I want to create a stream from a string in Python so that it's equivalent to reading the string as if it's read from a text file. something like:

for line in open('myfile.txt'): print line

except the contents of 'myfile.txt' are stored in a string s . Is this the correct/best way to do it?

s = StringIO.StringIO("atbnctdn")
for line in s: print line

---

I want to create a stream from a string in Python so that it's equivalent to reading the string as if it's read from a text file.

Is this the correct/best way to do it?

Yes, unless you really do want it in a list.

If it is intended to be consumed line by line, the way you are doing it makes sense.

StringIO() creates a file-like object.

File objects have a method, .readlines() , which materialize the object as a list. Instead of materializing the data in a list, you can iterate over it, which is more memory light:

# from StringIO import StringIO # Python 2 import
from io import StringIO # Python 3 import

txt = "foonbarnbaz"

Here we append each line into a list, so that we can demonstrate iterating over the file-like object and keeping a handle on the data. (More efficient would be list(file_like_io) .

m_1 = []
file_like_io = StringIO(txt)
for line in file_like_io:
    m_1.append(line)

and now:

>>> m_1
['foon', 'barn', 'baz']

you can return your io to any index point with seek :

>>> file_like_io.seek(0)
>>> file_like_io.tell() #print where we are in the object now
0

If you really want it in a list

.readlines() materializes the StringIO iterator as if one did list(io) - this is considered less preferable.

>>> m_2 = file_like_io.readlines() 

And we can see that our results are the same:

>>> m_1==m_2
True

Keep in mind that it is splitting after the newlines, preserving them in the text as well, so you'll get two newlines for every printed line, double-spacing on print.

---

You could roll your own with a simple generator function like this:

def string_stream(s, separators="n"):
    start = 0
    for end in range(len(s)):
        if s[end] in separators:
            yield s[start:end]
            start = end + 1
    if start < end:
        yield s[start:end+1]

Example usage:

>>> stream = string_stream("footbarnbazn", "tn")
>>> for s in stream:
...     print(s)
...
foo
bar
baz

cStringIO may be faster (I haven't tested), but this would give you flexibility in defining/consuming separators.

链接地址: http://www.djcxy.com/p/19166.html

上一篇: 从对象到字符串的熊猫dtype转换

下一篇: 创建流来从Python中的字符串迭代