Create reader for Arrow file format. Parameters ---------- source : bytes/buffer-like, pyarrow.NativeFile, or file-like Python object Either an in-memory buffer, or a readable file object. footer_offset : int, default None If the file is embedded in some larger
(source, footer_offset=None, *, options=None, memory_pool=None)
| 205 | |
| 206 | |
| 207 | def open_file(source, footer_offset=None, *, options=None, memory_pool=None): |
| 208 | """ |
| 209 | Create reader for Arrow file format. |
| 210 | |
| 211 | Parameters |
| 212 | ---------- |
| 213 | source : bytes/buffer-like, pyarrow.NativeFile, or file-like Python object |
| 214 | Either an in-memory buffer, or a readable file object. |
| 215 | footer_offset : int, default None |
| 216 | If the file is embedded in some larger file, this is the byte offset to |
| 217 | the very end of the file data. |
| 218 | options : pyarrow.ipc.IpcReadOptions |
| 219 | Options for IPC serialization. |
| 220 | If None, default values will be used. |
| 221 | memory_pool : MemoryPool, default None |
| 222 | If None, default memory pool is used. |
| 223 | |
| 224 | Returns |
| 225 | ------- |
| 226 | reader : RecordBatchFileReader |
| 227 | A reader for the given source |
| 228 | """ |
| 229 | return RecordBatchFileReader( |
| 230 | source, footer_offset=footer_offset, |
| 231 | options=options, memory_pool=memory_pool) |
| 232 | |
| 233 | |
| 234 | def serialize_pandas(df, *, nthreads=None, preserve_index=None): |
nothing calls this directly
no test coverage detected