Python：如何在reader中保存一行并保留其上下文？

Question

使用Python，我编写了以下脚本来索引csv.reader文件并将其与前一行中的相关值进行比较：

source = open('sourcefile')
reader = csv.reader(source)
with open('targetfile','w') as f:
    for row in reader:
        if row[1] == "NA" and row[2] == "NA":
            prevrow = row
            line = row
            writer = csv.writer(f)
            writer.writerow(line)
     else:
         while row[1] == prevrow[1]:
             if row[1] == prevrow[3]:
                cumbid = cumbid + row[2]
                line = row[0]  + "," + row[1]  + "," + cumbid  + "," + prevrow[3]  + "," + prevrow[4] + "," +  str(int(float(prevrow[5]))-int(float(cumbid)))  + "," + prevrow[6]  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + ","
                prevrow = line
            elif row[1] == prevrow[4]:
                cumask = cumask + row[2]
                line = row[0]  + "," + row[1]  + "," + cumask  + "," + prevrow[3]  + "," + prevrow[4] + "," +  prevrow[5]  + "," + str(int(float(prevrow[6]))-int(float(cumask)))  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + "," 
                prevrow = line
        else:
            if row[1] == prevrow[3] and row[2] < prevrow[5]:                       
                cumbid = row[2]
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[3]  + "," + prevrow[4] + "," +  str(int(float(prevrow[5]))-int(float(row[2])))  + "," + prevrow[6]  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + "," 

                prevrow = line
                print prevrow
                print prevrow[0]+ " " + prevrow[1]+ " " + prevrow[2]+ " " + prevrow[3]+ " " + prevrow[4]+ " " + prevrow[5]+ " " + prevrow[6]
                writer = csv.writer(f)
                writer.writerow(line.split())   
            elif row[1] == prevrow[4] and row[2] < prevrow[6]:                  
                cumask = row[2]
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[3]  + "," + prevrow[4] + "," +  prevrow[5]  + "," + str(int(float(prevrow[6]))-int(float(row[2])))  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + "," 
                prevrow = line
                print prevrow
                print prevrow[0]+ " " + prevrow[1]+ " " + prevrow[2]+ " " + prevrow[3]+ " " + prevrow[4]+ " " + prevrow[5]+ " " + prevrow[6]
                writer = csv.writer(f)
                writer.writerow(line.split())   
            elif row[1] == prevrow[3] and row[2] == prevrow[5] :                 
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[7]  + "," + prevrow[4] + "," +  prevrow[8] + "," + prevrow[6]  + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[9] + "," + prevrow[10] + "," + "," + "," + "," + "," + "," + "," + "," 
                prevrow = line
                print prevrow
                print prevrow[0]+ " " + prevrow[1]+ " " + prevrow[2]+ " " + prevrow[3]+ " " + prevrow[4]+ " " + prevrow[5]+ " " + prevrow[6]
                writer = csv.writer(f)
                writer.writerow(line.split())   
            elif row[1] == prevrow[4] and row[2] == prevrow[6]:                   #the trade was on the offer and took it all out--we have to refer to the 2nd level of the book
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[3]  + "," + prevrow[9] + "," +  prevrow[5]  + "," + prevrow[10]  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[13] + "," + prevrow[14] + "," + "," + "," + "," + "," + "," + "," + "," 
                prevrow = line
                print prevrow
                print prevrow[0]+ " " + prevrow[1]+ " " + prevrow[2]+ " " + prevrow[3]+ " " + prevrow[4]+ " " + prevrow[5]+ " " + prevrow[6]
                writer = csv.writer(f)
                writer.writerow(line.split())   
            elif row[1] == prevrow[1] and row[1] == prevrow[3] :                  #trade was on the bid, and the prev row was also a trade--decrement bid size
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[3]  + "," + prevrow[4] + "," +  str(int(float(prevrow[5]))-int(float(row[2])))  + "," + prevrow[6]  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + "," 
                prevrow = line
                writer = csv.writer(f)
                writer.writerow(line.split())   
            elif row[1] == prevrow[1] and row[1] == prevrow[4] :                    #trade was on the offer, and the prev row was also a trade--decrement ask size
                line = row[0]  + "," + row[1]  + "," + row[2]  + "," + prevrow[3]  + "," + prevrow[4] + "," +  prevrow[5]  + "," + str(int(float(prevrow[6]))-int(float(row[2])))  + "," + prevrow[7] + "," + prevrow[8] + "," + prevrow[9] + "," + prevrow[10] + "," + prevrow[11] + "," + prevrow[12] + "," + prevrow[13] + "," + prevrow[14] + "," + "," 
                prevrow = line
                writer = csv.writer(f)
                writer.writerow(line.split())   

但是，这些后续行生成以下内容：

代码

print prevrow
print prevrow[0]+ " " + prevrow[1]+ " " + prevrow[2]+ " " + prevrow[3]+ " " + prevrow[4]+ " " + prevrow[5]+ " " + prevrow[6]

输出

20100628000120702,12392.0,1.0,12392.0,12393.0,5.0
2 0 1 0 0 6 2

我希望输出的两行是相同的，这样像row[1]==prevrow[1]这样的比较将把12392.0与前一行（12392.0）中的第二个项目进行比较，而不是将行（0）中的第二个字符进行比较。所以，我的问题是prevrow变成了一个字符列表，当我试图用索引行中的项目的方式索引prevrow中的项目时，我似乎得到了列表中字符的索引。你知道吗

如何保存当前“行”并保留其上下文？你知道吗

Answer 1

使用tee创建两个虚拟迭代器，递增一个，然后压缩它们，以获得（lastrow，currentrow）元组序列。（参见itertools documentation中的pairwise配方。）

然后你可以写：

for prevrow,row in pairwise(reader):
    ... etc ...

Answer 2

您正在line中构造字符串并将其存储在prevrow；请存储一个列表：

line = row[:2] + [cumbid] + prevrow[3:5] + [str(int(float(prevrow[5])) - int(float(cumbid)))] + prevrow[6:15]