From c700c2a5e36adcdf053c3fef2f56207a3bc747f8 Mon Sep 17 00:00:00 2001
From: ghostiee-11 <168410465+ghostiee-11@users.noreply.github.com>
Date: Fri, 20 Mar 2026 16:47:40 +0530
Subject: [PATCH 1/2] Skip string coordinates in partition bound computation

String-typed coordinates (dtype U, S, O) crash in _block_metadata()
because numpy min/max ufuncs do not support string comparison.
Since ScalarBound only supports Int64/Float64/TimestampNanos, skip
string coordinates so the Rust pruning logic treats the dimension
conservatively (never prunes on it).

Partial fix for #121.
---
 tests/test_sql.py | 19 +++++++++++++++++++
 xarray_sql/df.py  | 44 +++++++++++++++++++++++++-------------------
 2 files changed, 44 insertions(+), 19 deletions(-)

diff --git a/tests/test_sql.py b/tests/test_sql.py
index 08bde14..b4e2bfd 100644
--- a/tests/test_sql.py
+++ b/tests/test_sql.py
@@ -1,5 +1,6 @@
 """SQL functionality tests for xarray-sql using pytest."""
 
+import numpy as np
 import pytest
 import xarray as xr
 
@@ -127,3 +128,21 @@ def test_cross_join(air_and_stations):
       "SELECT COUNT(*) AS total FROM air_data CROSS JOIN stations"
   ).to_pandas()
   assert result["total"].iloc[0] > 0
+
+
+def test_string_coordinates():
+  """String-typed coordinates should not crash during registration."""
+  ds = xr.Dataset(
+      {"score": (["student", "subject"], np.random.rand(3, 2))},
+      coords={
+          "student": ["alice", "bob", "charlie"],
+          "subject": ["math", "science"],
+      },
+  )
+  ctx = XarrayContext()
+  ctx.from_dataset("scores", ds.chunk({"student": 3, "subject": 2}))
+  result = ctx.sql("SELECT * FROM scores").to_pandas()
+  assert len(result) == 6
+  assert "student" in result.columns
+  assert "subject" in result.columns
+  assert "score" in result.columns
diff --git a/xarray_sql/df.py b/xarray_sql/df.py
index 756de20..d3f050f 100644
--- a/xarray_sql/df.py
+++ b/xarray_sql/df.py
@@ -323,25 +323,31 @@ def _block_metadata(coord_arrays: dict, block: Block) -> PartitionBounds:
   ranges: PartitionBounds = {}
   for dim, slc in block.items():
     coord_values = coord_arrays[str(dim)][slc]
-    if len(coord_values) > 0:
-      # Use actual min/max rather than first/last so that non-monotonic
-      # coordinate axes (e.g. descending latitude 90→-90) are handled
-      # correctly.  np.min/max work for both numeric and datetime64 arrays.
-      min_val = coord_values.min()
-      max_val = coord_values.max()
-
-      if isinstance(min_val, (np.datetime64, pd.Timestamp)):
-        min_val = int(pd.Timestamp(min_val).value)
-        max_val = int(pd.Timestamp(max_val).value)
-        ranges[str(dim)] = (min_val, max_val, "timestamp_ns")
-      elif hasattr(min_val, "item"):
-        min_val = min_val.item()
-        max_val = max_val.item()
-        dtype = "float64" if isinstance(min_val, float) else "int64"
-        ranges[str(dim)] = (min_val, max_val, dtype)
-      else:
-        dtype = "float64" if isinstance(min_val, float) else "int64"
-        ranges[str(dim)] = (min_val, max_val, dtype)
+    if len(coord_values) == 0:
+      continue
+    # String/object dtypes are not representable as ScalarBound
+    # (Int64/Float64/TimestampNanos) and numpy min/max ufuncs do not
+    # support them.  Skip so pruning treats the dimension conservatively.
+    if coord_values.dtype.kind in ('U', 'S', 'O'):
+      continue
+    # Use actual min/max rather than first/last so that non-monotonic
+    # coordinate axes (e.g. descending latitude 90→-90) are handled
+    # correctly.  np.min/max work for both numeric and datetime64 arrays.
+    min_val = coord_values.min()
+    max_val = coord_values.max()
+
+    if isinstance(min_val, (np.datetime64, pd.Timestamp)):
+      min_val = int(pd.Timestamp(min_val).value)
+      max_val = int(pd.Timestamp(max_val).value)
+      ranges[str(dim)] = (min_val, max_val, "timestamp_ns")
+    elif hasattr(min_val, "item"):
+      min_val = min_val.item()
+      max_val = max_val.item()
+      dtype = "float64" if isinstance(min_val, float) else "int64"
+      ranges[str(dim)] = (min_val, max_val, dtype)
+    else:
+      dtype = "float64" if isinstance(min_val, float) else "int64"
+      ranges[str(dim)] = (min_val, max_val, dtype)
   return ranges
 
 

From e57a2b430aea42db0753eeb038bfd2997617249b Mon Sep 17 00:00:00 2001
From: ghostiee-11 <168410465+ghostiee-11@users.noreply.github.com>
Date: Sat, 21 Mar 2026 03:11:20 +0530
Subject: [PATCH 2/2] style: use double quotes to satisfy pyink formatter

The project uses pyink-use-majority-quotes = true, which enforces
double quotes. Single quotes in the string dtype check caused the
lint CI to fail.
---
 xarray_sql/df.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/xarray_sql/df.py b/xarray_sql/df.py
index d3f050f..0945f06 100644
--- a/xarray_sql/df.py
+++ b/xarray_sql/df.py
@@ -328,7 +328,7 @@ def _block_metadata(coord_arrays: dict, block: Block) -> PartitionBounds:
     # String/object dtypes are not representable as ScalarBound
     # (Int64/Float64/TimestampNanos) and numpy min/max ufuncs do not
     # support them.  Skip so pruning treats the dimension conservatively.
-    if coord_values.dtype.kind in ('U', 'S', 'O'):
+    if coord_values.dtype.kind in ("U", "S", "O"):
       continue
     # Use actual min/max rather than first/last so that non-monotonic
     # coordinate axes (e.g. descending latitude 90→-90) are handled