/returned DataArrays with name=None. Fix adds name='cumulative_viewshed'/'visibility_frequency' params (Optional[str]) and sets result.name; additive, non-breaking, no shim. coords/attrs were already preserved on both. (MEDIUM Cat 3) line_of_sight (visibility.py:162) annotated frequency_mhz: float = None; default contradicts the float hint and the docstring already says optional. Fix -> Optional[float] (imported typing.Optional). Tests added: cumulative/frequency default+custom name. No Cat 1 naming drift: observer_elev/target_elev/max_distance/x/y and the x0/y0/x1/y1 two-point extension match viewshed and the observers dict keys. No Cat 2 arbitrary return drift: line_of_sight -> Dataset fits its per-sample multi-variable result; the two cumulative funcs -> DataArray like viewshed. No Cat 4 default drift (observer_elev=0/target_elev=0/max_distance=None match). No Cat 5 orphan API: all 3 funcs re-exported in __init__.py; no __all__ but consistent with module convention. cuda-validated: CUDA_AVAILABLE=True on this host; cupy entry points accept the new name= kwarg and the line_of_sight Optional hint. PRE-EXISTING backend bug (out of scope, not an api-consistency issue, NOT filed here): cumulative_viewshed on a cupy raster raises TypeError 'Unsupported type numpy.ndarray' in the count + (vs_data != INVISIBLE) accumulation (numpy accumulator vs cupy viewshed result); reproduced on origin/main without this branch's changes -- a backend-parity gap for a future backend-parity sweep."
-zonal,2026-06-10,3188,MEDIUM,1;3;5,"Re-sweep 2026-06-10 (deep-sweep-api-consistency-zonal-2026-06-10). Prior sweep's HIGH zones_ids/zone_ids typo confirmed already fixed on main (#2521). Several previously-documented MEDIUM Cat 3 items also fixed on main since 2026-05-27: crosstab layer docstring now says default=None; hypsometric_integral now has param+return annotations; apply now has -> xr.DataArray. Two remaining safe Cat 3 fixes filed+PR'd this run (issue #3188 / PR #3196): (1) crosstab zone_ids/cat_ids annotated List[...]=None -> wrapped in Optional[...] to match stats()/crop(); (2) crosstab nodata_values docstring said 'Cells with nodata' (copy-paste from apply) -> now references nodata_values. Non-breaking, 17 crosstab tests pass. Documented-not-fixed: (MEDIUM Cat 1) nodata vs nodata_values drift across stats/crosstab (nodata_values, default None, filters VALUES raster) vs apply/hypsometric_integral (nodata, default 0, filters ZONES raster) -- names differ but so do the concepts and defaults, so a blanket rename would conflate two distinct meanings; needs a design decision, not a mechanical shim. (MEDIUM Cat 5) get_full_extent has a public-style docstring+example but is not in __init__.py -- borderline orphan, minor utility, left as-is. (LOW Cat 3) crop() lacks a return type annotation while stats/crosstab/apply/regions/trim annotate theirs. Cross-cutting (not filed): first-arg name varies (stats/crosstab/crop use zones; regions/trim use raster) but regions/trim operate on the raster itself so the name matches the role; library-wide agg vs raster vs values naming spans 20+ modules, out of per-module scope. cuda-validated: CUDA_AVAILABLE=True on this host."
+zonal,2026-06-10,3188,MEDIUM,1;3;5,"Re-sweep 2026-06-10 (deep-sweep-api-consistency-zonal-2026-06-10). Prior sweep's HIGH zones_ids/zone_ids typo confirmed already fixed on main (#2521). Several previously-documented MEDIUM Cat 3 items also fixed on main since 2026-05-27: crosstab layer docstring now says default=None; hypsometric_integral now has param+return annotations; apply now has -> xr.DataArray. Two remaining safe Cat 3 fixes filed+PR'd this run (issue #3188 / PR #3196): (1) crosstab zone_ids/cat_ids annotated List[...]=None -> wrapped in Optional[...] to match stats()/crop(); (2) crosstab nodata_values docstring said 'Cells with nodata' (copy-paste from apply) -> now references nodata_values. Non-breaking, 17 crosstab tests pass. Documented-not-fixed: (MEDIUM Cat 1) nodata vs nodata_values drift across stats/crosstab (nodata_values, default None, filters VALUES raster) vs apply/hypsometric_integral (nodata, default 0, filters ZONES raster) -- names differ but so do the concepts and defaults, so a blanket rename would conflate two distinct meanings; needs a design decision, not a mechanical shim. (MEDIUM Cat 5) get_full_extent has a public-style docstring+example but is not in __init__.py -- borderline orphan, minor utility, left as-is. (LOW Cat 3) crop() lacks a return type annotation while stats/crosstab/apply/regions/trim annotate theirs. Cross-cutting (not filed): first-arg name varies (stats/crosstab/crop use zones; regions/trim use raster) but regions/trim operate on the raster itself so the name matches the role; library-wide agg vs raster vs values naming spans 20+ modules, out of per-module scope. cuda-validated: CUDA_AVAILABLE=True on this host."

xarray_spatial-0.10.13/.claude/sweep-documentation-state.csv

@@ -0,0 +1,6 @@
+module,last_inspected,issue,severity_max,categories_found,notes,doc_coverage
+flood,2026-06-25,,HIGH,1;4;5,"Cat4 HIGH: vegetation_roughness, vegetation_curve_number, flood_depth_vegetation public but absent from reference/flood.rst; Cat1 MEDIUM: no Examples on any of 7 public funcs; Cat5 MEDIUM: backend support undocumented (all 4 backends) + NaN propagation undocumented for curve_number_runoff/travel_time. Fixed in deep-sweep-documentation-flood-2026-06-25: added 3 rst entries, Examples+backend Notes to all 7 funcs (examples executed OK on CUDA host), NaN notes. PR #3502 opened with the fix; gh issue create blocked by auto-mode classifier so no issue number.",7/7
+classify,2026-06-25,3506,MEDIUM,1;3,"Cat3: reclassify (numpy/dask/cupy blocks) + equal_interval example outputs were stale/wrong, binary used np.nan in array repr; corrected to actual output (tests confirm code is correct). Cat1: added missing Examples to std_mean, head_tail_breaks, percentiles, maximum_breaks, box_plot. Fixed in deep-sweep-documentation-classify-2026-06-25 (PR for #3506). Cat2 natural_breaks num_sample-None omission already tracked in #3501 (left alone). All 10 public funcs listed in reference/classification.rst (no Cat4 gap). CUDA available: ran numpy examples; cupy/dask reprs reviewed statically.",10/10
+geotiff,2026-06-25,,MEDIUM,1,"to_geotiff (public write entry point) had Parameters/Returns/Raises but no Examples section while open_geotiff does (Cat1 MEDIUM); added Examples block (plain GeoTIFF, cog=True, .vrt mosaic) modeled on open_geotiff; fixed on deep-sweep-documentation-geotiff-2026-06-25; repo issues disabled so no issue number. Cat2/3/4/5 clean: open_geotiff/to_geotiff signature-docstring parity locked by parity/test_signature_contract.py + write/test_bigtiff.py + test_polish.py; both funcs in autosummary in reference/geotiff.rst; reference page mirrors SUPPORTED_FEATURES tiers (tier-parity gate); CUDA available, all docstring examples are +SKIP illustrative only",2/2
+fire,2026-06-25,,MEDIUM,1;5,"all 7 public funcs (dnbr, rdnbr, burn_severity_class, fireline_intensity, flame_length, rate_of_spread, kbdi) lacked Examples section (Cat1 MEDIUM) and backend-support note (Cat5 MEDIUM); fixed in deep-sweep-documentation-fire-2026-06-25-01; repo issues disabled so no issue number; examples run and outputs match numpy backend; all 7 listed in reference/fire.rst; no Cat2/Cat3/Cat4 issues",7/7
+perlin,2026-06-23,,MEDIUM,2;5,"name param undocumented (Cat2) + float-dtype requirement/ValueError undocumented, no Raises section (Cat5); fixed in deep-sweep-documentation-perlin-2026-06-23; repo has issues disabled so no issue number; example runs and output matches; 1 public func (perlin) listed in reference/surface.rst",1/1

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-metadata-state.csv

@@ -1,12 +1,15 @@
 module,last_inspected,issue,severity_max,categories_found,notes
 aspect,2026-05-29,2682,MEDIUM,4;5,"Audited 2026-05-29 (agent-a3b7c82e34312ffcb worktree, branch deep-sweep-metadata-aspect-2026-05-29). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live for aspect/northness/eastness across planar and geodesic methods. Cat 1 attrs, Cat 2 coords, Cat 3 dims, and .name all preserved correctly on every backend: the 3 public functions re-emit coords=agg.coords, dims=agg.dims, attrs=agg.attrs at the xr.DataArray constructor. NEW MEDIUM finding #2682 (Cat 4 + Cat 5): the planar dask backends (_run_dask_numpy, _run_dask_cupy) called map_overlap with a default-dtype meta (np.array(()) / cupy.array(())), so the lazy DataArray advertised float64 while the chunk functions _cpu / _run_cupy cast to and return float32. numpy and cupy backends already reported float32, and the geodesic dask paths already passed dtype=np.float32, so only the two planar dask paths were inconsistent: a backend-inconsistent metadata bug where agg.dtype differs by backend and silently flips float64->float32 on .compute(). Fix in PR #2741: pass dtype=np.float32 / dtype=cupy.float32 to the planar dask meta. northness/eastness derive from aspect so they inherit the corrected dtype. 5 new tests (test_dask_numpy_advertised_dtype_matches_computed parametrized over 4 boundary modes, plus test_dask_cupy_advertised_dtype_matches_computed) assert lazy dtype == computed dtype == float32. Full aspect suite 69 passed. slope.py and curvature.py share the same default-dtype meta pattern on their planar dask paths (out of scope for this aspect-only sweep; likely same inconsistency). No CRITICAL/HIGH/LOW findings."
+classify,2026-06-25,3508,MEDIUM,4;5,"Audited 2026-06-25 (agent-a5f16f6137723fc77 worktree, branch deep-sweep-metadata-classify-2026-06-25). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live. All 10 public classifiers (binary/reclassify/quantile/natural_breaks/equal_interval/std_mean/head_tail_breaks/percentiles/maximum_breaks/box_plot) re-emit name=, dims=agg.dims, coords=agg.coords, attrs=agg.attrs at the xr.DataArray constructor, so Cat 1 attrs (res/crs/transform/nodatavals), Cat 2 coords (values+dtype), Cat 3 dims, and .name all preserved and identical across the 4 backends. NEW MEDIUM finding #3508 (Cat 4 + Cat 5): binary() output dtype differed by backend -- _cpu_binary allocated dtype=data.dtype so numpy/dask+numpy returned the input dtype while _run_cupy_binary used dtype='f4'. The docstring documents float32 and every other classifier emits float32 via _bin/_cpu_bin; binary was the only outlier, and for integer input the numpy path returned an integer dtype that can't hold the NaN sentinel. The _cpu_binary float32 fix + verify_dtype=True backend tests + a float64/float32/int32 dtype test landed on main via the duplicate accuracy-sweep PR #3514; PR #3513 was then rebased onto that and is now scoped to the remaining piece: _run_dask_cupy_binary passed an untyped meta=cupy.array(()) (float64) so the lazy dask+cupy array advertised float64 while computing float32 -- the same advertised-vs-computed mismatch class as aspect #2682 / focal #3217. #3513 types the meta as cupy.array((), dtype='f4') and asserts the lazy dtype in test_binary_dask_cupy. Full classify suite passes, GPU paths run live. The sibling classifiers' dask+cupy helpers (_run_dask_cupy_bin and friends) share the same untyped meta and likely the same latent lazy-dtype mismatch (out of scope, follow-up). Cat 4 nodatavals-vs-NaN is the library-wide attrs=agg.attrs convention, not classify-specific (documented, not fixed). No CRITICAL/HIGH/LOW findings."
 contour,2026-05-29,2700,HIGH,1;5,"Audited 2026-05-29 (agent-ab7fff484a8f57de2 worktree, branch deep-sweep-metadata-contour-2026-05-29). CUDA available; cupy and dask+cupy paths exercised live. contours() returns a list of (level, ndarray) tuples or a GeoDataFrame, not a DataArray, so Cat 2/3 DataArray checks reinterpreted as coordinate-transform + CRS propagation. Coordinate transform (np.interp over input dims, descending y respected) is correct and identical across all 4 backends (tracing is host-side via _contours_numpy). Cat 4 N/A: library convention is NaN-as-nodata; slope/aspect/curvature/focal do not read attrs['nodatavals'] either, so contour not reading it is consistent, not a bug. NEW HIGH finding #2700 (Cat 1/Cat 5): contours(return_type='geopandas') crashed with 'Assigning CRS to a GeoDataFrame without a geometry column is not supported' whenever the input had attrs['crs'] but the result was empty (flat raster, levels outside data range) because _to_geopandas built gpd.GeoDataFrame([], crs=crs) with no geometry column; separately the all-NaN early-return passed crs=None and silently dropped the CRS. Fix (PR #2708): _to_geopandas builds an empty frame with an explicit geometry column so the CRS attaches; all-NaN early-return forwards agg.attrs['crs']. Both empty paths now return a well-formed empty GeoDataFrame carrying the CRS. 4 new tests in TestGeoDataFrame cover populated-CRS, empty-with-CRS, all-NaN-with-CRS, and empty-without-CRS. Full contour suite 28 passed. numpy-return path emits no DataArray attrs by design (list of tuples)."
+corridor,2026-06-22,3446,HIGH,1;5,"Audited 2026-06-22 (agent-a8b2674b815bdfa3f worktree, branch deep-sweep-metadata-corridor-2026-06-22). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live end-to-end for least_cost_corridor across single/threshold/relative/unreachable/pairwise paths. Cat 2 coords (x/y values + float64 dtype) and Cat 3 dims (y,x) preserved on every backend: they flow through cost_distance (coords=raster.coords, dims=raster.dims) and survive xarray's binary intersection. NEW HIGH finding #3446 (Cat 1 + Cat 5): the corridor is cd_a + cd_b where each cost-distance surface carries its SOURCE raster's attrs (cost_distance copies attrs from the source, not friction). xarray's default keep_attrs on binary + keeps only attrs present-and-equal in both operands, so when the source masks are plain marker rasters with no geo-attrs (the common case) the corridor came back with attrs=={} even though the friction surface that defines the grid had res/crs/transform/nodatavals; a downstream slope/clip on the corridor silently lost cellsize/CRS. Secondary Cat 5: .name was None whenever the two sources had different names (cost_distance renames each surface to its source .name; summing differently-named arrays drops the name). Fix (PR on this branch): non-precomputed path re-emits friction.attrs + friction.name on every output via new _apply_geo_metadata helper (single, threshold, all-NaN-unreachable, and pairwise-Dataset paths); precomputed path left on the existing source-derived behaviour since there is no friction to draw from. Only .attrs/.name set -- data values, coords, dims, dtype untouched, dask stays lazy (no compute). 10 new tests (test_corridor_inherits_friction_geo_attrs x4 backends, test_corridor_threshold_keeps_geo_attrs x4 backends, test_corridor_unreachable_keeps_geo_attrs, test_pairwise_inherits_friction_geo_attrs, test_precomputed_keeps_source_attrs_not_friction). Full corridor suite 43 passed. Cat 4 N/A: NaN-as-nodata is the library convention; corridor never reads attrs['nodatavals'] for masking. No CRITICAL/MEDIUM/LOW findings."
 cost_distance,2026-06-15,3344,MEDIUM,5,"Audited 2026-06-15 (agent-ad0b84e7f7b212360 worktree, branch deep-sweep-metadata-cost_distance-2026-06-15). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live end-to-end with a rich attrs set (res/crs/transform/nodatavals/_FillValue/units). Cat 1 attrs, Cat 2 coords (values + float64 dtype), and Cat 3 dims (y,x) all preserved and identical across the 4 backends -- public cost_distance() wraps with xr.DataArray(coords=raster.coords, dims=raster.dims, attrs=raster.attrs). NEW MEDIUM finding #3344 (Cat 5): the dask+numpy and dask+cupy backends leaked the internal dask graph name (_trim-<hash> from map_overlap, asarray-<hash> from the dask+cupy convert-back path) into result.name while numpy/cupy returned None; .name was a nondeterministic per-run token that breaks .to_dataset() variable keys and any name-keyed pipeline. Same .name-leak class as proximity #2723 and zonal #2611. Fix (PR #3349 on this branch): return result.rename(raster.name) -- a constructor name= kwarg does not override a named dask array, and name=None is treated as infer-from-data, so .rename() is required. supports_dataset path unaffected (keys by var_name, verified live). New parametrized regression test test_result_name_matches_input over 4 backends x {None, named}; full cost_distance suite 63 passed (post-merge with origin/main). LOW (documented, not fixed): output float32 uses NaN as the unreachable sentinel but input nodatavals/_FillValue (e.g. -9999) are carried through verbatim, so a downstream reader masks a value that never appears -- this is the library-wide attrs=raster.attrs convention shared by proximity/slope/aspect/focal, not a cost_distance-specific bug, so fixing it in isolation would diverge this module from every peer. No CRITICAL/HIGH findings."
 focal,2026-06-10,3217,MEDIUM,4;5,"Re-audited 2026-06-10 (agent-ad0d55a894c6abc60 worktree, branch deep-sweep-metadata-focal-2026-06-10). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live for mean, apply, focal_stats, hotspots. Cats 1-3 clean: attrs (res/crs/nodatavals/_FillValue/unit), coords (values, dtype, coord attrs), dims, .name, 3D per-band path, and hotspots unit=% all preserved and identical across the 4 backends. NEW MEDIUM finding #3217 (Cat 4 + Cat 5): (a) mean() hardcoded float32 on the GPU paths (_mean_cupy cupy.asarray(dtype=float32), _mean_dask_cupy astype(float32)) while numpy/dask+numpy returned float64 (mean() casts astype(float) before dispatch), so float64 input silently lost precision on cupy/dask+cupy; dask+cupy also advertised float64 (untyped meta) but computed float32. (b) apply()/focal_stats() dask paths passed untyped meta (np.array(()) / cupy.array(())) to map_overlap, so for float32/int input the lazy DataArray advertised float64 but computed the promoted float32 (#2805 typed the chunk fns but not the meta). Same class as aspect #2682 and proximity #2723. Fix: the mean() GPU dtype half landed on main first via duplicate issue #3214/PR #3221 (_promote_float contract: float dtypes preserved, ints->float32, GPU bit-exact vs CPU in float64); PR #3226 (branch deep-sweep-metadata-focal-2026-06-10-01) types every map_overlap meta with data.dtype and aligns tests to the _promote_float contract; 25 new parametrized regression tests (4 backends x 3 dtypes mean; dask backends x 3 dtypes apply/focal_stats; exact CPU/GPU parity). Full focal suite 258 passed. No other CRITICAL/HIGH/MEDIUM/LOW findings."
 geotiff,2026-06-09,3116,HIGH,2;3,"Re-audited 2026-06-09 (agent-ae89ff94a64e3ee8f worktree, branch deep-sweep-metadata-geotiff-2026-06-09). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live. Focus: surfaces changed since the 2026-05-18 audit (unpack rename + GPU/dask+GPU support #3075, pack=True #3065/#3079, masked int->float promotion #2994, bbox= reads, rioxarray param alignment #2963, no-georef VRT coord synthesis #2824, GeoTransform omission #2971). Live probes: unpack attrs (scale_factor/add_offset/mask_and_scale_dtype/nodata/masked_nodata), masked=True promotion, default masked=False, bbox window+transform shift, multi-band band=N, dims/name/coords (incl. coord dtype) all identical across the 4 backends; nodata_pixels_present absent on dask paths is the documented lazy contract, not a bug. pack->unpack round trips verified on numpy/dask/gpu-write; pack of a cupy-backed read raises via the known cupy+xarray xp.astype incompat (see memory cupy_where_astype_incompat; dependency-pin fix, raises loudly, not a metadata bug). VRT reads (full/masked/window/bbox) and no-georef TIFF reads agree across the 4 backends. NEW HIGH finding #3116 (Cat 2+3): to_geotiff(non_georef_da, out.vrt, tile_size=N) wrote a corrupt index for arrays spanning >1 tile -- write_vrt derives placement from each source GeoTransform and non-georef tiles all carry the identity transform, so rasterX/YSize collapsed to one tile and every DstRect landed at the origin; reads silently returned a single tile (24x32 in -> 16x16 out). Gap left by #2966/#2971 (tests only covered one non-georef source). Fix: _write_vrt_tiled threads per-tile pixel offsets through _build_vrt -> write_vrt via internal dst_offsets kwarg; write_vrt refuses >1 all-non-georef sources without explicit placement and rejects dst_offsets alongside georeferenced sources. 18 new tests in tests/vrt/test_non_georef_placement_3116.py incl. 4-backend round trip, dask-backed and plain-ndarray writes, XML DstRect assertions, georef placement regression, and the write_vrt error contract. Full vrt suite 520 passed; write+round-trip suites 1292 passed."
 interpolate,2026-06-12,3288,MEDIUM,5,kriging K_inv-None fallback was numpy-backed on all backends and misnamed the variance raster; fixed via #3288. All 4 backends verified end-to-end on GPU host. LOW (documented only): template nodatavals/_FillValue copied verbatim while fill_value is the actual output sentinel; tests codify attrs==template.attrs
 mcda,2026-06-10,3147,HIGH,1,"constrain() dropped all attrs (res/crs/nodatavals) whenever exclude non-empty (xr.where takes attrs from scalar fill); fixed via attrs restore, tests for numpy/dask/dask+cupy. All other mcda funcs keep attrs/coords/dims on all 4 backends. Out-of-scope crashes noted for backend-parity: owa broken on cupy (numpy order-weights x cupy) and on dask (da.sort does not exist); sensitivity monte_carlo crashes on cupy/dask+cupy (.values on cupy); xr.where compute on cupy/dask+cupy hits known cupy13.6/xarray2025.12 incompat."
 multispectral,2026-06-20,3429,MEDIUM,2;3,"true_color() hardcoded y/x dims + dropped extra coords; fixed PR #3434 (all 4 backends verified, CUDA available)"
+perlin,2026-06-23,,MEDIUM,2,perlin() dropped input x/y coords (kept attrs incl crs); fixed by passing coords=agg.coords; verified numpy/cupy/dask/dask+cupy; issues disabled on fork so no issue number; PR pending
 polygonize,2026-06-12,3293,MEDIUM,1,"Audited 2026-06-12 (agent-a86d90abea41b04cf worktree, branch deep-sweep-metadata-polygonize-2026-06-12). CUDA available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live for int, float+NaN, and no-georef rasters. polygonize returns vector output (numpy/awkward/geopandas/spatialpandas/geojson), not a DataArray, so Cats 2-4 reinterpreted as transform/CRS/value-dtype propagation. Transform auto-detect (attrs['transform'] -> rio.transform() -> x/y coords, #2536/#2607) and CRS resolution run in public polygonize() before dispatch, so all 4 backends emit identical columns, bounds, and CRS (verified live). Column value dtype follows input dtype on every backend. NEW MEDIUM finding #3293 (Cat 1): _detect_raster_crs ignored the _xrspatial_no_georef marker that _detect_raster_transform honours, so a geotiff-reader crs_only raster (attrs carry both crs and the marker; metadata_to_attrs writes crs independent of has_georef) produced a GeoDataFrame claiming EPSG:#### over pixel-space geometries -- the #2536 metadata-lies-about-the-data mismatch through the marker channel. contour.py imports the same helper and inherits the fix. Fix: early return None in _detect_raster_crs on the marker + docstring note; 2 new tests in TestPolygonizeCRSPropagation. polygonize+contour suites 274 passed; all 9 auxiliary polygonize test files 303 passed. rotated-read path unaffected (reader drops CRS there). No CRITICAL/HIGH/LOW findings."
 proximity,2026-05-29,2723,MEDIUM,4;5,"Audited 2026-05-29 (agent-a61dbadc2452a2003 worktree, branch deep-sweep-metadata-proximity-2026-05-29). CUDA+cupy available; all 4 backends (numpy/cupy/dask+numpy/dask+cupy) run live end-to-end for proximity/allocation/direction, both bounded (finite max_distance) and unbounded. Cat 1 (attrs res/crs/transform/nodatavals/_FillValue), Cat 2 (coords + coord dtype), and Cat 3 (dims) all preserved and identical across the 4 backends -- public funcs wrap with xr.DataArray(coords=raster.coords, dims=raster.dims, attrs=raster.attrs). NEW MEDIUM finding #2723 (Cat 4 + Cat 5): (a) bounded dask+numpy path (_process_dask -> da.map_overlap with meta=np.array(())) declared output dtype float64 while the chunk fn returns float32 and numpy/cupy/dask+cupy + the unbounded KDTree path all declare float32; docstrings show dtype=float32. Fix: meta=np.array((), dtype=np.float32). (b) dask backends leaked an internal dask op name (_trim-<hash>, _kdtree_chunk_fn-<hash>, asarray-<hash>) into result.name while numpy/cupy return None. Fix: assign result.name=None after construction in all 3 public funcs (xarray ignores a name=None kwarg for named dask arrays, so the reset must happen post-construction). Same .name-leak class as zonal #2611. PR #2728 off child branch deep-sweep-metadata-proximity-2026-05-29-01. New parametrized regression test test_output_metadata_consistent_across_backends asserts declared dtype float32 + name None across all 4 backends x 3 funcs x bounded/unbounded; full test_proximity.py suite 93 passed. No other CRITICAL/HIGH/MEDIUM/LOW findings."
 rasterize,2026-06-09,3087,MEDIUM,1,GeoDataFrame .crs dropped on no-like path (Cat 1); fixed via #3087 emitting attrs crs/crs_wkt when output has no CRS. like-path attrs/coords/dims/nodata verified live on all 4 backends (CUDA available); Cats 2-5 clean.

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-performance-state.csv

@@ -15,7 +15,7 @@ edge_detection,2026-03-31T18:00:00Z,SAFE,compute-bound,0,,
 emerging_hotspots,2026-03-31T18:00:00Z,SAFE,compute-bound,0,,
 erosion,2026-03-31T18:00:00Z,WILL OOM,memory-bound,2,1120,Memory guard added. Algorithm inherently global.
 fire,2026-03-31T18:00:00Z,SAFE,compute-bound,0,,
-flood,2026-03-31T18:00:00Z,SAFE,compute-bound,0,,
+flood,2026-06-25,RISKY,compute-bound,1,3503,"Cat1 HIGH: _validate_mannings_n_dataarray used .values, eagerly materialized dask/cupy roughness raster (OOM); fixed lazy-safe. Core elementwise ops SAFE. LOW: dask+cupy host round-trip (map_blocks b.get) shared with cost_distance/surface_distance, documented not fixed."
 focal,2026-05-29,SAFE,compute-bound,1,2734,"HIGH: _hotspots_dask_cupy chunk fn round-tripped each chunk host<->GPU (cupy.asnumpy classify cupy.asarray); fixed PR 2739 to reuse _run_gpu_hotspots on device. LOW (not fixed): _apply_numpy/_hotspots_cupy use zeros_like where empty would suffice. CUDA kernels regs<=62, no register-pressure issue."
 geodesic,2026-03-31T18:00:00Z,N/A,compute-bound,0,,
 geotiff,2026-06-11,SAFE,IO-bound,0,3235,"Pass 15 (2026-06-11): 1 MEDIUM found and fixed. _pack (_attrs.py:~1795) guarded the no-sentinel integer restore with an eager bool(out.isnull().any()), which executed the whole upstream dask graph at to_geotiff(pack=True) call time; the streaming writer then executed it again, so every source chunk computed twice (measured 32 decode-task executions for 16 chunks on a 512x512 int16 SCALE/OFFSET no-GDAL_NODATA source; 71->33 total task starts post-fix). Filed #3235, fixed by mapping a per-chunk NaN guard (_pack_guard_no_nan) into the graph for dask-backed data (raises from the write's single compute; numpy keeps the eager call-time check; meta= preserves cupy backing). 9 new tests in test_pack_lazy_nan_guard_3235.py incl. fusion-proof execution counter and cupy-chunk guard unit test (dask+cupy e2e still blocked upstream by #3112). Scrutinised all 16 commits since 2026-06-08 (pack/unpack series #3065/#3075/#3079/#3129/#3174/#3175, VRT placement #3135, compression_level gate #3176, streaming banding #3136, dask+cupy writer order fix #3171): no other regressions; #3171's get-then-asarray order is intentional D2H for gpu=False. GPU validated on-device this pass: eager GPU unpack returns cupy with exact parity (387ms incl warmup, only 0-d scalar .get()s -- no bulk host round trip), dask+GPU unpack lazy (112 tasks/16 chunks, cupy meta, compute returns cupy, parity 0.0), GDS fast path intact without unpack (4 tasks/chunk); unpack disqualifying GDS is documented intentional. Dask CPU probe 4 tasks/chunk, 50k-task cap intact. Note: #1714 (_write_vrt_tiled synchronous scheduler) is now FIXED+CLOSED (scheduler='threads' at _writers/eager.py:1517) -- drop from the open-issue list. LOW noted (no fix): _pack does identity (data-0.0)/1.0 arithmetic allocating two full-array temporaries when scale==1/offset==0 (masked_nodata-only pack); prior deferred LOWs unchanged. SAFE/IO-bound holds. | Pass 14 (2026-06-09): MEDIUM found and fixed -- _write_streaming ran one dask .compute() per 256-row tile-row/strip, so a source chunk taller than the band re-executed once per band it overlapped (measured 2x at chunks=512, 4x at chunks=1024, whole upstream graph re-runs for computed pipelines). Filed #3117, fixed via _stream_row_bands: consecutive tile-rows/strips group into row bands sized by the source chunk-row span (one-chunk halo, #3007 accounting) under streaming_buffer_bytes; each band computes once and tiles/strips are carved from the materialised band. Wide rasters needing column segmentation keep the per-tile-row path. Post-fix per-chunk executions == 1 on the default read->write round trip. 5 new tests (TestRowBandRecompute3117 + _stream_row_bands unit); write/integration/parity suites pass (2195). LOW deferred (no fix): _read_geotiff_gpu_chunked parses header+all IFDs twice at graph build (_backends/gpu.py ~1367-1419, cap check then GDS probe; build-time only). GPU paths validated on-device this pass: eager gpu read returns cupy with parity, dask+GPU chunked read lazy (17 tasks/4 chunks) with parity; GPU writer full materialisation is documented intentional (streaming_buffer_bytes no-op). Read path keeps 50k-task graph cap; dask read probe 4 tasks/chunk. SAFE/IO-bound holds. | Pass 13 (2026-05-20): 1 MEDIUM found and fixed. _nvjpeg_batch_encode (_gpu_decode.py:~L1560) and _nvjpeg2k_batch_encode (~L2958) called cupy.cuda.Device().synchronize() inside the per-tile encode loops, a whole-device fence that blocked every CUDA stream and serialised concurrent work (e.g. predictor encodes on other streams). The decode-side counterpart _try_nvjpeg_batch_decode already used cupy.cuda.Stream.null.synchronize() at L1442; the encoder side was inconsistent. Filed #2212 and fixed both encoders to use Stream.null.synchronize(), scoping the per-tile sync to the default stream the encode/retrieve calls were issued on. nvJPEG / nvJPEG2000 encoders maintain a single shared state per encoder so encodes within a batch are inherently serial; the fix removes the device-wide blocker without changing the API ordering contract. 5 new tests in test_nvjpeg_encode_stream_sync_2212.py (AST checks that neither encoder contains Device().synchronize() inside a for-loop, that both call Stream.null.synchronize() in the loop, and that the decoder reference pattern stays pinned). All 5 new tests + 19 existing related encode/decode tests pass. nvjpeg/nvjpeg2k shared libs not present on this host so end-to-end encode verification is gated; add cuda-unavailable-libs note to re-validate on a host with the RAPIDS conda env. SAFE/IO-bound verdict holds; no change in dask graph cost. Dask probe: 2560x2560 deflate-tiled file via read_geotiff_dask(chunks=256) yields 400 tasks for 100 chunks (4 tasks/chunk), well under the 50K cap. LOW deferred (no fix in this PR): _build_ifd called twice per IFD level in _assemble_standard_layout (_writer.py:1531+1543), _assemble_cog_layout (1582+1625), and the COG overview path (2519+2546+2740) -- the first call's bytes are discarded; only the overflow byte length is used to compute pixel_data_offset. Cost is bounded by IFD count (typically 1-5 overview levels) so absolute impact is minor. Pre-existing pattern. | Pass 12 (2026-05-18): 1 MEDIUM found and fixed. _try_nvjpeg2k_batch_decode at _gpu_decode.py:~L2725-2778 allocated per-tile per-component cupy.empty buffers (N*S round-trips through the cupy memory pool) and called cupy.cuda.Device().synchronize() once per tile, forcing default-stream serialisation that defeats nvJPEG2000's internal pipelining. Filed #2107 and fixed: pre-allocate a single d_comp_pool sized n_tiles*samples*tile_height*pitch under a _check_gpu_memory guard, derive per-tile/per-component views as slab offsets, and replace the per-tile sync with a single batch-end sync. Same pattern as #1659 (_try_nvcomp_from_device_bufs), #1688 (_try_kvikio_read_tiles), #1712 (_nvcomp_batch_compress). 7 new tests in test_nvjpeg2k_single_alloc_2107.py: AST-level structural assertions confirm no cupy.empty inside the for-loop and no Device().synchronize() inside the loop, plus pool/per_tile_comp_bytes presence and _check_gpu_memory guard checks; lib-absent short-circuit; unsupported-dtype cleanup contract; cupy-only pool slab-non-overlap test (gpu-marked). libnvjpeg2k.so not present on this host so the end-to-end nvJPEG2000 decode is gated -- note added to re-validate on a host with the RAPIDS conda env. All 30 jpeg2000/compression tests + 7 new tests pass. SAFE/IO-bound verdict holds (no change in dask graph cost). Dask probe: 4096x4096 deflate-tiled file via read_geotiff_dask(chunks=512) yields 256 tasks for 64 chunks (4 tasks/chunk), well under the 50K cap. | Pass 11 (2026-05-18): 1 MEDIUM found and fixed. _read_strips (_reader.py:~L1972) and _fetch_decode_cog_http_strips (_reader.py:~L2670) decoded strips sequentially in a Python for-loop while the tile counterparts (_read_tiles L2146, _fetch_decode_cog_http_tiles L2898) gated parallel decode on _PARALLEL_DECODE_PIXEL_THRESHOLD via ThreadPoolExecutor. Filed #2100 and fixed: both strip paths now collect jobs, parallel-decode when n_strips > 1 and strip_pixels >= 64K, then place sequentially. Measured (uint16, 4-core): 4096x4096 deflate 130ms->34ms (3.82x), 8192x8192 deflate 531ms->146ms (3.63x), 8192x8192 zstd 211ms->85ms (2.48x), uncompressed 25ms->22ms (1.14x). 5 new tests in test_parallel_strip_decode_2100.py (parallel/serial parity, pool-engaged on multi-strip, serial-path for single-strip, windowed cross-strip read, HTTP COG strip parity). 3998 tests pass; 8 pre-existing failures predating this change (predictor2 BE + size_param_validation_gpu_vrt reference now-private read_to_array attr). SAFE/IO-bound verdict holds. | Pass 10 (2026-05-15): 1 new MEDIUM found and fixed; 2 LOW noted. MEDIUM (_reader.py:2737): _fetch_decode_cog_http_tiles decoded tiles sequentially in a Python for-loop after the concurrent fetch landed (issue #1480). Local _read_tiles parallelises decode whenever tile_pixels >= 64K via ThreadPoolExecutor (_reader.py:2017); the HTTP path was structurally similar but never picked up the same gate, so wide windowed reads of multi-tile COGs left deflate/zstd decode single-threaded. Mirrored the local-path threshold + pool. 5 new tests in test_cog_http_parallel_decode_2026_05_15.py (parallel + serial round-trip correctness, pool-instantiation branch selection above the threshold, single-tile path skips the pool, structural _decode_strip_or_tile call count == n_tiles). All 262 COG/HTTP tests pass; 3162 of 3164 selected geotiff tests pass overall (2 pre-existing failures predating Pass 9 per prior notes -- test_predictor2_big_endian_gpu_1517 references the now-private read_to_array attr, and the test_size_param_validation_gpu_vrt_1776 tile_size=4 validator failure). LOW deferred (no fix in this PR): (1) _block_reduce_2d_gpu (_gpu_decode.py:3142/3163/3189) does bool(mask.any().item()) per overview level when nodata is set, paying one device sync per level; the alternative (unconditional cupy.putmask) always pays the work cost and the short-circuit is correct under the current API. (2) _nvcomp_batch_compress adler32 staging (_gpu_decode.py:2543-2546) issues n_tiles slice-assign kernels into a fresh contig buffer despite all callers passing slices of a single underlying d_tile_buf; an API refactor to accept the source buffer directly would skip the rebuild. SAFE/IO-bound verdict holds. Dask probe: 2560x2560 chunks=256 yields 400 tasks (4 per chunk), well under the 50000 cap. GPU probe: 1024x1024 float32 zstd read returns CuPy-backed in 236 ms with no host round-trip. | Rockout 2026-05-15: LOW filed #1934 -- _apply_nodata_mask_gpu used cupy.where (allocating); switched to cupy.putmask on the already-owned buffer (float path) and on the post-astype float64 buffer (int path). Saves one chunk-sized device allocation per call. 7 new tests in test_apply_nodata_mask_gpu_inplace_1934.py; 52 related nodata tests pass. | Pass 8 (2026-05-12): 1 new MEDIUM found and fixed. _assemble_standard_layout/_assemble_cog_layout returned bytes(bytearray), doubling peak memory transiently during eager writes. Filed #1756, fixed by returning the bytearray directly. Measured: 95 MB uint8 raster peak drops 202 MB -> 107 MB. _write_bytes / parse_header already accepted the buffer protocol so the change is transparent to callers. 6 new tests in test_assemble_layout_no_bytes_copy_1756.py. 2123 existing geotiff tests pass; the 10 unrelated failures (test_no_georef_windowed_coords_1710, test_predictor2_big_endian_gpu_1517) reference the now-private read_to_array attribute (commit 8adb749, issue #1708) and predate this change. SAFE/IO-bound verdict holds. | Pass 7 (2026-05-12): re-audit identified 4 MEDIUM findings, all real, all backed by microbenches. (1) unpack_bits sub-byte loops for bps=2/4/12 in _compression.py:836-878 were 100-200x slower than vectorised numpy (filed #1713, fixed in this branch: bps=4 2M pixels drops from 165ms to 3ms = 55x; bps=2/12 similar). (2) _write_vrt_tiled at __init__.py:1708 uses scheduler='synchronous' on independent tile writes; measured 33% slowdown on 256-tile zstd write vs threads scheduler (filed #1714, no fix yet). (3) _nvcomp_batch_compress at _gpu_decode.py:2522-2526 still does per-tile cupy.get().tobytes() despite #1552 / #1659 fixing the same pattern elsewhere; measured 45% reduction with concat+single get on n=1024 (filed #1712, no fix yet). (4) _nvcomp_batch_compress at _gpu_decode.py:2457 uses per-tile cupy.empty allocations; 1024 tiles 16KB drops from 4.7ms to 1.0ms with single contiguous + views (bundled into #1712). Cat 6 OOM verdict: SAFE/IO-bound holds -- read_geotiff_dask caps task count at _MAX_DASK_CHUNKS=50_000 and per-chunk memory is bounded by chunk size. _inflate_tiles_kernel resource usage on Ampere: 67 regs/thread, 2896B local/thread, 8192B shared/block (LZW kernel: 29 regs, 24576B shared) -- register pressure under control; high local memory in inflate is unavoidable (LZ77 state) but only thread 0 in each block uses it. | Pass 4 (2026-05-10): re-audit after #1559 (centralise attrs across all read backends). New _populate_attrs_from_geo_info helper at __init__.py:301 runs once per read, not per-chunk -- no perf impact. Probe: 2560x2560 deflate-tiled file opened via read_geotiff_dask yields 400 tasks (4 tasks/chunk for 100 chunks), well under 1M cap. read_geotiff_gpu(1024x1024) returns cupy.ndarray end-to-end with no host round-trip (226ms incl. write+decode). No new HIGH/MEDIUM findings. SAFE/IO-bound holds. | Pass 3 (2026-05-10): SAFE/IO-bound. Audited 4 perf commits: #1558 (in-place NaN writes on uniquely-owned buffers correct), #1556 (fp-predictor ngjit ~297us/tile for 256x256 float32), #1552 (single cupy.concatenate + one .get() for batched D2H at _gpu_decode.py:870-913), #1551 (parallel decode threshold >=65536px engages 256x256 default at _reader.py:1121). Bench: 8192x8192 f32 deflate+pred2 256-tile write 782ms; 4096x4096 f32 deflate read 83ms with parallel decode. Deferred LOW (none filed, all <10% MEDIUM threshold): _writer.py:459/1109 redundant .copy() before predictor encode (~1% per tile), _compression.py:280 lzw_decompress dst[:n].copy() (~2% per LZW tile decode), _writer.py:1419 seg_np.copy() before in-place NaN substitution (negligible, conditional path), _CloudSource.read_range opens fresh fsspec handle per range (pre-existing, predates audit scope). nvCOMP per-tile D2H batching break-even confirmed (variable sizes need staging buffer, no win). | Pass 3 (2026-05-10): audited f157746,39322c3,f23ec8f,1aac3b7. All 5 commits correct. Redundant .copy() in _writer.py:459,1109 and _compression.py:280 (1-2% overhead, LOW). _CloudSource.read_range() per-call open is pre-existing arch issue. No HIGH/MEDIUM regressions. SAFE. | re-audit 2026-05-02: 6 commits since 2026-04-16 (predictor=3 CPU encode/decode, GPU predictor stride fix, validate_tile_layout, BigTIFF LONG8 offsets, AREA_OR_POINT VRT, per-tile alloc guard). 1M dask chunk cap intact at __init__.py:948; adler32 batch transfer intact at _gpu_decode.py:1825. New code is metadata validation and dispatcher logic with no extra materialization or per-tile sync points. No HIGH/MEDIUM regressions. | Pass 5 (2026-05-12): re-audit identified MEDIUM in _gpu_decode.py:1577 _try_nvcomp_from_device_bufs: per-tile cupy.empty + trailing cupy.concatenate doubled peak VRAM and added serial concat. Filed #1659 and fixed to single-buffer + pointer offsets (matches LZW/deflate/host-buffer patterns at L1847/L1878/L1114). Microbench (alloc+concat overhead only, not full nvCOMP latency): n=256 tile_bytes=65536 drops 3.66ms->0.69ms, n=256 tile_bytes=262144 drops 8.18ms->0.13ms. Tests: 5 new tests in test_nvcomp_from_device_bufs_single_alloc_1659.py (codec short-circuit, no-lib short-circuit, memory-guard contract, real ZSTD round-trip via nvCOMP, structural single-buffer check). 1458 existing geotiff tests pass, 3 unrelated matplotlib/py3.14 failures pre-existing. SAFE/IO-bound verdict holds. | Pass 6 (2026-05-12): re-audit on top of #1659. New HIGH in _try_kvikio_read_tiles at _gpu_decode.py:941: per-tile cupy.empty() + blocking IOFuture.get() inside loop serialised GDS reads to ~1 outstanding pread, missed parallelism the kvikio worker pool was designed for, paid per-tile cupy.empty setup (matches #1659 anti-pattern in nvCOMP path), and lacked _check_gpu_memory guard. Filed #1688 and fixed to single contiguous buffer + batched submit + guard. Microbench with 8-worker pool simulation: 256 tiles@1ms latency drops 256ms->38.7ms (~6.6x); single-thread simulation 256ms->28.5ms (9x). Tests: 9 new tests in test_kvikio_batched_pread_1688.py (kvikio-absent path, single-buffer pointer arithmetic, submit-before-get ordering, memory guard, partial-read fallback, round-trip data, zero-size/all-sparse tiles). All 1577 geotiff tests pass except pre-existing matplotlib/py3.14 failures."
@@ -32,7 +32,7 @@ morphology,2026-06-20,SAFE,compute-bound,1,3401,memory guard fired on full lazy-
 multispectral,2026-05-02,SAFE,compute-bound,0,,"Re-audit 2026-05-02 after PRs 1292 (true_color memory guard) and 1301 (validate_arrays in true_color). Verified SAFE. No HIGH. MEDIUM: da.stack in _true_color_dask/_true_color_dask_cupy at L1702/L1731 creates (1,1,1,1) chunks along band axis (4 bands so impact is minor, scheduling overhead not OOM). LOW: np.zeros((h,w,4)) at L1681 then full overwrite -- np.empty would suffice. All 17 indices use plain map_blocks with no halo; 8192x8192 ndvi graph is 80 tasks, evi/arvi/ebbi 112 tasks."
 normalize,2026-03-31T18:00:00Z,SAFE,compute-bound,0,1124,Boolean indexing replaced with lazy nanmin/nanmax/nanmean/nanstd.
 pathfinding,2026-04-15T12:00:00Z,SAFE,compute-bound,0,false-positive,Downgraded. CuPy .get() is required -- A* has no GPU kernel. Per-pixel .compute() is only 2 calls for start/goal validation. seg.values in multi_stop_search collects already-computed results for stitching.
-perlin,2026-03-31T18:00:00Z,WILL OOM,memory-bound,0,,
+perlin,2026-06-23,SAFE,memory-bound,1,3469,"Re-audit 2026-06-23 (CUDA host). 1 HIGH: _perlin_dask_numpy (L128) and _perlin_dask_cupy (L279) called dask.persist on the full noise array before min/ptp(max) reductions -> every chunk resident at once, WILL OOM at 30TB. Persist redundant: both reductions share the named noise subgraph in one dask.compute (verified min/ptp graphs ref same 100 noise keys). Fixed by removing persist -> peak mem chunk-bounded, SAFE. map_blocks graph embarrassingly parallel (3.3 tasks/chunk). cupy kernel _perlin_gpu 36 regs/thread (no pressure); cupy path stays on GPU, no host round trip. numpy/cupy single-GPU unaffected. 13 perlin tests pass incl GPU+dask+GPU. PR #3469-fix."
 polygon_clip,2026-06-10,SAFE,graph-bound,0,3191,"crop=True picked tiny leading edge chunk as rasterize mask size -> 13169-task graph; fixed to max(rc),max(cc) -> 1045 tasks. crop=False/numpy/cupy clean. Cat1-5 clean. GPU+dask+cupy run-validated."
 polygonize,2026-06-12,RISKY,compute-bound,0,3303,"Pass 3 (2026-06-12): re-audit after #2817/#2913/#3041. 0 HIGH. 1 MEDIUM fixed (#3303): _compute_region_value_ranges ran a pure-Python per-pixel loop (95% of float chunk time; 0.283s of 0.299s on 1024x1024, float chunks ~30x int) and re-ran _calculate_regions on an already-labelled block; moved to jitted _region_ranges_scan + _polygonize_numpy_regions label reuse (0.299s -> 0.015s/chunk). Side fix: w_match/s_match flags were always-truthy (_is_close numba overload generator called from pure Python returns impl function); output-neutral by chunk geometry, now computed correctly in jit. Cat1/2 clean (dask.compute batching is the documented #2673 design). Cat3 validated on GPU: cupy int/float + dask+cupy run end-to-end, single documented transfer, no round-trip. Cat4/5 LOW unchanged: _calculate_regions_cupy per-unique-value labeling (low impact); per-polygon Python classify loop in _polygonize_chunk dominates only on pathological many-polygon chunks (788K polys -> 7.8s). Cat6 RISKY unchanged: driver accumulates O(total polygons); 32-chunk batches bound transient peak. 527 polygonize tests + 40 new pass."
 proximity,2026-06-09,RISKY,graph-bound,0,3103,"Pass 2 (2026-06-09): re-audit after 16 fix commits since 2026-03-31. 0 HIGH, 2 MEDIUM found and fixed: (1) #3103/PR #3126 line-sweep @ngjit closure inside _process recompiled per call (~0.42s constant overhead; 10x10 warm call 0.44s->1ms after module-level hoist with explicit args, 1000x1000 0.49s->35ms); (2) #3132/PR #3137 dask xs/ys coordinate grids built via da.tile/da.repeat+rechunk cost ~185 tasks/chunk with the ys term scaling O(raster height) (~4.3 tasks/row, 44K tasks at 10240 rows); chunk-aligned da.broadcast_to gives identical values, bounded graph 18535->5554 tasks (3.3x) on 2560^2/256 chunks; regression test bounds tasks/chunk<80 (old 100.4, new 58.7) + ragged-chunk parity. LOW not fixed: zeros+fill(-1) row buffers in line-sweep; numpy backend materializes full float64 xs/ys grids (guarded since #1111); unbounded KDTree streaming count pass computes chunks on driver by design (gh-879). GPU validated on CUDA host: cupy 1024^2 proximity 6ms device-resident with exact numpy parity, dask+cupy bounded parity exact, _proximity_cuda_kernel 56 regs/thread (no register pressure). _halo_depth python loop measured 58ms at 100K coords - not a finding. Verdict RISKY (was WILL OOM): unbounded paths either guarded (MemoryError at 80% mem) or stream via kdtree; bounded map_overlap peak scales with chunk size."

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-style-state.csv

@@ -7,13 +7,14 @@ dasymetric,2026-06-20,3411,HIGH,1;3;4,"F401 is_dask_cupy unused (HIGH); E128 lin
 diffusion,2026-06-20,3421,HIGH,3;4,F401 unused not_implemented_func (HIGH); isort import-ordering drift (MEDIUM); no Cat1/2/5. Fixed in PR off issue 3421.
 fire,2026-06-19,3395,HIGH,3;4,Cat3 F401 math.log unused + F841 S_T unused local; Cat4 isort utils import reorder. No Cat1/2/5. Fixed in #3395 rockout.
 focal,2026-05-29,2731,HIGH,3;4;5,"F401 not_implemented_func (import line 36, unused, not re-exported). isort: stdlib reorder (import math before from-imports), dropped stray blank lines in import groups, alphabetised+rewrapped convolution/utils from-imports, moved dataset_support import into order. Cat 5: mutable default excludes=[np.nan] in mean() (line 238) -> None sentinel, resolved to [np.nan] in body; never mutated so behaviour preserved; regression test test_mean_default_excludes_does_not_leak added. Cat 1/2 clean. 115 focal tests pass. PR pending."
-geotiff,2026-06-14,3329,MEDIUM,4,"Re-sweep after #3259: flake8 baseline 0 across all 33 production files (flake8 7.3.0). Cat 4 only: isort (line_length=100) drift in _writers/eager.py (the from .._validation import block wrapped at ~80 cols, not 100); rewrapped, no behavioural change. Cat 1/2/3/5 clean: no E/W/F codes; grep hits for == False (comment), ': dict'/': list' annotations, and type=getaddrinfo kwarg are all false positives, not shadowed builtins or comparisons. isort+flake8 clean after fix; import smoke test ok. PR via #3329."
+geotiff,2026-06-25,,MEDIUM,4,"Re-sweep 2026-06-25 (file last modified 2026-06-24). flake8 baseline 0 across all 35 production files AND all tests (flake8 7.x, max-line-length=100). Cat 4 only: one isort (line_length=100) finding in tests/unit/test_exception_exports_3265.py - the deferred 'from xrspatial.geotiff._reader import PixelSafetyLimitError as reader_pixel_err' was written with a backslash continuation but joins to 83 chars, so isort wants it on one line. Joined it; formatting only, no behaviour change. Cat 1/2/3/5 clean: no E/W/F codes; grep hits for == False (a comment), ': dict'/': list' annotations, and type=getaddrinfo kwarg are false positives (not shadowed builtins or comparisons). isort+flake8 clean after fix; tests collect ok. Issue-create denied by auto-mode classifier; fix committed to deep-sweep branch; PR #3517 opened."
 hydro-d8,2026-05-29,2705,HIGH,1;3;4,"flake8+isort over the 13 D8 files only (dinf/mfd out of scope). Cat 3 HIGH: F401 x2 (flow_length_d8 function-local _compute_accum_seeds never called; snap_pour_point_d8 module-level cuda_args unused) - both confirmed dead, no re-export. Cat 1: E127/E128 continuation-indent x90 (mostly multi-line def signatures); E302/E303 blank-line cluster in watershed_d8; E501 x4 (flow_path_d8 + snap_pour_point_d8, wrapped ternaries). Cat 4: isort import-block reordering on all 13 files. No Cat 2 (W-codes), no Cat 5 (grep clean: no bare except, mutable defaults, ==None/==True, or shadowed builtins). flake8+isort clean after fix; 385 D8 tests pass. flow_direction_d8 needed manual blank-line placement to satisfy both isort and E302."
 interpolate,2026-06-12,3286,HIGH,3;4,"Full subpackage sweep (_idw, _kriging, _spline, _validation). Cat 3 F401: unused 'import math' in _idw.py L5 (IDW kernels are pure arithmetic; _spline.py keeps math for math.log in TPS kernels; not re-exported, __init__ exports only idw/kriging/spline). Cat 4 isort: _idw.py + _spline.py 5-line xrspatial.utils from-import reflowed to 2 lines under line_length=100, matching _kriging.py from #2916. Cat 1/2/5 clean (no E/W codes; grep: no bare except, mutable defaults, ==None/True, shadowed builtins). flake8+isort clean after fix; 66 interpolation tests pass (CUDA available). PR open."
 interpolate-kriging,2026-06-04,2916,MEDIUM,1;4,"flake8 E128 x2: continuation-line under-indent at the _chunk_var kriging-predict calls in _kriging_dask_numpy (L234) and _kriging_dask_cupy (L324); re-indented to visual-indent column. Cat 4 isort: 5-line from xrspatial.utils (...) block collapses to one 88-char line under line_length=100. Cat 2/3/5 grep clean (no W-codes, F-codes, bare except, mutable defaults, ==None/True, or shadowed builtins). flake8+isort clean after fix; 14 kriging tests pass. PR open."
 mcda,2026-06-10,3143,HIGH,3;4,F401 dead function-local warnings import in standardize.py; isort drift in weights.py combine.py __init__.py; flake8 otherwise clean; Cat 5 greps clean; fixed via PR for #3143
 morphology,2026-06-20,3400,HIGH,3;4,"F401 xr/has_cuda_and_cupy/not_implemented_func + F841 size (dead); isort ordering. /rockout PR off issue 3400. flake8+isort clean, 74 tests pass."
 multispectral,2026-06-20,3428,HIGH,3;4,F401 unused not_implemented_func (Cat3 HIGH); isort dataset_support before utils (Cat4 MEDIUM); Cat1/2/5 clean; fixed via #3428
+perlin,2026-06-23,,HIGH,3;4,"Cat3 F401: unused not_implemented_func import from xrspatial.utils (refactor leftover, not re-exported; __init__ exports only perlin). Cat4 isort: utils from-import reflowed to line_length=100 after the name was dropped. No Cat1/2/5 (grep clean: no bare except, mutable defaults, ==None/True, or shadowed builtins). flake8+isort clean after fix; import smoke test ok; 12 perlin tests pass (CUDA available). GitHub issues disabled on repo so no issue filed; fix on deep-sweep branch + PR."
 polygon_clip,2026-06-10,3184,HIGH,3;4,"Cat 3 F401: dropped unused typing imports Sequence/Tuple/Union (line 10, only Optional used) and the dead 'import cupy' inside the _apply_mask dask+cupy closure (line 245); the other 'import cupy' at line 261 is used (cupy.asarray) and kept. None re-exported (__init__ re-exports only clip_polygon). Cat 4 isort: reflowed the 5-name xrspatial.utils from-import block to line_length=100. Cat 1/2/5 grep clean (no E/W codes, no bare except, mutable defaults, ==None/True, or shadowed builtins). flake8+isort clean after fix; 23 test_polygon_clip tests pass. PR open."
 polygonize,2026-05-27,2534,HIGH,1;3;4,"F401 line 58 (is_cupy_array unused, not re-exported). E127 lines 83/88 (overload continuation indent in generated_jit). isort: 5-line .utils import block collapses to one line at 100-char limit. Cat 2 clean. Cat 5 grep clean."
 proximity,2026-06-18,,MEDIUM,4,"Re-audit 2026-06-18 (file last modified 2026-06-14). Cat 4 isort: from xrspatial.utils import (...) block wrapped one name early for line_length=100; pulled has_cuda_and_cupy onto first continuation line (both lines <=100, flake8 clean). isort+flake8 now clean. Fix applied + verified imports resolve. Cat 1/2/3 clean. Cat 5 grep: target_values: list = None on L1540/1700/1859 are type annotations not builtin-shadowing vars (false positive); mutable-default already None-sentinel from prior #2725. MEDIUM. /rockout issue+PR creation BLOCKED by auto-mode classifier (external publish denied); fix committed to deep-sweep branch only."

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/.claude/sweep-test-coverage-state.csv

@@ -2,11 +2,14 @@ module,last_inspected,issue,severity_max,categories_found,notes
 aspect,2026-06-21,3439,MEDIUM,2,"#3439 (Cat 2): Inf/-Inf elevation input and all-NaN raster were untested. Added Inf-input finite-neighbors + 4-backend parity, and all-NaN -> all-NaN shape-preserved for aspect/northness/eastness on all 4 backends; GPU-validated (CUDA available). Both behaviors defined and consistent across backends -> coverage gap, not a bug (supersedes prior 'Inf possible source bug, not filed' note). Prior: #2742 degenerate shapes + geodesic boundary modes; #2829 northness/eastness geodesic branch -- all still covered."
 classify,2026-06-20,,MEDIUM,3;4,"Deep-sweep 2026-06-20 test-coverage on a CUDA host. Backend matrix was already complete: all 10 public classifiers (binary/reclassify/quantile/natural_breaks/equal_interval/std_mean/head_tail_breaks/percentiles/maximum_breaks/box_plot) x 4 backends present and green (Cat 1 no gap). Cat 2 (NaN/Inf) covered: input_data() fixture embeds -inf/nan/+inf at corners on every numpy/dask/cupy test, plus all-nan/all-inf/all-same dedicated tests. Cat 5 covered via general_output_checks(verify_attrs=True) asserting attrs/dims/coords on every backend test. Found two MEDIUM gaps: Cat 3 (no 1x1 single-pixel, no Nx1/1xN strip tests for any classifier) and Cat 4 (the _validate_scalar k<min_val guards were never exercised: quantile/natural_breaks/maximum_breaks k>=2, equal_interval k>=1). Probed live: single-pixel and strip shapes all work correctly (no source bug -- lone finite pixel -> class 0; binary match -> 1; reclassify -> bin's new_value), so these are untested-but-passing paths. Added test-only (numpy, since the gaps are in shared CPU-side validation + bin-edge logic and the 4-backend dispatch is already fully covered): test_classify_single_pixel, test_binary_single_pixel, test_reclassify_single_pixel, test_classify_nx1_strip, test_classify_1xn_strip, and 4 k-below-min ValueError tests. All RAN and PASSED on the CUDA host; full file 98 passed. classify.py untouched. LOW (documented, not fixed): empty raster (0 rows) raises on numpy equal_interval (zero-size reduction) and differs across backends -- degenerate input, not pinned. Non-square cellsize not exercised (classifiers ignore cellsize, so out of scope)."
 contour,2026-06-08,2704;2710;3044,MEDIUM,1;2,"Pass 2 (2026-06-08, deep-sweep test-coverage): re-swept on a CUDA host. Verified issue #2704 is CLOSED (fixed 2026-06-01 via #2749, kernel now uses np.isfinite at contour.py:73-74); the two prior xfail(strict) #2704 pins were already flipped to plain passing assertions in TestInfHandling (test_inf_corner_no_nan_coords / test_neg_inf_corner_no_nan_coords) -- no stale xfail remained. Found one MEDIUM Cat 1+2 gap: no cross-backend parity test fed NaN input -- TestBackendEquivalence uses elevation_raster_no_nans and test_partial_nan is numpy-only, so numpy-interior NaN-skip vs dask NaN-halo (da.overlap) parity at chunk edges was unpinned. Filed #3044, added TestNaNBackendParity (3 tests: dask, cupy, dask+cupy each assert _segments_by_level equality vs numpy on a partial-NaN ramp with a NaN edge row + interior NaN cell inside a non-edge chunk). All 3 RAN and PASSED on a CUDA host; full file 89 passed, 0 skipped. Probed and verified now-resolved: the prior-pass LOW items are no longer real gaps -- the levels=None all-NaN early-return IS asserted (result==[]) on all 4 backends (numpy L148/dask L163/cupy L178/dask+cupy L194) plus geopandas (test_geopandas_all_nan_keeps_crs). LOW (documented, NOT fixed): non-square cellsize (res[0]!=res[1]) still never exercised -- all tests use create_test_raster res (0.5,0.5); probed live that anisotropic coords transform correctly (y scaled 2.0, x scaled 0.5 -> crossing x=1.25, y spans 0..8), works, so it is a LOW coverage gap not a bug. Cat 3 1x1/Nx1/1xN remain rejected by the >=2x2 guard (tested). Test-only PR for #3044; contour.py untouched. | Pass 1 (2026-05-29): added TestInfHandling, TestCRSPropagation, TestNonDefaultDims to test_contour.py (5 passed + 2 strict-xfail on a CUDA host; full file 29 passed, 2 xfailed). All four backends (numpy / cupy / dask+numpy / dask+cupy) were already exercised with cross-backend segment-equality assertions (TestBackendEquivalence), and ran green locally on the CUDA host -- Cat 1 well covered, no new backend tests needed. Cat 2 HIGH (Inf): the marching-squares NaN-skip guard at contour.py:67 uses x!=x which does not catch infinity, so a finite level near a +/-inf corner leaks NaN coordinates into the output. Filed source bug #2704 and added two xfail(strict=True) tests pinning it (+inf and -inf) plus test_inf_far_level_no_crossing covering the safe path where the inf quad classifies as all-above (idx 15) and is skipped before any interpolation. Cat 5 MEDIUM: no test asserted gdf.crs propagation from agg.attrs['crs'] (contour.py:660) -- added test_geopandas_crs_from_attrs (to_epsg()==5070) + test_geopandas_no_crs_attr. Cat 5 MEDIUM: the index-to-coordinate transform (contour.py:644-654) reads agg.dims[0]/[1] coords but no test used non-y/x dims -- added test_lat_lon_dims_coordinate_transform + test_lat_lon_matches_yx_equivalent. PR #2710 (test-only, source untouched). LOW (documented, not fixed): non-square cellsize (cellsize_x != cellsize_y) never exercised -- all tests use res (0.5,0.5); levels=None early-return on all-NaN/all-equal works (probed) but only the explicit-levels all-NaN path is asserted. Cat 3 1x1/Nx1/1xN are rejected by the >=2x2 validation guard and that rejection is already tested (test_too_small, test_minimum_raster)."
+corridor,2026-06-22,,HIGH,1;3;4;5,"Deep-sweep 2026-06-22 test-coverage on a CUDA host. least_cost_corridor is the only public function; it delegates all backend math to cost_distance (4 backends) plus pure xarray arithmetic. Backend matrix for the core paths was already complete: symmetry/optimal-path/abs-threshold/rel-threshold/barrier/unreachable each parametrized over numpy/dask+numpy/cupy/dask+cupy. Cat 2 (NaN barriers + all-NaN unreachable) covered. Found gaps (all paths probed live and work -> coverage gaps, not source bugs): Cat 3 HIGH (no Nx1/1xN strip), Cat 5 HIGH (no attrs/coords/dim-name preservation test; corridor reads res for cost_distance math), Cat 1 MEDIUM (no numpy==other-backend exact equivalence assertion -- per-backend property checks only), Cat 4 MEDIUM (connectivity=4 and finite max_cost forwarding untested at corridor level). Added test-only: test_1xn_strip, test_nx1_strip, test_numpy_matches_other_backends (dask+numpy/cupy/dask+cupy), test_connectivity_4, test_max_cost_forwarded, test_attrs_coords_preserved, test_custom_dim_names_preserved. All RAN and PASSED on the CUDA host (cupy + dask+cupy not skipped); full file 41 passed."
 cost_distance,2026-06-16,3367,MEDIUM,1;2,"Pass (2026-06-16 deep-sweep test-coverage, CUDA host). cost_distance is heavily tested: 1122 test lines for 1354 src lines, all 4 backends parametrized + regression tests for #1191/#880/#1252/#1262/#3340/#3341/#3343/#3344. Found one MEDIUM Cat 1+2 gap: _cost_distance_dask f_min<=0 early return (all-impassable friction, finite max_cost -> da.full NaN preserving chunks) was unreached -- numpy equiv covered by test_source_on_impassable_cell, iterative dask by test_iterative_narrow_corridor, but the bounded map_overlap wrapper shortcut was not. Filed #3367, added test_dask_all_impassable_friction_returns_nan (all-zero friction, dask+numpy chunks(3,3), max_cost=5; asserts all-NaN, dask-backed, npartitions>1). RAN + PASSED; -W error::UserWarning confirms early return taken (no iterative warning). Full file 85 passed on CUDA host. LOW (documented, not fixed): non-square cellsize numeric correctness untested (_make_meta_raster uses res=(2,3) but test_metadata_preserved checks metadata only)."
 dasymetric,2026-06-20,3407;3406,HIGH,2;3;4;5,"deep-sweep test-coverage on a CUDA host (CUDA available, GPU tests ran). Module is well covered (813 test loc / 834 src): 4-backend equivalence for disaggregate weighted+binary, conservation, NaN/nodata/negative-weight, limiting_variable + cupy/dask NotImplemented guards, pycnophylactic numpy+cupy+dask-raises, validate_disaggregation all backends, memory guards (#1261). Filed #3407 (test-only) for real gaps and added 4 new test classes (11 passed, 2 xfailed). Cat5 HIGH: metadata (attrs res/crs + coords) never asserted -> TestMetadataPreservation (numpy/dask). Cat3 HIGH: true 1x1 raster untested (only 1x2 strip) -> TestSinglePixel for disaggregate weighted/binary + pycnophylactic (degenerate no-shift smoothing) + dask parity. Cat2 MEDIUM: Inf weight collapses zone total to 0 (silent conservation break) -> TestInfWeight pins current behaviour. Cat4 MEDIUM: 3-class limiting_variable (multi-break + per-class caps) untested despite docstring -> TestLimitingVariableThreeClass. SOURCE BUG found (filed #3406, NOT fixed - test-only sweep): pycnophylactic raises ValueError (np.nanmax on zero-size array) when no pixel is valid for smoothing (all-NaN zones or no zone id in values); disaggregate handles same input gracefully (all-NaN). Pinned with TestPycnophylacticEmptyValid xfail(strict, raises=ValueError) -> flips red when #3406 fixed. LOW (documented, not fixed): non-square cellsize never exercised (all tests use res 0.5/0.5); disaggregate cupy/dask+cupy 1x1 + metadata not separately added (eager numpy gap was the real one, GPU dispatch already covered by TestCrossBackend)."
 diffusion,2026-06-20,3422,HIGH,1;2;3;4,"Pass 1 (2026-06-20, deep-sweep test-coverage, CUDA host). diffuse() dispatch table registers all 4 backends but test_diffusion.py only exercised numpy + dask+numpy. Cat 1 HIGH: cupy (_diffuse_cupy/_diffuse_step_gpu) and dask+cupy (_diffuse_dask_cupy/_diffuse_chunk_cupy) registered but never invoked -- no test ran them. Cat 4 HIGH: boundary accepts nan/nearest/reflect/wrap; only nearest+wrap tested, reflect had none. Cat 3 HIGH: 1x1 single-pixel and Nx1/1xN strip rasters never tested. Cat 2 MEDIUM: NaN tested numpy-only; Inf and all-NaN inputs untested. Filed #3422, added 14 tests (PR #3424, test-only, source untouched): cupy/dask+cupy parity vs numpy (incl. spatially-varying alpha + NaN propagation), reflect boundary across all 4 backends, 1x1 + Nx1 + 1xN (numpy + chunked dask strip), all-NaN stays NaN, Inf contamination smoke test. All 14 RAN+PASSED on a CUDA host; the 4 cupy/dask+cupy tests genuinely executed (not skipped); full file 39 passed. All paths verified correct before the tests were added -- coverage gap, not a bug. LOW (documented, not fixed): non-square cellsize (res[0]!=res[1]) never exercised -- diffuse uses res[0] as dx and assumes square cells; empty 0-row/0-col raster untested; asv benchmark absent; 'nan' boundary-mode edge=NaN behaviour not directly asserted on diffuse (covered indirectly via wrap/nearest)."
+fire,2026-06-25,,HIGH,2,"Deep-sweep 2026-06-25 test-coverage on a CUDA host. Backend matrix already complete: all 7 public funcs (dnbr/rdnbr/burn_severity_class/fireline_intensity/flame_length/rate_of_spread/kbdi) x 4 backends present and green (Cat 1 no gap). NaN covered (per-func nan_propagation + #3394 dtype parity). Cat 4 covered: rate_of_spread tests all 13 fuel models + invalid 0/14; kbdi annual_precip invalid 0/-100; fireline heat_content default+custom. Cat 5 covered via general_output_checks on every func. Found one gap: Cat 2 +Inf/-Inf inputs were untested on every function. Probed all 4 backends live: behavior is fully consistent and well-defined (no divergence, no bug) -- e.g. dnbr inf-inf->nan, burn_severity_class +inf->7/-inf->1, kbdi prev=inf clamps to 800, rate_of_spread slope=inf->nan. Added test-only regression: per-func numpy Inf contract (locks exact values) + 4-backend Inf parity (28 new tests, all RAN and PASSED on GPU). No source change; the kernels' only finite guard is v!=v so these lock that contract. Cat 3 1x1/strip: per-pixel kernels (no neighborhood window) so no degeneracy risk, and 1x1/1xN already exercised by kbdi/rdnbr/flame tests -> LOW, not added."
+flood,2026-06-25,,MEDIUM,1,"Deep-sweep 2026-06-25 test-coverage on a CUDA host. Module is densely tested (1051 test LOC vs 966 source). Backend matrix nearly complete: all 7 public funcs x 4 backends present and green EXCEPT vegetation_roughness mode='ndvi' on dask+cupy -- _veg_roughness_ndvi_dask_cupy was dispatched (flood.py:585) but never invoked by any test (nlcd dask+cupy, ndvi cupy, ndvi dask all tested). Cat 1 MEDIUM: added TestVegRoughnessDaskCuPy::test_ndvi_numpy_equals_dask_cupy mirroring the nlcd case; GPU-validated locally (passed, full file 89 passed). Cat 2 NaN well covered per-func incl #1104 (NaN curve_number) and #1437 (mannings_n DataArray) regressions; Inf inputs untested but low-risk (HAND/rainfall Inf -> NaN), not flagged. Cat 3 1xN strips + 1x1 covered for several funcs. Cat 5 metadata preserved is asserted on every backend test via general_output_checks (verify_attrs defaults True), so inundation/curve_number_runoff lacking a dedicated coords test is NOT a real gap. No source bugs found."
 focal,2026-06-10,3220;3219;3225,HIGH,1;2;3;4,"Deep-sweep 2026-06-10 on CUDA host, all 4 backends executed. Filed #3220 (coverage) and added 36 tests in PR branch: Inf inputs for mean/focal_stats (HIGH Cat2 - no Inf test existed anywhere), mean NaN input (HIGH Cat2 - default excludes=[nan] semantics never asserted), 1x1 + 1xN/Nx1 strips (HIGH Cat3), empty 0-row raster numpy-only (MEDIUM Cat3), mean passes=2 == mean(mean) and excludes sentinel -9999 behavioral tests (MEDIUM Cat4), dask+cupy non-default boundary modes for mean/apply/focal_stats (MEDIUM Cat1/4). Bugs surfaced, filed separately (NOT fixed here): #3219 hotspots silently returns all zeros on Inf input (nan global std passes the std==0 guard, all 4 backends); #3225 empty raster works on numpy but crashes cupy (raw CudaAPIError) and dask (map_overlap depth ValueError). hotspots+Inf and non-numpy empty behavior left unpinned until those are fixed. Backend matrix for the 4 public funcs was already solid (all 4 backends + parity); boundary modes covered except dask+cupy. Siblings filed #3214-3217 same day (dtype/docstring/apply-default-func) - no overlap."
-geotiff,2026-06-12,3266,MEDIUM,1,"Pass 22 (2026-06-12, deep-sweep test-coverage): delta audit of the ~20 commits since pass 21 (06-09..06-12, mostly pack/unpack fixes + #3241 GPU streaming writer + coregister #3254/#3248). Filed #3266 (tests). Cat 1 MEDIUM: pack=True gained working gpu/dask+gpu support in #3240, but three pack features were tested numpy+dask only: float32 width preservation (#3080, test_pack_float_width_3080.py), nodata kwarg fill (#3168, test_pack_nodata_kwarg_3168.py), band-subset per-band SCALE/OFFSET rewrite (#3161, test_pack_band_subset_3161.py). Live probe on this CUDA host: all six gpu/dask+gpu legs pass today (no source bug, pure coverage gap). Added one gpu/dask-gpu parametrized round-trip test per file (6 tests, requires_gpu, RUN+passing locally) and fixed two stale docstrings claiming unpack/pack is CPU-only (wrong since #3075/#3240). Verified NOT gaps this pass: #3128 int64 sentinel tests cover eager+dask+gpu; #3241 streaming writer landed with byte-identical band-first/band-last/BytesIO/small-buffer tests; #3104 scale-zero rejection has gpu legs; #3169 revived the dead compression-corpus oracle gate. Out of scope: coregister=True lives in accessor.py (excluded module); its multi-band + polar gaps are documented as experimental caveats in docs/source/reference/geotiff.rst (#3248). LOW (carried, documented not fixed): Inf as the declared nodata sentinel never tested. || PREVIOUS: Pass 21 (2026-06-09, deep-sweep test-coverage): filed #3114 (tests) + #3112 (source bug). Cat 1 HIGH: to_geotiff(pack=True) round-trip was tested only on numpy and dask+numpy (write/test_pack_3064.py); #3075 made unpack=True work on gpu and dask+gpu reads, but no test packed a GPU-read array back. Live probe on this CUDA host: BOTH GPU legs crash today -- eager gpu raises AttributeError (cupy has no astype, the known cupy 13.6/xarray 2025.12 where/astype incompat) and dask+gpu raises TypeError (numpy fill value inside cupy.where) -- both from _pack's out.fillna(nodata) in _attrs.py; _writers/gpu.py says the pre-dispatch re-pack is supposed to make every write path work. Source bug filed as #3112; test-only PR adds test_pack_round_trip_gpu (gpu + dask-gpu params, requires_gpu, xfail(strict=True) on #3112 so the fix flips them loudly) and fixes the stale module docstring claiming GPU rejects mask_and_scale. Ran on CUDA host: 13 passed, 2 xfailed. Verified NOT gaps this pass (probed before flagging): empty/zero-band writer guard is covered (test_basic.py 2075/2095 blocks incl. gpu + dask + streaming entry points); degenerate shapes covered on all 4 read backends (read/test_degenerate_shapes.py); overview_resampling all 7 modes parametrized; missing_sources raise/warn + invalid, band_nodata first/invalid, unpack on all 4 backends, masked/parse_coordinates/lock/cache/default_name/name-deprecation all exercised; attrs contract per-backend (attrs/test_contract.py). LOW (documented, not fixed): Inf as the declared nodata sentinel is never tested (only one nan+inf data round-trip in test_edge_cases.py). || PREVIOUS: Pass 20 (2026-06-06, deep-sweep test-coverage): filed #2984 and added test_writer.py degenerate-shape GPU write coverage (Cat 1 backend + Cat 3 geometric edge). Read side already covers 1x1/1xN/Nx1 on all 4 backends (read/test_degenerate_shapes.py) and the dask streaming writer covers them (integration/test_dask_pipeline.py); the GPU write path was the gap (smallest shape in gpu/test_writer.py was 2x2). Added test_write_geotiff_gpu_degenerate_round_trip (1x1/1xN/Nx1 x none/deflate) + test_to_geotiff_dask_gpu_degenerate_round_trip (dask+cupy via gpu=True). 9 new tests RUN+passing on a CUDA host. Verified paths work first (not a source bug); transform supplied explicitly via attrs. Wider tree audit (~92k test LOC vs ~33k source): rioxarray-compat (#2961), bbox NaN/Inf/rotated, 8-backend parity matrix, codec round-trips already covered -- no other real gaps. | Pass (2026-06-05 test-coverage sweep): mature module (~31k src / ~124k test LOC, 9 test dirs). Exhaustive existing coverage -- parity/test_backend_matrix.py runs all 4 backends + VRT + HTTP + fsspec; golden_corpus full-manifest parity; read_rioxarray_compat_2961 covers masked/mask_and_scale/parse_coordinates/default_name on eager+dask. Cat1+Cat3 gap found (MEDIUM): degenerate-shape READS (1x1/1xN/Nx1) were tested only on the eager numpy reader (test_edge_cases.py) and the dask streaming WRITE path (integration/test_dask_pipeline.py); the windowed dask READ (chunks=) and GPU READ (gpu=True) on a single-pixel dimension were never exercised (smallest dask-read source in read/test_tiling is 8x8/2x32, parity fixtures 32x32/64x64). Probed: paths work today, no source bug -- pure coverage gap. Added read/test_degenerate_shapes.py (18 tests): dask read x{chunks 1,3,4} x{1x1,1xN,Nx1} + coord/transform/crs parity + GPU read + dask+gpu read. GPU cells RAN and PASSED on this CUDA host (grid-size-1 launch validated). Fixture supplies explicit attrs['transform'] (writer cannot infer pixel size from a 1-element coord axis). Branch deep-sweep-test-coverage-geotiff-degenerate-read-01. NOTE: pre-existing union-merge CRLF/duplicate-record corruption in this CSV left untouched -- appended one clean record; DictReader last-write-wins picks this one."
+geotiff,2026-06-25,3518,MEDIUM,1,"Pass 23 (2026-06-25, deep-sweep test-coverage, CUDA host): delta audit of the ~15 geotiff commits since pass 22 (06-12..06-24): #3483 categorical PAM sidecar, #3375/#3376/#3380 xarray backend engine, #3373/#3374 chunked GPU read-once, #3371/#3372 reject predictor+lossy codec, #3331/#3332 reject zero/non-finite ModelPixelScale, #3327 gate dict gdal_metadata behind rich-tag opt-in, #3323/#3325 masked_nodata dtype-cast, #3277 pack nodata native width. Filed #3518 (tests). Cat 1 MEDIUM: the #3483 categorical PAM sidecar (_pam.py + _write_category_sidecar in _writers/eager.py) is round-trip tested only on the eager numpy write path (xrspatial/tests/test_rasterize_categorical_3482.py); the dask streaming (eager.py:1063) and GPU/nvCOMP (eager.py:880) write branches each have their own sidecar emit call that no test touches. Live probe on this CUDA host: dask + GPU categorical round-trips both emit the sidecar and read back names+RGBA colors today (no source bug, pure coverage gap). Added geotiff/tests/write/test_category_sidecar_backends_3483.py: dask write round-trip + GPU write round-trip (requires_gpu, RAN+passing locally) + names-only (category_colors=None build branch + names-only read path, never hit by the existing colors-always suite). 3 tests pass. Verified NOT gaps this pass: #3327 dict gdal_metadata gate has 43-line test_contract.py coverage; #3331/#3332 zero pixel scale covered by unit/test_degenerate_pixel_size_3331.py; #3371/#3372 predictor+lossy-codec reject has 2 test files; #3375/#3376/#3380 xarray engine covered by test_xarray_backend_3365.py + coregister 3376/3379; #3277 pack native width has 3 test files. LOW (carried, documented not fixed): Inf as the declared nodata sentinel still never tested. || Pass 22 (2026-06-12, deep-sweep test-coverage): delta audit of the ~20 commits since pass 21 (06-09..06-12, mostly pack/unpack fixes + #3241 GPU streaming writer + coregister #3254/#3248). Filed #3266 (tests). Cat 1 MEDIUM: pack=True gained working gpu/dask+gpu support in #3240, but three pack features were tested numpy+dask only: float32 width preservation (#3080, test_pack_float_width_3080.py), nodata kwarg fill (#3168, test_pack_nodata_kwarg_3168.py), band-subset per-band SCALE/OFFSET rewrite (#3161, test_pack_band_subset_3161.py). Live probe on this CUDA host: all six gpu/dask+gpu legs pass today (no source bug, pure coverage gap). Added one gpu/dask-gpu parametrized round-trip test per file (6 tests, requires_gpu, RUN+passing locally) and fixed two stale docstrings claiming unpack/pack is CPU-only (wrong since #3075/#3240). Verified NOT gaps this pass: #3128 int64 sentinel tests cover eager+dask+gpu; #3241 streaming writer landed with byte-identical band-first/band-last/BytesIO/small-buffer tests; #3104 scale-zero rejection has gpu legs; #3169 revived the dead compression-corpus oracle gate. Out of scope: coregister=True lives in accessor.py (excluded module); its multi-band + polar gaps are documented as experimental caveats in docs/source/reference/geotiff.rst (#3248). LOW (carried, documented not fixed): Inf as the declared nodata sentinel never tested. || PREVIOUS: Pass 21 (2026-06-09, deep-sweep test-coverage): filed #3114 (tests) + #3112 (source bug). Cat 1 HIGH: to_geotiff(pack=True) round-trip was tested only on numpy and dask+numpy (write/test_pack_3064.py); #3075 made unpack=True work on gpu and dask+gpu reads, but no test packed a GPU-read array back. Live probe on this CUDA host: BOTH GPU legs crash today -- eager gpu raises AttributeError (cupy has no astype, the known cupy 13.6/xarray 2025.12 where/astype incompat) and dask+gpu raises TypeError (numpy fill value inside cupy.where) -- both from _pack's out.fillna(nodata) in _attrs.py; _writers/gpu.py says the pre-dispatch re-pack is supposed to make every write path work. Source bug filed as #3112; test-only PR adds test_pack_round_trip_gpu (gpu + dask-gpu params, requires_gpu, xfail(strict=True) on #3112 so the fix flips them loudly) and fixes the stale module docstring claiming GPU rejects mask_and_scale. Ran on CUDA host: 13 passed, 2 xfailed. Verified NOT gaps this pass (probed before flagging): empty/zero-band writer guard is covered (test_basic.py 2075/2095 blocks incl. gpu + dask + streaming entry points); degenerate shapes covered on all 4 read backends (read/test_degenerate_shapes.py); overview_resampling all 7 modes parametrized; missing_sources raise/warn + invalid, band_nodata first/invalid, unpack on all 4 backends, masked/parse_coordinates/lock/cache/default_name/name-deprecation all exercised; attrs contract per-backend (attrs/test_contract.py). LOW (documented, not fixed): Inf as the declared nodata sentinel is never tested (only one nan+inf data round-trip in test_edge_cases.py). || PREVIOUS: Pass 20 (2026-06-06, deep-sweep test-coverage): filed #2984 and added test_writer.py degenerate-shape GPU write coverage (Cat 1 backend + Cat 3 geometric edge). Read side already covers 1x1/1xN/Nx1 on all 4 backends (read/test_degenerate_shapes.py) and the dask streaming writer covers them (integration/test_dask_pipeline.py); the GPU write path was the gap (smallest shape in gpu/test_writer.py was 2x2). Added test_write_geotiff_gpu_degenerate_round_trip (1x1/1xN/Nx1 x none/deflate) + test_to_geotiff_dask_gpu_degenerate_round_trip (dask+cupy via gpu=True). 9 new tests RUN+passing on a CUDA host. Verified paths work first (not a source bug); transform supplied explicitly via attrs. Wider tree audit (~92k test LOC vs ~33k source): rioxarray-compat (#2961), bbox NaN/Inf/rotated, 8-backend parity matrix, codec round-trips already covered -- no other real gaps. | Pass (2026-06-05 test-coverage sweep): mature module (~31k src / ~124k test LOC, 9 test dirs). Exhaustive existing coverage -- parity/test_backend_matrix.py runs all 4 backends + VRT + HTTP + fsspec; golden_corpus full-manifest parity; read_rioxarray_compat_2961 covers masked/mask_and_scale/parse_coordinates/default_name on eager+dask. Cat1+Cat3 gap found (MEDIUM): degenerate-shape READS (1x1/1xN/Nx1) were tested only on the eager numpy reader (test_edge_cases.py) and the dask streaming WRITE path (integration/test_dask_pipeline.py); the windowed dask READ (chunks=) and GPU READ (gpu=True) on a single-pixel dimension were never exercised (smallest dask-read source in read/test_tiling is 8x8/2x32, parity fixtures 32x32/64x64). Probed: paths work today, no source bug -- pure coverage gap. Added read/test_degenerate_shapes.py (18 tests): dask read x{chunks 1,3,4} x{1x1,1xN,Nx1} + coord/transform/crs parity + GPU read + dask+gpu read. GPU cells RAN and PASSED on this CUDA host (grid-size-1 launch validated). Fixture supplies explicit attrs['transform'] (writer cannot infer pixel size from a 1-element coord axis). Branch deep-sweep-test-coverage-geotiff-degenerate-read-01. NOTE: pre-existing union-merge CRLF/duplicate-record corruption in this CSV left untouched -- appended one clean record; DictReader last-write-wins picks this one."
 idw,2026-06-04,2919,HIGH,1;4,"cupy/dask+cupy backends untested (Cat1 HIGH); GPU k-reject error path untested (Cat4 MED). Added 6 GPU tests, validated on CUDA host. Inf-in-points (Cat2) and attrs-preservation (Cat5) are LOW, documented not fixed."
 interpolate,2026-06-12,3290,MEDIUM,2;3;4;5,"Deep-sweep 2026-06-12 on CUDA host. Backend coverage already complete: all 4 backends exercised for idw/kriging/spline incl. cross-backend equivalence and variance paths; no Cat 1 gaps. Filed #3290 for MEDIUM gaps, all verified correct-by-probe before filing (test-only fix): idw fill_value zero-weight branch (deterministic via 1e200 distance weight underflow; added numpy+dask+cupy, cupy RAN+PASSED), idw power only tested at default (exact oracle 10/(2^p+1)), spline collinear lstsq fallback, kriging duplicate points + all-equal-z (zero-variance variogram) + exactly-singular K regularisation retry (unit test on _build_kriging_matrix with all-zero variogram), spline/kriging 1x1 template, Inf/-Inf point filtering (only NaN was tested), lat/lon dim-name propagation (parametrized all 3 funcs), idw attrs preservation, 0-column template. Remaining minor untested: _build_kriging_matrix warn-then-NaN branch (needs mocked LinAlgError on retry). LOW documented not fixed: no asv benchmarks, non-uniform cell spacing unasserted. Full file 82 passed 0 skipped locally."
 interpolate-kriging,2026-06-04,2920;2921,HIGH,1;2;3;4;5,"Single public fn kriging(); all 4 backends already had cross-backend parity tests (numpy/cupy/dask+numpy/dask+cupy) incl. cupy & dask+cupy variance -- ran green on CUDA host. Gaps closed (test-only, #2921): Cat1 dask+numpy return_variance branch (_chunk_var) was untested -> added test_dask_return_variance_matches_numpy (atol=1e-12, var ~1e-14). Cat4 nlags only default(15) tested -> added non-default nlags=5 + invalid paths (nlags=0/-1 ValueError, nlags=2.5 TypeError). Cat2/3 two-point <3-lag-bins UserWarning branch -> test_two_point_warns_few_lag_bins. Cat2 all-NaN kriging input -> test_kriging_all_nan_points (only idw covered before). Cat5 output metadata (coords/dims/attrs/name) untested -> added test_output_metadata. Single-point kriging CRASHES (zero-size array reduction in _experimental_variogram, N=1) -- real source bug filed #2920; added xfail(strict, raises=ValueError) test_single_point documenting expected graceful behavior; source fix left to #2920 (test-only PR). LOW/not filed: singular-matrix K_inv-is-None all-NaN branch is defensive and unreachable via public API. GPU-validated."
@@ -14,6 +17,7 @@ interpolate_spline,2026-06-04,,HIGH,1;3;5,scope=spline-only; cupy+dask_cupy spli
 mcda,2026-06-10,3149,HIGH,1;2;5,"Pass 1 (2026-06-10, deep-sweep test-coverage): test_mcda.py had 175 tests, all numpy or dask+numpy -- zero cupy/dask+cupy coverage despite explicit cupy branches in standardize._get_xp and combine._sort_descending (Cat 1 HIGH). Filed #3149, added ~70 tests: cross-backend parity for standardize (7 methods) x cupy/dask+numpy/dask+cupy, combine (wlc/wpm/fuzzy and-or-sum-product-gamma/owa) x 3 backends, constrain, boolean_overlay, sensitivity OAT+MC on GPU backends; metadata preservation (attrs/coords/dims/name) for every stage (Cat 5 MEDIUM); wpm all-NaN criterion + Inf propagation through wlc/fuzzy-and (Cat 2 MEDIUM). All RUN on a CUDA host: 233 passed, 11 xfailed. Probing surfaced real source bugs already filed by sibling sweeps as #3146 (owa raises on ALL dask backends -- _sort_descending calls nonexistent da.sort; owa cupy mixes numpy order weights into cupy stack; piecewise standardize broken on cupy + dask+cupy and categorical on dask+cupy via np.asarray on cupy chunks; monte_carlo sensitivity reads .values on cupy data) and #3147 (constrain drops attrs when masks applied) -- those paths pinned with strict xfail markers to flip on fix; constrain cupy/dask+cupy xfail(strict=False) on the known cupy 13.6 + xarray xr.where dependency incompat, not an mcda bug. Source untouched (test-only PR). LOW (documented, not fixed): name= output parameter untested across combine functions; empty (0-row) raster untested -- elementwise ops, judged low value. weights.py (ahp/rank) is pure-numpy metadata, backend matrix N/A, already well covered."
 morphology,2026-06-20,3404,MEDIUM,2;3,"Added Inf/-Inf, all-NaN, Nx1/1xN strip, integer-dtype tests; source already correct, regression guards only; cupy + dask+cupy ran on GPU host"
 multispectral,2026-06-20,3431,MEDIUM,2;3;4,true_color NaN/alpha + all-equal range_val==0 + nondefault nodata/c/th; evi & savi validation error paths; GPU tests ran (cupy+dask+cupy)
+perlin,2026-06-23,,HIGH,3;4;5,"added tests: seed/freq/name params, single-row, 1x1+Nx1 degenerate (pins all-NaN bug), dims+attrs preserve, coords-drop (pins bug); GPU tests ran on CUDA host; 2 source bugs surfaced (coords dropped; 1x1/Nx1 all-NaN ptp div0) need issues filed (gh issue create blocked by auto-mode)"
 polygon_clip,2026-06-10,3197,MEDIUM,1;2;3;5,"deep-sweep test-coverage 2026-06-10 on a CUDA host. Existing file covered numpy well + one parity test per dask/cupy/dask+cupy backend. Filed #3197 (test-only) and added 13 tests: Cat1 GPU param/NaN coverage (cupy + dask+cupy each get custom nodata, all_touched=True, and NaN-input preservation vs numpy; previously only crop=False inner polygon ran on GPU); Cat2 Inf/-Inf preserved, all-NaN input -> all-NaN, int32 + sentinel nodata=-1; Cat3 Nx1 + 1xN strip rasters; Cat5 coords preserved (crop=False) + crop coords are a contiguous subset of input coords (crop=True). All 13 RAN+PASSED on GPU (6 GPU tests not skipped); full file 36 passed 0 skipped. LOW (documented, NOT fixed): rasterize_kw forwarding never tested; non-square cellsize never tested. SOURCE NOTE (out of scope, not filed): clip_polygon docstring says 'named y and x dims' but rasterize() hard-requires literal y/x, so lat/lon-dim rasters raise -- dim-name preservation (Cat5) is therefore unsupported by contract, not a test gap. SOURCE NOTE 2: polygon_clip.py:216 still passes rasterize(use_cuda=True) on dask+cupy (renamed to gpu= in #3089); harmless deprecation alias today, candidate for an api-consistency follow-up."
 polygonize,2026-06-12,3299,MEDIUM,1,"Pass 4 (2026-06-12): added test_polygonize_mask_chunk_mismatch_3299.py (25 tests, all passing on a CUDA host incl. dask+cupy). Closes Cat 1 MEDIUM: the _polygonize_dask mask-rechunk branch (mask_data.chunks != dask_data.chunks -> rechunk) was never exercised; every prior dask masked test used mask chunks identical to the raster's. Mismatched layouts pinned against same-backend aligned-mask reference: (6,7) same-grid-shape misalignment (the silent-corruption layout for int rasters), single-chunk (15,18), more-blocks (4,5); int+float rasters, connectivity 4/8, dask+numpy and dask+cupy; plus exact-geometry single-masked-pixel hole anchor. Mutation (delete the rechunk guard) flips all 25 red; clean md5 restore. Full polygonize suite 486 passed / 16 skipped. Test-only; source untouched. Issue #3299. Audit re-confirmed Cat 2/3/4 closed by passes 1-3 and post-2026-05-29 changes (#2913 float-mask fix flipped prior xfails, #3041 has issue-2677 test file, #2673/#2817 covered by batch-invariance and heap tests); Cat 5 N/A (no DataArray output; CRS/transform propagation already tested). | Pass 3 (2026-05-29): added test_polygonize_mask_dtype_coverage_2026_05_29.py (41 passed, 8 xfailed on a CUDA host). Closes Cat 4 MEDIUM parameter-coverage gap: mask= is documented to accept bool/integer/float values but every prior test passed only a bool mask. Integer masks (int32/int64) now pinned against the same-backend bool-mask output on all four backends x both raster dtypes x connectivity 4/8; float-mask-on-integer-raster also pinned. Each backend is compared to its OWN bool reference to isolate mask-dtype from the unrelated numpy-vs-dask hole-vs-single-ring representation difference. Mutation (drop the not-mask[ij] exclusion in _calculate_regions) flips 11 tests red incl. the pixel-exclusion sanity anchor; clean md5 restore. Surfaced source bug #2623: a float-dtype mask on a float-dtype raster raises TypeError at polygonize.py:918 (mask & nan_mask; bitwise_and undefined for float&bool; cupy/dask route floats through _polygonize_numpy so they crash too; int masks coerce fine). 8 float-mask cases marked xfail(strict, raises=TypeError) referencing #2623. Test-only; source untouched. | Pass 2 (2026-05-27): added test_polygonize_atol_rtol_backend_coverage_2026_05_27.py with 15 tests, all passing on a CUDA host. Closes Cat 4 MEDIUM parameter-coverage gap on atol/rtol forwarding through the cupy and dask+cupy backends. atol/rtol were exposed by #2173 / #2194 and thread through _polygonize_cupy (polygonize.py:808) and _polygonize_dask (polygonize.py:1719); the dask path further plumbs them into dask.delayed(_polygonize_chunk)(...) at lines 1748-1754 and into _bucket_key_for_value for cross-chunk merge bucketing at lines 1757-1758. Pre-existing tests covered non-default atol/rtol only on numpy and dask+numpy. The cupy and dask+cupy dispatchers were untested -- a regression dropping the kwargs there would silently change the float polygon count and would not be caught. Same dispatcher-silently-drops-kwarg pattern fixed by #1561 / #1605 / #1685 / #1810 / #1974 on adjacent GeoTIFF surfaces. 15 tests: cupy strict-equality + default-tolerance pin on _REPRO_2173, dask+cupy strict-equality single-chunk + multi-chunk (engages cross-chunk merge bucket) + default-tolerance multi-chunk pin, cupy intermediate-atol small/large pair, dask+cupy intermediate-atol single/multi-chunk small + single-chunk large, cupy integer atol-ignored matrix, dask+cupy integer atol-ignored single-chunk + multi-chunk, cupy rtol-only large/small matrix. Mutation against _polygonize_cupy float branch (drop atol/rtol kwargs in the _polygonize_numpy forward call at polygonize.py:823-825) flips 3 of 5 cupy tests red; mutation against dask.delayed(_polygonize_chunk)(...) at polygonize.py:1748-1754 (drop atol, rtol args) flips 2 of 6 dask+cupy tests red. Confirmed clean restore via md5sum. Source untouched. Filed issue #2537 (test-only). Cat 4 MEDIUM (parameter coverage on cupy + dask+cupy atol/rtol forwarding). Pass 1 (2026-05-19): added test_polygonize_coverage_2026_05_19.py with 58 tests, all passing on a CUDA host. Closes Cat 3 HIGH 1x1 / Nx1 single-column geometric gaps (Nx1 exercises the nx==1 padding path at polygonize.py:565 and the cupy nx==1 numpy-fallback at polygonize.py:671), Cat 3 MEDIUM 1xN single-row and all-equal-value rasters on all four backends. Closes Cat 2 HIGH NaN parity for cupy + dask+cupy (numpy/dask were already covered by test_polygonize_nan_pixels_excluded*), Cat 2 MEDIUM all-NaN raster on all four backends, Cat 2 HIGH +/-Inf pins on all four backends. Filed source-bug issue #2155: numpy/dask/dask+cupy backends silently absorb Inf cells into adjacent finite polygons because _is_close reduces abs(inf-inf) to nan; cupy backend handles Inf correctly. Pins lock the asymmetric behaviour so the fix is visible. Closes Cat 1 MEDIUM simplify_tolerance + mask= parity gaps on dask+cupy backend (numpy/cupy/dask were already covered). Closes Cat 4 MEDIUM column_name non-default value across geopandas/spatialpandas/geojson return types and Cat 4 MEDIUM validation error paths (bad connectivity, bad transform length, mask shape mismatch, mask underlying-type mismatch). Cat 5 N/A: polygonize returns lists/dataframes, not a DataArray with attrs to propagate."
 proximity,2026-06-18,2692;3139,MEDIUM,1;4,"Pass 4 (2026-06-18, deep-sweep test-coverage): 1 MEDIUM, 1 LOW. MEDIUM (Cat 4/Cat 1): all three public funcs are @supports_dataset and document Dataset-in/Dataset-out, but no test ever passed a Dataset; the shared decorator is covered generically in test_dataset_support.py which never lists proximity/allocation/direction, so per-variable _process dispatch + attrs/coords round-trip + result.name=None reset were unpinned. Added test_dataset_input_processes_each_variable (3 funcs x 4 backends, 12 tests); numpy+dask+cupy+dask+cupy all RUN and PASS on this CUDA host, expected built from numpy baseline to avoid implicit cupy host conversion. Verified working first -- no source bug, source untouched. Full file 528 passed. LOW (documented, not fixed): public exports euclidean_distance / manhattan_distance have no direct unit test (only great_circle_distance does); both are trivial pure fns exercised indirectly through proximity. || Pass 3 (2026-06-09, deep-sweep test-coverage): module grew since Pass 2 (#2807 metric validation, #2812 GREAT_CIRCLE brute force, #2850/#2851 input validation, #2854/#2908 halo fixes, tie-break routing) and each landed with its own tests; Pass 2's stale LOW (invalid distance_metric fallback) is FIXED and tested (#2807). Found 3 MEDIUM gaps, filed #3139, added 40 tests (all RUN and PASS on a CUDA host; full file 450 passed): (1) Cat 2 integer-dtype raster untested on any backend -- bounded dask pads int arrays with boundary=np.nan which casts to INT_MIN phantom targets, only neutralized because the coordinate-grid pads are real NaNs; pinned int32 x 3 funcs x 4 backends x bounded/unbounded vs float64 numpy baseline + explicit target_values; (2) Cat 1 bounded dask+cupy (_process_dask_cupy) only ever ran EUCLIDEAN; pinned MANHATTAN+GREAT_CIRCLE x 3 funcs with a routing spy; mutation (pad=0) flips all 6 red, clean md5 restore; (3) Cat 3 empty 0-row/0-col raster unpinned; fails fast with IndexError, pinned raises. All behaviors verified correct before tests were added -- no source bug, source untouched. LOW (documented, not fixed): -inf pixel input never tested (+inf is; isfinite is symmetric). || Pass 2 (2026-06-02): added 18 tests to test_proximity.py closing the two MEDIUM gaps Pass 1 left open, all RUN and passing on a CUDA host across numpy/cupy/dask+numpy/dask+cupy (15 cross-backend + 3 error-path). Source untouched. Cat 4 MEDIUM (error path): _process raises ValueError when raster.dims != (y, x) (proximity.py:1043) but no test exercised the swapped x/y guard; test_wrong_dim_order_raises pins it for proximity/allocation/direction. Cat 2 MEDIUM (all-NaN input): Pass 1 noted all-NaN/all-zero on eager numpy+cupy was unpinned; test_all_nan_raster_all_nan_output pins an all-NaN 6x6 raster -> all-NaN float32 output on all four backends x three functions. Remaining LOW (documented): invalid distance_metric string silently falls back to EUCLIDEAN (proximity.py:1049-1051). || PREVIOUS: Pass 1 (2026-05-29): added 65 tests to test_proximity.py closing three coverage gaps, all RUN and passing on a CUDA host (numpy/cupy/dask+numpy/dask+cupy). Issue #2692, PR opened. Source untouched. Cat 3 HIGH: degenerate raster shapes (1x1 single pixel, Nx1 column strip, 1xN row strip) had zero coverage for proximity/allocation/direction on any backend; they stress the line-sweep kernel boundaries (_process_proximity_line) and the GPU brute-force kernel grid sizing (_proximity_cuda_kernel via cuda_args). Pinned all three shapes x three functions x four backends against hand-checked expected values; mutation of a pinned direction expectation confirms teeth. Cat 1/4 HIGH: allocation and direction only ran EUCLIDEAN across backends; MANHATTAN and GREAT_CIRCLE were cross-backend-tested for proximity only. Pinned both metrics x two functions x four backends against the numpy baseline (all match). Cat 5 MEDIUM: no test set non-empty res/crs attrs so the attrs-preservation assertion in general_output_checks compared two empty dicts. proximity reads attrs['res'] via get_dataarray_resolution for bounded-dask chunk padding, so added attrs round-trip tests on four backends plus a bounded-dask test where a res attr matching the coordinate spacing must equal the numpy baseline. A res attr that lies about the spacing mis-sizes the map_overlap depth; source fragility, not a test gap, left for a separate accuracy issue. Cat 2 (NaN/Inf input) already covered by the shared test_raster fixture (embeds np.inf and np.nan, runs on four backends). Remaining LOW: all-NaN / all-zero input on eager numpy+cupy not directly pinned."

{xarray_spatial-0.10.11 → xarray_spatial-0.10.13}/CHANGELOG.md

@@ -2,6 +2,66 @@
 -----------
 
 
+### Version 0.10.13 - 2026-06-26
+
+#### New features
+- geotiff: write continuous-raster symbology sidecars (.qml + PAM statistics) from to_geotiff (#3537) (#3538)
+- templates: add list_templates() to discover from_template names (#3536)
+- templates: add major world cities as bounding-box templates in from_template (#3534)
+- fire: accept m/min spread rate in fireline_intensity by default (#3527) (#3529)
+- hydrology: make routing the sole public hydrology API surface (#3528) (#3530)
+- accessor: add coregister=True to .xrs.rasterize (#3492) (#3493)
+
+#### Bug fixes and improvements
+- flood: fix float32 dtype leak in flood_depth/curve_number_runoff dask backends (#3499) (#3509)
+- flood: fix dask backends emitting float32 output for float32 input (#3498)
+- flood: make mannings_n DataArray validation lazy-safe (#3503) (#3507)
+- flood: fix E305 and collapse utils import (#3504)
+- flood: add missing reference entries, examples, and backend notes (#3502)
+- binary: output float32 on all backends (#3514)
+- binary: type the dask+cupy meta so the lazy dtype matches float32 (#3508) (#3513)
+- geotiff: stop a short-row PAM RAT from crashing open_geotiff (#3522)
+- geotiff: add Examples section to to_geotiff docstring (#3521)
+- geotiff: cover categorical PAM sidecar round-trip on dask and GPU write paths (#3518) (#3519)
+- geotiff: join backslash-continued import in test (isort) (#3517)
+- templates: use CF Conventions metadata instead of crs_units/crs_name (#3532)
+- templates: fill all-NaN templates in generate_terrain (numpy/cupy) (#3526)
+- templates: honor requested resolution exactly in from_template() (#3494) (#3495)
+- classify: fix stale example outputs and add missing examples (#3516)
+- classify: cover all-NaN degenerate input across classifiers (#3515)
+- classify: document num_sample=None in natural_breaks docstring (#3505)
+- fire: add Examples sections and backend-support notes to docstrings (#3512)
+- fire: add Inf/-Inf regression tests (#3500)
+- multispectral: test vegetation_roughness ndvi on dask+cupy backend (#3497)
+
+
+### Version 0.10.12 - 2026-06-25
+
+#### New features
+- templates: add from_template() for empty study-area DataArrays over common regions (#3484) (#3487)
+- accessor: add .xrs.validate() to check a raster against the xarray-spatial contract (#3486)
+- accessor: add a categorized repr to the .xrs accessor (#3476) (#3477)
+- accessor: forward per-tool repr to the accessed tool (#3478) (#3479)
+- interpolation: accept GeoDataFrames and expose on the .xrs accessor with coregister (#3481)
+- rasterize: support string/categorical columns with QGIS-visible labels (#3483)
+
+#### Bug fixes and improvements
+- proximity: fix EUCLIDEAN proximity/allocation/direction dropping targets at exact max_distance (#3443)
+- perlin: preserve input x/y coordinates on the output (#3470)
+- perlin: preserve input dtype on the dask backends (#3471)
+- perlin: fix dask OOM by dropping a redundant dask.persist (#3469) (#3473)
+- perlin: document the name parameter in the docstring (#3468)
+- perlin: document name param, float-dtype requirement, and ValueError (#3467)
+- perlin: drop unused not_implemented_func import, fix isort wrap (#3466)
+- perlin: add test coverage for params, degenerate shapes, and metadata (#3472)
+- perlin: fix stale test_perlin coords assertion after coords preservation (#3489)
+- generate_terrain: preserve caller coords/res/crs (#3474) (#3475)
+- least_cost_corridor: propagate friction geo-attrs (#3446) (#3449)
+- corridor: validate matching shapes for precomputed surfaces (#3447)
+- corridor: add strip, cross-backend, param, and metadata test coverage (#3448)
+- fire: add test coverage for dask+cupy dispatch and metadata preservation (#3444)
+
+
 ### Version 0.10.11 - 2026-06-21
 
 #### Bug fixes and improvements