Iterating from SOMADataFrame

Author: johnkerl

.gitignore

@@ -52,5 +52,6 @@ apis/python/src/tiledbsoma/libtiledb.*
 apis/python/src/tiledbsoma/libtiledbsoma.*
 
 /.quarto/
+/tags
 
 /NOTES/

apis/python/src/tiledbsoma/soma_dataframe.py

@@ -5,8 +5,11 @@
 import pyarrow as pa
 import tiledb
 
+import tiledbsoma.libtiledbsoma as clib
+
 from . import util, util_arrow, util_tiledb
 from .logging import log_io
+from .query_condition import QueryCondition
 from .soma_collection import SOMACollectionBase
 from .tiledb_array import TileDBArray
 from .types import Ids, NTuple
@@ -157,6 +160,55 @@ def is_indexed(self) -> Literal[False]:
     def get_index_column_names(self) -> Sequence[str]:
         return []
 
+    def read_using_lib_temp(
+        self,
+        *,
+        # TODO: find the right syntax to get the typechecker to accept args like ``ids=slice(0,10)``
+        # ids: Optional[Union[Sequence[int], Slice]] = None,
+        ids: Optional[Any] = None,
+        value_filter: Optional[str] = None,
+        column_names: Optional[Sequence[str]] = None,
+        result_order: Optional[str] = None,
+        # TODO: batch_size
+        # TODO: partition,
+        # TODO: platform_config,
+    ) -> Iterator[pa.Table]:
+        """
+        TODO: copy the text
+        """
+
+        with self._tiledb_open("r") as A:
+            dim_names, attr_names = util_tiledb.split_column_names(
+                A.schema, column_names
+            )
+
+            query_condition = None
+            if value_filter is not None:
+                # query_condition = tiledb.QueryCondition(value_filter)
+                query_condition = QueryCondition(value_filter)
+
+            # As an arg to this method, `column_names` is optional-None. For the pybind11
+            # code it's optional-[].
+            lib_column_names = [] if column_names is None else column_names
+
+            sr = clib.SOMAReader(
+                self._uri,
+                name=self.name,
+                schema=A.schema,  # query_condition needs this
+                column_names=lib_column_names,
+                query_condition=query_condition,
+            )
+
+            # TODO: platform_config
+            # TODO: batch_size
+            # TODO: result_order
+
+            sr.submit()
+
+            while arrow_table := sr.read_next():
+                # yield util_arrow.ascii_to_unicode_pyarrow_readback(batch)
+                yield arrow_table  # XXX what other post-processing
+
     def read(
         self,
         *,

apis/python/tests/test_soma_dataframe.py

@@ -192,7 +192,7 @@ def _check_tbl(tbl, col_names, ids):
         print(r)
         print("INPUT SCHEMA")
         print(i)
-        print("CMP", r==i)
+        print("CMP", r == i)
 
         print("")