遍历建议直接用pandas的itertuples()，去除前面的空白字符串用lstrip()就行
df = pd.read_excel("test.xlsx")
for row in df.itertuples():
row.行名称=row.行名称.lstrip()
以上应该就可以了，注意缩进。
按 指定行数 分页读取好处理：
def iterpage(istream, pagesize):buffer = []
for data in istream:
buffer.append(data)
if len(buffer)>=pagesize:
yield buffer
buffer = []
if buffer:
yield buffer
with open("source.txt", 'rt') as handle:
for page in iterpage(handle, 1000):
print page # or your business logical
print "-"*32 # page break
删除文本文件的前N行：
def removehead(filename, headlines):buffer = []
with open(filename, 'rt') as handle:
for i, ln in enumerate(handle):
if ln < headlines:
continue
buffer.append(ln)
with open(filename, 'wt') as handle:
handle.writelines(buffer)
或者：
def getandremovehead(filename, headlines):with open(filename, 'rt') as handle:
buffer = handle.readlines()
with open(filename, 'wt') as handle:
handle.writelines(buffer[headlines:])
return buffer[:headlines]
但遇到大文本文件时，删除其中N行不是很理想的业务方案
在遍历数据结构的时候，是不可以修改原数据结构的。不然就会抛出错误。我常用的解决办法是做一份拷贝，遍历这个拷贝。（如果数据不是很大的话）
比如，这个代码：
C#代码
1.<SPAN style="FONT-SIZE: x-small">IDictionary<int, string>ht = new Dictionary<int, string>()
2.ht.Add(1, "one")
3.ht.Add(2, "two")
4.
5.// Print "one,two"
6.Console.WriteLine(String.Join(",", ht.Values.Select(i =>i.ToString()).ToArray()))
7.
8.foreach (int key in new List<int>(ht.Keys)) {
9.if (key == 1) ht.Remove(key)
10.}
11.
12.// Print "two"
13.Console.WriteLine(String.Join(",", ht.Values.Select(i =>i.ToString()).ToArray()))</SPAN>
IDictionary<int, string>ht = new Dictionary<int, string>()
ht.Add(1, "one")
ht.Add(2, "two")
// Print "one,two"
Console.WriteLine(String.Join(",", ht.Values.Select(i =>i.ToString()).ToArray()))
foreach (int key in new List<int>(ht.Keys)) {
if (key == 1) ht.Remove(key)
}
// Print "two"
Console.WriteLine(String.Join(",", ht.Values.Select(i =>i.ToString()).ToArray()))我在遍历的时候，做了一份拷贝。代码是 new List<int>(ht.Keys)，用到了 List 的构造拷贝函数，
