normalizer.py 文件源码

def read_excel(self, file):
        # TODO: add iterator and return columns
        excel_tab = pd.read_excel(file, dtype=str)
        columns = excel_tab.columns

        def make_gen(excel_tab, chunksize):
            cursor = 0
            chunk = excel_tab.iloc[:chunksize]
            while chunk.shape[0]:
                yield chunk
                cursor += chunksize
                chunk = excel_tab.iloc[cursor:cursor+chunksize]
        tab = make_gen(excel_tab, self.CHUNKSIZE) 

        tab = (self._clean_header(tab_part) for tab_part in tab)

        return tab, None, None, self._clean_column_names(columns)