| 222 | |
| 223 | |
| 224 | class ORCWriter: |
| 225 | __doc__ = f""" |
| 226 | Writer interface for a single ORC file |
| 227 | |
| 228 | Parameters |
| 229 | ---------- |
| 230 | where : str or pyarrow.io.NativeFile |
| 231 | Writable target. For passing Python file objects or byte buffers, |
| 232 | see pyarrow.io.PythonFileInterface, pyarrow.io.BufferOutputStream |
| 233 | or pyarrow.io.FixedSizeBufferWriter. |
| 234 | {_orc_writer_args_docs} |
| 235 | """ |
| 236 | |
| 237 | is_open = False |
| 238 | |
| 239 | def __init__(self, where, *, |
| 240 | file_version='0.12', |
| 241 | batch_size=1024, |
| 242 | stripe_size=64 * 1024 * 1024, |
| 243 | compression='uncompressed', |
| 244 | compression_block_size=65536, |
| 245 | compression_strategy='speed', |
| 246 | row_index_stride=10000, |
| 247 | padding_tolerance=0.0, |
| 248 | dictionary_key_size_threshold=0.0, |
| 249 | bloom_filter_columns=None, |
| 250 | bloom_filter_fpp=0.05, |
| 251 | ): |
| 252 | self.writer = _orc.ORCWriter() |
| 253 | self.writer.open( |
| 254 | where, |
| 255 | file_version=file_version, |
| 256 | batch_size=batch_size, |
| 257 | stripe_size=stripe_size, |
| 258 | compression=compression, |
| 259 | compression_block_size=compression_block_size, |
| 260 | compression_strategy=compression_strategy, |
| 261 | row_index_stride=row_index_stride, |
| 262 | padding_tolerance=padding_tolerance, |
| 263 | dictionary_key_size_threshold=dictionary_key_size_threshold, |
| 264 | bloom_filter_columns=bloom_filter_columns, |
| 265 | bloom_filter_fpp=bloom_filter_fpp |
| 266 | ) |
| 267 | self.is_open = True |
| 268 | |
| 269 | def __del__(self): |
| 270 | self.close() |
| 271 | |
| 272 | def __enter__(self): |
| 273 | return self |
| 274 | |
| 275 | def __exit__(self, *args, **kwargs): |
| 276 | self.close() |
| 277 | |
| 278 | def write(self, table): |
| 279 | """ |
| 280 | Write the table into an ORC file. The schema of the table must |
| 281 | be equal to the schema used when opening the ORC file. |