How to unread a line in python

Question

I am new to Python (2.6), and have a situation where I need to un-read a line I just read from a file. Here's basically what I am doing.

  for line in file:
     print line
     file.seek(-len(line),1)
     zz = file.readline()
     print zz

However I notice that "zz" and "line" are not the same. Where am I going wrong?

Thanks.

@quamrana : It's kind of hard to explain. That's how the code that I am manipulating is written :) — user721975, May 01 '11 at 18:05
Here's an example (my current use): You're reading a block of data from a file, and there's no explicit end-of-block marker, but you can recognize when you've hit the next block. In that case, you don't want the processing of block i to consume the beginning of block i+1, so "un-reading" makes a lot of sense — Eric Anderson, Aug 14 '12 at 15:43

score 12 · Accepted Answer · answered May 01 '11 at 17:45

12

I don't think for line in file: and seek make a good combination. Try something like this:

while True:
    line = file.readline()
    print line
    file.seek(-len(line),1)
    zz = file.readline()
    print zz

    # Make sure this loop ends somehow

answered May 01 '11 at 17:45

Gustav Larsson

8,199
3
31
51

I just found it out the hard way. I am coming from a Perl background and am still making adjustments to Python. Thanks for your answer, that worked. – user721975 May 01 '11 at 18:01
3

`.readline()` returns an empty string `''` on EOF, so the exit condition is `if not line: break`. – jfs May 01 '11 at 18:54

score 3 · Answer 2 · answered May 01 '11 at 17:45

3

You simply cannot mix iterators and seek() this way. You should pick one method and stick to it.

answered May 01 '11 at 17:45

David Heffernan

601,492
42
1,072
1,490

score 2 · Answer 3 · answered May 01 '11 at 18:47

You can combine the iteration over lines with a .seek() operation:

for i, line in enumerate(iter(f.readline, ''), 1):
    print i, line,
    if i == 2: # read 2nd line two times
       f.seek(-len(line), os.SEEK_CUR)

If a file contains:

a
b
c

Then the output would be:

1 a
2 b
3 b
4 c

score 2 · Answer 4 · answered May 01 '11 at 21:09

Untested. Basically, you want to maintain a lifo cache of 'unread' lines. On each read of a line, if there is something in the cache, you take it out of the cache first. If there's nothing in the cache, read a new line from the file. This is rough, but should get you going.

lineCache = []

def pushLine(line):
    lineCache.append(line)

def nextLine(f):
    while True:
        if lineCache:
            yield lineCache.pop(0)
        line = f.readline()
        if not line:
            break
        yield line
    return

f = open('myfile')

for line in nextLine(f):
    # if we need to 'unread' the line, call pushLine on it.  The next call to nextLine will
    # return the that same 'unread' line.
    if some_condition_that_warrants_unreading_a_line:
        pushLine(line)
        continue
    # handle line that was read.

http://code.activestate.com/recipes/502304-iterator-wrapper-allowing-pushback-and-nonzero-tes/ — tzot, May 02 '11 at 00:06

How to unread a line in python

4 Answers4

Linked