| 1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768 |
- from typing import List
- from pandas._typing import FilePathOrBuffer, Scalar
- from pandas.compat._optional import import_optional_dependency
- from pandas.io.excel._base import _BaseExcelReader
- class _PyxlsbReader(_BaseExcelReader):
- def __init__(self, filepath_or_buffer: FilePathOrBuffer):
- """Reader using pyxlsb engine.
- Parameters
- __________
- filepath_or_buffer: string, path object, or Workbook
- Object to be parsed.
- """
- import_optional_dependency("pyxlsb")
- # This will call load_workbook on the filepath or buffer
- # And set the result to the book-attribute
- super().__init__(filepath_or_buffer)
- @property
- def _workbook_class(self):
- from pyxlsb import Workbook
- return Workbook
- def load_workbook(self, filepath_or_buffer: FilePathOrBuffer):
- from pyxlsb import open_workbook
- # Todo: hack in buffer capability
- # This might need some modifications to the Pyxlsb library
- # Actual work for opening it is in xlsbpackage.py, line 20-ish
- return open_workbook(filepath_or_buffer)
- @property
- def sheet_names(self) -> List[str]:
- return self.book.sheets
- def get_sheet_by_name(self, name: str):
- return self.book.get_sheet(name)
- def get_sheet_by_index(self, index: int):
- # pyxlsb sheets are indexed from 1 onwards
- # There's a fix for this in the source, but the pypi package doesn't have it
- return self.book.get_sheet(index + 1)
- def _convert_cell(self, cell, convert_float: bool) -> Scalar:
- # Todo: there is no way to distinguish between floats and datetimes in pyxlsb
- # This means that there is no way to read datetime types from an xlsb file yet
- if cell.v is None:
- return "" # Prevents non-named columns from not showing up as Unnamed: i
- if isinstance(cell.v, float) and convert_float:
- val = int(cell.v)
- if val == cell.v:
- return val
- else:
- return float(cell.v)
- return cell.v
- def get_sheet_data(self, sheet, convert_float: bool) -> List[List[Scalar]]:
- return [
- [self._convert_cell(c, convert_float) for c in r]
- for r in sheet.rows(sparse=False)
- ]
|
