The Problem Creation of fancy reports using Pandas and Python.
Proposed Solution Using a template xlsx file containing a template sheet nicely formatted with references to another pre-populated worksheet, delete the pre-populated sheet and insert the new worksheet from pandas. The template sheet will lose the links reverting to #REF so these will need to be renamed.
I tried:
import os
import xlrd, xlwt
import envconfig
swb1 = xlrd.open_workbook(os.path.join(envconfig.REPORT_WRITER_PATH,'TEMPLATE.xls'), on_demand=True, formatting_info=True)
swb2 = xlrd.open_workbook(os.path.join(envconfig.REPORT_WRITER_PATH,'REPORT.xls'), on_demand=True, formatting_info=True)
swb1s1 = swb1.sheet_by_name('Template')
swb2s1 = swb2.sheet_by_name('Report')
twb = xlwt.Workbook()
sheet1 = twb.add_sheet(swb1s1)
sheet2 = twb.add_sheet(swb2s1)
twb.save("python_spreadsheet.xls")
The above errors with:
sheet1 = twb.add_sheet(swb1s1)
File "C:\Users\pa003202\AppData\Local\Continuum\Anaconda3\lib\site-packages\xlwt\Workbook.py", line 366, in add_sheet
sheetname = sheetname.decode(self.encoding)
AttributeError: 'Sheet' object has no attribute 'decode'
sheetname = sheetname.decode(self.encoding)
AttributeError: 'Sheet' object has no attribute 'decode'
Is there a way to inject data from pandas into a workbook or to open a workbook and insert a sheet?