Reading and manipulating a file in python

Question

def ints(filename):
    a = []
    f = open(filename, "r")
    lines = f.readlines()
    f.close()
    for line in lines:
        numbers = line.split()
        for number in numbers:            
            a.append(int(number))
    return a

This is my function so far, I want to be able to read a file containing integers and characters like "x" and "b" etc and return a list of only the integers. At the moment the function can only deal with a file containing integers.

How can I modify this to exclude characters or letters?

Can you give an example of the file contents you expect?

thegrinner
– thegrinner

2013-01-16 17:07:56 +00:00
Commented Jan 16, 2013 at 17:07 — thegrinner
– thegrinner, Commented Jan 16, 2013 at 17:07

NlightNFotis · Accepted Answer · 2013-01-16 17:55:37Z

4

This is my edit of your code, which performs as you expect it to.

  def ints(filename):
        """A function that takes the filename of a file as an input arguement, computs and returns a list of
        integers of all the numbers in the file."""
        a = []
        f = open(filename, "r")
        lines = f.readlines()
        f.close()
        for line in lines:
            for character in line:
                try:
                    a.append(int(character))
                except ValueError:
                    pass
        return a

edited Jan 16, 2013 at 17:55

answered Jan 16, 2013 at 17:10

NlightNFotis

9,8716 gold badges46 silver badges67 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

user1632861 Over a year ago

Weird that this got accepted, only working for numbers surrounded by spaces: abc123cba will ignore 123

NlightNFotis Over a year ago

@Mahi Oh thanx for the notice, however, this is something so easy to fix.

user1632861 Over a year ago

Still could be improved, for "1 19 532 2" it returns [1, 1, 9, 5, 3, 2, 2] instead of [1, 19, 532, 2] as does Chris Harris' answer.

NlightNFotis Over a year ago

@Mahi I'm sorry, but the question leave's me under the impression that the op wants single digit numbers. If he requests otherwise, then I shall change it.

Ashwini Chaudhary · Accepted Answer · 2013-01-16 17:24:39Z

2

regex can be helpful here:

A simple example:

In [22]: import re

In [23]: strs="121 some 34useless text 56"

In [24]: map(int,re.findall("\d+",strs))
Out[24]: [121, 34, 56]

# or this If you want the individual digits:

In [40]: map(int,re.findall("\d",strs))
Out[40]: [1, 2, 1, 3, 4, 5, 6]

For your code this should work:

for line in lines:
    numbers = map(int,re.findall("\d+",line))
    a.extend(numbers)

edited Jan 16, 2013 at 17:24

answered Jan 16, 2013 at 17:09

Ashwini Chaudhary

252k60 gold badges478 silver badges519 bronze badges

2 Comments

NlightNFotis Over a year ago

+1 nice idea. Although I firmly believe that regex are hard at this point for the op, since it's obvious he has recently got started with python.

user1632861 Over a year ago

you mean line instead of strs on the second example? :)

Ci3 · Accepted Answer · 2013-01-16 17:25:32Z

1

I just simply tested if characters were numbers:

sample_string = "Test4. 2325This string3"
a_list = []
for x in sample_string:
    if x.isdigit():
        a_list.append(x)

edited Jan 16, 2013 at 17:25

answered Jan 16, 2013 at 17:17

Ci3

4,87211 gold badges36 silver badges49 bronze badges

2 Comments

Ashwini Chaudhary Over a year ago

1) It returns ['4', '2', '3', '2', '5', '3'] instead of [4, 2325, 3]. 2) By a.append(x) I guess you actually meant a_list.append(x)?

Ci3 Over a year ago

@AshwiniChaudhary Yes. I was sloppy.

XORcist · Accepted Answer · 2013-01-16 17:10:49Z

0

for number in numbers:
    try:
        a.append(int(number))
    except ValueError:
        pass

answered Jan 16, 2013 at 17:10

XORcist

4,36726 silver badges32 bronze badges

Comments

Keith · Accepted Answer · 2013-01-16 17:18:11Z

0

Using more modern Python idioms:

def ints(filename):
    with open(filename, "r") as f:
        for line in f:
            for number in line.split():
                try:
                    yield int(number)
                except ValueError:
                    pass


a = list(ints("testdata.txt"))
print(a)

Basically, try to convert to int, which will raise ValueError if the string is not a decimal number. Catch it and ignore it and carry on.

answered Jan 16, 2013 at 17:18

Keith

43.2k11 gold badges62 silver badges77 bronze badges

Comments

BenDundee · Accepted Answer · 2013-01-16 17:20:21Z

0

Try/catch might help:

for thing in line.split():
    i_thing = None
    try:
        i_thing = int(thing)
    except ValueError:
        pass

    s_thing = None
    try:
        s_thing = str(thing)
    except:
        raise Exception("OH NOES!")

It's pretty ugly, but I haven't found a nicer way to do what you're trying to do.

edited Jan 16, 2013 at 17:20

answered Jan 16, 2013 at 17:13

BenDundee

4,5613 gold badges32 silver badges35 bronze badges

Comments

sotapme · Accepted Answer · 2013-01-16 17:43:19Z

How about using string.translate etc. to just replaces all non digits with spaces then leverage the power of split() and map().

Granted it's a little obscure and my default response would be to just use the re module as on the whole there's so much more you can do with Regular Expressions, that they're worth the effort in learning.

In [119]: import string
In [120]: allchars = string.maketrans('', '')
In [121]: delchars = allchars.translate(allchars, "0123456789")
In [122]: emptychars = string.maketrans(delchars, ' ' * len(delchars))
In [123]: "Welcome home 1234 56 ol".translate(emptychars)
Out[123]: '             1234 56   '
In [124]: "Welcome home 1234 56 ol".translate(emptychars).split()
Out[124]: ['1234', '56']
In [125]: map(int, "Welcome home 1234 56 ol".translate(emptychars).split())
Out[125]: [1234, 56]

Dan · Accepted Answer · 2013-01-16 18:07:12Z

This is untested sudo-code, but should be valid. Also the best advice I can give is watch David Beazley's Generator Tricks for Systems programmers and Mastering Python 3 I/O talks. They were a great help in learning python.

This is just a simple generator to fetch each line of a file and close the file afterwards.

def getLine(fileName):
    file = open(fileName, "r")

    for line in file.readLines():
        yield line

    file.close()


def getNumbers(line):
// I'm lazy and stole this one from Keith
    for number in line.split():
        try:
            yield int(number)
        except ValueError:
            pass

def generatorChain(fileName):
    // I'm a little iffy on the syntax here, but shouldn't be to hard with a bit of googling
    fileGen = getLine(fileName);
    yield getNumber( fileGen.next() ).next()

def listCompressionMagic():
    return [x for x in generatorChain("foo.txt") ]

Collectives™ on Stack Overflow

Reading and manipulating a file in python

8 Answers 8

4 Comments

2 Comments

2 Comments

Comments

Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

8 Answers 8

4 Comments

2 Comments

2 Comments

Comments

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related