diff of excel file in Git repository index and working copy

excel_diff.py

#! /usr/bin/env python

"""
diff of excel files
"""

from dulwich.repo import Repo  # import Repo object from dulwich
import sys
import os
from openpyxl import load_workbook
from StringIO import StringIO


def diff_excel(filename, sheets, repo='.'):
    """
    diff excel file
    """
    try:
        wb_new = load_workbook(filename)
    except:
        raise IOError('"%s" does not exist' % filename)
    r = Repo(repo)  # open repository, raises not a repository
    idx = r.open_index()  # open index to get excel file's sha
    f = idx[filename]  # get excel file info from index
    b = r[f.sha]  # get blob of excel file
    s = StringIO(b)  # file like object of excel file's blob data
    # StringIO(b) actually returns StringIO(b.data)
    wb_old = load_workbook(s)
    for sh in sheets:
        print 'sheet: %s' % sh
        try:
            ws_old = wb_old.get_sheet_by_name(sh)
        except:
            print 'old workbook does not have sheet: "%s"' % sh
        try:
            ws_new = wb_new.get_sheet_by_name(sh)
        except:
            print 'new workbook does not have sheet: "%s"' % sh
        n = 0
        for rold, rnew in zip(ws_old.rows, ws_new.rows):
            rvlist = [rv.value for rv in rold]
            fmt = ('old row: %5d' % (n + 1)) + ''.join(['\t%s'] * len(rvlist))
            print fmt % tuple(rvlist)
            rvlist = [rv.value for rv in rnew]
            fmt = ('new row: %5d' % (n + 1)) + ''.join(['\t%s'] * len(rvlist))
            print fmt % tuple(rvlist)
            n += 1
        if n < len(ws_old.rows):
            print 'old file has %d extra rows' % (len(ws_old.rows) - n)
        elif n < len(ws_new.rows):
            print 'new file has %d extra rows' % (len(ws_new.rows) - n)

if __name__ == '__main__':
    diff_excel(filename=sys.argv[1], sheets=sys.argv[2:])