[python] support table scan with row range

Author: universe-hcy

paimon-python/pypaimon/read/scanner/full_starting_scanner.py

@@ -63,6 +63,8 @@ def __init__(self, table, predicate: Optional[Predicate], limit: Optional[int]):
 
         self.idx_of_this_subtask = None
         self.number_of_para_subtasks = None
+        self.start_row_of_this_subtask = None
+        self.end_row_of_this_subtask = None
 
         self.only_read_real_buckets = True if options.bucket() == BucketMode.POSTPONE_BUCKET.value else False
         self.data_evolution = options.data_evolution_enabled()
@@ -125,33 +127,16 @@ def with_shard(self, idx_of_this_subtask, number_of_para_subtasks) -> 'FullStart
         self.number_of_para_subtasks = number_of_para_subtasks
         return self
 
-    def _append_only_filter_by_shard(self, partitioned_files: defaultdict) -> (defaultdict, int, int):
-        """
-        Filter file entries by shard. Only keep the files within the range, which means
-        that only the starting and ending files need to be further divided subsequently
-        """
-        total_row = 0
-        # Sort by file creation time to ensure consistent sharding
-        for key, file_entries in partitioned_files.items():
-            for entry in file_entries:
-                total_row += entry.file.row_count
-
-        # Calculate number of rows this shard should process using balanced distribution
-        # Distribute remainder evenly among first few shards to avoid last shard overload
-        base_rows_per_shard = total_row // self.number_of_para_subtasks
-        remainder = total_row % self.number_of_para_subtasks
-
-        # Each of the first 'remainder' shards gets one extra row
-        if self.idx_of_this_subtask < remainder:
-            num_row = base_rows_per_shard + 1
-            start_row = self.idx_of_this_subtask * (base_rows_per_shard + 1)
-        else:
-            num_row = base_rows_per_shard
-            start_row = (remainder * (base_rows_per_shard + 1) +
-                         (self.idx_of_this_subtask - remainder) * base_rows_per_shard)
-
-        end_row = start_row + num_row
+    def with_row_range(self, start_row, end_row) -> 'FullStartingScanner':
+        if start_row >= end_row:
+            raise Exception("start_row must be less than end_row")
+        self.start_row_of_this_subtask = start_row
+        self.end_row_of_this_subtask = end_row
+        return self
 
+    def _append_only_filter_by_row_range(self, partitioned_files: defaultdict,
+                                         start_row: int,
+                                         end_row: int) -> (defaultdict, int, int):
         plan_start_row = 0
         plan_end_row = 0
         entry_end_row = 0  # end row position of current file in all data
@@ -183,12 +168,16 @@ def _append_only_filter_by_shard(self, partitioned_files: defaultdict) -> (defau
 
         return filtered_partitioned_files, plan_start_row, plan_end_row
 
-    def _data_evolution_filter_by_shard(self, partitioned_files: defaultdict) -> (defaultdict, int, int):
+    def _append_only_filter_by_shard(self, partitioned_files: defaultdict) -> (defaultdict, int, int):
+        """
+        Filter file entries by shard. Only keep the files within the range, which means
+        that only the starting and ending files need to be further divided subsequently
+        """
         total_row = 0
+        # Sort by file creation time to ensure consistent sharding
         for key, file_entries in partitioned_files.items():
             for entry in file_entries:
-                if not self._is_blob_file(entry.file.file_name):
-                    total_row += entry.file.row_count
+                total_row += entry.file.row_count
 
         # Calculate number of rows this shard should process using balanced distribution
         # Distribute remainder evenly among first few shards to avoid last shard overload
@@ -206,6 +195,11 @@ def _data_evolution_filter_by_shard(self, partitioned_files: defaultdict) -> (de
 
         end_row = start_row + num_row
 
+        return self._append_only_filter_by_row_range(partitioned_files, start_row, end_row)
+
+    def _data_evolution_filter_by_row_range(self, partitioned_files: defaultdict,
+                                            start_row: int,
+                                            end_row: int) -> (defaultdict, int, int):
         plan_start_row = 0
         plan_end_row = 0
         entry_end_row = 0  # end row position of current file in all data
@@ -244,6 +238,30 @@ def _data_evolution_filter_by_shard(self, partitioned_files: defaultdict) -> (de
 
         return filtered_partitioned_files, plan_start_row, plan_end_row
 
+    def _data_evolution_filter_by_shard(self, partitioned_files: defaultdict) -> (defaultdict, int, int):
+        total_row = 0
+        for key, file_entries in partitioned_files.items():
+            for entry in file_entries:
+                if not self._is_blob_file(entry.file.file_name):
+                    total_row += entry.file.row_count
+
+        # Calculate number of rows this shard should process using balanced distribution
+        # Distribute remainder evenly among first few shards to avoid last shard overload
+        base_rows_per_shard = total_row // self.number_of_para_subtasks
+        remainder = total_row % self.number_of_para_subtasks
+
+        # Each of the first 'remainder' shards gets one extra row
+        if self.idx_of_this_subtask < remainder:
+            num_row = base_rows_per_shard + 1
+            start_row = self.idx_of_this_subtask * (base_rows_per_shard + 1)
+        else:
+            num_row = base_rows_per_shard
+            start_row = (remainder * (base_rows_per_shard + 1) +
+                         (self.idx_of_this_subtask - remainder) * base_rows_per_shard)
+
+        end_row = start_row + num_row
+        return self._data_evolution_filter_by_row_range(partitioned_files, start_row, end_row)
+
     def _compute_split_start_end_row(self, splits: List[Split], plan_start_row, plan_end_row):
         """
         Find files that needs to be divided for each split
@@ -451,7 +469,13 @@ def _create_append_only_splits(
         for entry in file_entries:
             partitioned_files[(tuple(entry.partition.values), entry.bucket)].append(entry)
 
-        if self.idx_of_this_subtask is not None:
+        if self.start_row_of_this_subtask is not None:
+            # shard data range: [plan_start_row, plan_end_row)
+            partitioned_files, plan_start_row, plan_end_row = \
+                self._append_only_filter_by_row_range(partitioned_files,
+                                                      self.start_row_of_this_subtask,
+                                                      self.end_row_of_this_subtask)
+        elif self.idx_of_this_subtask is not None:
             partitioned_files, plan_start_row, plan_end_row = self._append_only_filter_by_shard(partitioned_files)
 
         def weight_func(f: DataFileMeta) -> int:
@@ -467,7 +491,7 @@ def weight_func(f: DataFileMeta) -> int:
             packed_files: List[List[DataFileMeta]] = self._pack_for_ordered(data_files, weight_func,
                                                                             self.target_split_size)
             splits += self._build_split_from_pack(packed_files, file_entries, False, deletion_files_map)
-        if self.idx_of_this_subtask is not None:
+        if self.start_row_of_this_subtask is not None or self.idx_of_this_subtask is not None:
             # When files are combined into splits, it is necessary to find files that needs to be divided for each split
             self._compute_split_start_end_row(splits, plan_start_row, plan_end_row)
         return splits
@@ -555,7 +579,14 @@ def sort_key(manifest_entry: ManifestEntry) -> tuple:
         for entry in sorted_entries:
             partitioned_files[(tuple(entry.partition.values), entry.bucket)].append(entry)
 
-        if self.idx_of_this_subtask is not None:
+        print("self.start_row_of_this_subtask:{}".format(self.start_row_of_this_subtask))
+        if self.start_row_of_this_subtask is not None:
+            # shard data range: [plan_start_row, plan_end_row)
+            partitioned_files, plan_start_row, plan_end_row = \
+                self._data_evolution_filter_by_row_range(partitioned_files,
+                                                         self.start_row_of_this_subtask,
+                                                         self.end_row_of_this_subtask)
+        elif self.idx_of_this_subtask is not None:
             # shard data range: [plan_start_row, plan_end_row)
             partitioned_files, plan_start_row, plan_end_row = self._data_evolution_filter_by_shard(partitioned_files)
 
@@ -584,7 +615,7 @@ def weight_func(file_list: List[DataFileMeta]) -> int:
 
             splits += self._build_split_from_pack(flatten_packed_files, sorted_entries, False, deletion_files_map)
 
-        if self.idx_of_this_subtask is not None:
+        if self.start_row_of_this_subtask is not None or self.idx_of_this_subtask is not None:
             self._compute_split_start_end_row(splits, plan_start_row, plan_end_row)
         return splits
 

paimon-python/pypaimon/read/table_scan.py

@@ -71,3 +71,7 @@ def _create_starting_scanner(self) -> Optional[StartingScanner]:
     def with_shard(self, idx_of_this_subtask, number_of_para_subtasks) -> 'TableScan':
         self.starting_scanner.with_shard(idx_of_this_subtask, number_of_para_subtasks)
         return self
+
+    def with_row_range(self, start_row, end_row) -> 'TableScan':
+        self.starting_scanner.with_row_range(start_row, end_row)
+        return self

paimon-python/pypaimon/tests/blob_table_test.py

@@ -2373,6 +2373,68 @@ def test_blob_large_data_volume_with_shard(self):
         actual = pa.concat_tables([actual1, actual2, actual3]).sort_by('id')
         self.assertEqual(actual, expected)
 
+    def test_data_blob_writer_with_row_range(self):
+        """Test DataBlobWriter with mixed data types in blob column."""
+
+        # Create schema with blob column
+        pa_schema = pa.schema([
+            ('id', pa.int32()),
+            ('type', pa.string()),
+            ('data', pa.large_binary()),
+        ])
+
+        schema = Schema.from_pyarrow_schema(
+            pa_schema,
+            options={
+                'row-tracking.enabled': 'true',
+                'data-evolution.enabled': 'true'
+            }
+        )
+        self.catalog.create_table('test_db.with_row_range_test', schema, False)
+        table = self.catalog.get_table('test_db.with_row_range_test')
+
+        # Use proper table API to create writer
+        write_builder = table.new_batch_write_builder()
+        blob_writer = write_builder.new_write()
+
+        # Test data with different types of blob content
+        test_data = pa.Table.from_pydict({
+            'id': [0, 1, 2, 3, 4],
+            'type': ['text', 'json', 'binary', 'image', 'pdf'],
+            'data': [
+                b'This is text content',
+                b'{"key": "value", "number": 42}',
+                b'\x00\x01\x02\x03\xff\xfe\xfd',
+                b'PNG_IMAGE_DATA_PLACEHOLDER',
+                b'%PDF-1.4\nPDF_CONTENT_PLACEHOLDER'
+            ]
+        }, schema=pa_schema)
+
+        # Write mixed data
+        total_rows = 0
+        for batch in test_data.to_batches():
+            blob_writer.write_arrow_batch(batch)
+            total_rows += batch.num_rows
+
+        # Test prepare commit
+        commit_messages = blob_writer.prepare_commit()
+        # Create commit and commit the data
+        commit = write_builder.new_commit()
+        commit.commit(commit_messages)
+        blob_writer.close()
+
+        # Read data back using table API
+        read_builder = table.new_read_builder()
+        table_scan = read_builder.new_scan().with_row_range(2, 4)
+        table_read = read_builder.new_read()
+        splits = table_scan.plan().splits()
+        result = table_read.to_arrow(splits)
+
+        # Verify the data was read back correctly
+        self.assertEqual(result.num_rows, 2, "Should have 2 rows")
+        self.assertEqual(result.num_columns, 3, "Should have 3 columns")
+        self.assertEqual(result["id"].unique().to_pylist(), [2, 3], "Get incorrect column ID")
+
     def test_data_blob_writer_with_shard(self):
         """Test DataBlobWriter with mixed data types in blob column."""