我对Python很新颖,并且有想法如何优化这个程序以避免内存错误。优化数据转换程序以避免内存错误
我想从两个工作簿中读取数据:raw_data和映射。我想使用映射文档将raw_data转换为新的电子表格来转换数据。所以我加载工作簿,从映射数据中创建数据字典并开始转换。但是,我遇到了内存错误。
有什么办法来优化下面的代码来避免这个错误?
import openpyxl
from openpyxl.utils import get_column_letter
mapping = openpyxl.load_workbook(r'C:...\mapping.xlsx') #load mapping doc
wb = openpyxl.load_workbook(r'C:...\raw_data.xlsx') #load raw data
sheet = wb.active #look at the active sheet in the raw data file
user_map_raw = mapping.get_sheet_by_name('User ID Mapping') #for user ids
item_map_raw = mapping.get_sheet_by_name('Item ID Mapping') #for item ids
...other mappings here
def load(sheet):
user_dict = {}
print "creating user dictionary..."
for row in range(1, user_map_raw.max_row+1):
old_name = user_map_raw['A' + str(row)].value #old user name
new_name = user_map_raw['B' + str(row)].value #new user name
user_dict[old_name] = new_name #old name is key for the new name
item_dict = {}
print "creating item id dictionary..."
for row in range(1, item_map_raw.max_row+1):
old_item = item_map_raw['A' + str(row)].value #old item id
new_item = item_map_raw['B' + str(row)].value #new item id
item_dict[old_item] = new_item #old item id is key for new item id
raw = [] #empty list to store data before writing to new file
for row in range(2, sheet.max_row+1): #loop thru raw data and map
print "loading row %s" % row
user_ID = user_dict[sheet['A' + str(row)].value]
item_type = sheet['B' + str(row)].value
item_ID = item_dict[sheet['C' + str(row)].value]
...other transformations here
add = [user_ID, item_type, item_ID, ...]
raw.append(add) #add transformed data to list
new = openpyxl.Workbook() #create new workbook
output = new.active #select the active sheet
for i in range(len(raw)): #loop through transformed data list
"print writing row %s" %i
for j in range(len(raw[i])): #write to new sheet
output[get_column_letter(j+1) + str(i+1)] = raw[i][j]
new.save('new_doc.xlsx')
load(sheet)
嘿谢谢 - 这个作品。我唯一的问题:你为什么使用row_offset = 2? – kbball
在问题示例中,您将通过主表跳过几行并从第2行开始循环:'range(2,sheet.max_row + 1):'。希望'row_offset'做同样的事情。很难从openpyxl文档中知道。您可能需要验证您是否获得了所需的所有行。 – systemjack
对,我认为它应该是row_offset = 1。但无论如何现在工作。感谢你在这方面的所有努力。关键要点:将文件读取为只读并将其写入为只写。使用iter_rows提取单元格值 – kbball