IndexError: string index out of range

[Steve Holden]

"michael" <mhyoung at valdosta.edu> wrote in message
news:mailman.987034096.12363.python-list at python.org...
> #!/usr/bin/python
>
> import sys
>
> FILENAME = "./testfile"
> EXCLUDE='#'   #Lines beginning with this should not be printed.
>
> def get_line() :
>   buf = f.readline()
>   if (buf[0] == EXCLUDE):
>     return 0
>   else:
>     return buf
>
The problem here is that you are testing for the line beginning with a "#"
before you know whether you have encountered the end of the file. That's why
you get an error at EOF.

> f = open(FILENAME, "r")
>
Also note that you set f here, but use it (as a global) int he function.
While this is possible, it's generally better practice (and more flexible)
to pass it in to your function as an argument. I wouldn;t worry too much
about that just now, though.

> while 1:
>   buf = get_line()
>   if (buf == ''):
>     break
Here you correctly test for EOF before testing for a comment line, but it's
too late, the damage was done in get_line()!

>   if (buf == 0):   # If 0 is returned then the line read on this loop
> begins with
>     continue        #    the EXCLUDE char and is not to be printed
>   else:
>     sys.stdout.write(buf)
>
The above logic might be more succinctly expressed as

if buf:            # zero tests false. None might be better?
    sys.stdout.write(buf)

> f.close()
>
> #############   END OF PROGRAM   ##############
>
> Im new to Python.This is my prgram. It reads the contents of testfile
> line by line and prints it out unless the line begins with a #. I know
> that it might be pointless to use a function for this but it is for
> learning purposes. This program works fine until it gets to the last
> line in the testfile and I get this error.
>
> Traceback (innermost last):
>   File "./test3.py", line 18, in ?
>     buf = get_line()
>   File "./test3.py", line 10, in get_line
>     if (buf[0] == EXCLUDE):
> IndexError: string index out of range
>
> Can someone please tell me what I'm doing wrong? I've looked all over
> the net and in several books for a solution.
>
You didn't go too far wrong, just a simple sequencing error. In this case,
would it actually be better to have the read routine keep going until it
either hits EOF or finds a line which needs processing? The following code
is untested, just to give you the idea...

def get_line():        # return a non-comment line or EOF
    l = f.readline()
    if not f:            # empty strings also test false
        return f
    if f[0] == "#":
        return get_line()

This recursive solution removes the comments before they have time to get
troublesome. If you don;t like recursion (a function calling itself) then
you could equivalently write:

def get_line():    # return a non-comment line or EOF
    while 1:
        l = f.readline()
        if not f:
            return f
        if f[0] != "#":
            return l

This simplifies your main program, because now it never sees the comment
lines and so only needs to worry about end of file.

Does this help?

regards
 Steve