How to edit a file in python 2.7.10?

Question

I am trying to edit a file as follows in python 2.7.10 and running into below error, can anyone provide guidance on what the issue is on how to edit files?

import fileinput,re
filename = 'epivers.h'
text_to_search = re.compile("#define EPI_VERSION_STR         \"(\d+\.\d+) (TOB) (r(\d+) ASSRT)\"")
replacement_text = "#define EPI_VERSION_STR         \"9.130.27.50.1.2.3 (r749679 ASSRT)\""
with fileinput.FileInput(filename, inplace=True, backup='.bak') as file:
    for line in file:
        print(line.replace(text_to_search, replacement_text))
file.close()

Error:-

Traceback (most recent call last):
  File "pythonfiledit.py", line 5, in <module>
    with fileinput.FileInput(filename, inplace=True, backup='.bak') as file:
AttributeError: FileInput instance has no attribute '__exit__'

UPDATE:

import fileinput,re
import os
import shutil
import sys
import tempfile
filename = 'epivers.h'
text_to_search = re.compile("#define EPI_VERSION_STR         \"(\d+\.\d+) (TOB) (r(\d+) ASSRT)\"")
replacement_text = "#define EPI_VERSION_STR         \"9.130.27.50.1.2.3 (r749679 ASSRT)\""

with open(filename) as src, tempfile.NamedTemporaryFile(
        'w', dir=os.path.dirname(filename), delete=False) as dst:

    # Discard first line
    for line in src:
        if text_to_search.search(line):
            # Save the new first line
            line = text_to_search .sub(replacement_text,line)
            dst.write(line + '\n')
        dst.write(line)

# remove old version
os.unlink(filename)

# rename new version
os.rename(dst.name,filename)

I am trying to match line define EPI_VERSION_STR "9.130.27 (TOB) (r749679 ASSRT)"

Answer 1

If r is a compiled regular expression and line is a line of text, the way to apply the regex is

r.match(line)

to find a match at the beginning of line, or

r.search(line)

to find a match anywhere. In your particular case, you simply need

line = r.sub(replacement, line)

though in addition, you'll need to add a backslash before the round parentheses in your regex in order to match them literally (except in a few places where you apparently put in grouping parentheses around the \d+ for no particular reason; maybe just take those out).

Your example input string contains three digits, and the replacement string contains six digits, so \d+\.\d+ will never match either of those. I'm guessing you want something like \d+(?:\.\d+)+ or perhaps very bluntly [\d.]+ if the periods can be adjacent.

Furthermore, a single backslash in a string will be interpreted by Python, before it gets passed to the regex engine. You'll want to use raw strings around regexes, nearly always. For improved legibility, perhaps also prefer single quotes or triple double quotes over regular double quotes, so you don't have to backslash the double quotes within the regex.

Finally, your usage of fileinput is wrong. You can't use it as a context manager. Just loop over the lines which fileinput.input() returns.

import fileinput, re

filename = 'epivers.h'
text_to_search = re.compile(r'#define EPI_VERSION_STR         "\d+(?:\.\d+)+ \(TOB\) \(r\d+ ASSRT\)"')
replacement_text = '#define EPI_VERSION_STR         "9.130.27.50.1.2.3 (r749679 ASSRT)"'
for line in fileinput.input(filename, inplace=True, backup='.bak'):
    print(text_to_search.sub(replacement_text, line))

In your first attempt, line.replace() was a good start, but it doesn't accept a regex argument (and of course, you don't close() a file you opened with with ...). In your second attempt, you are checking whether the line is identical to the regex, which of course it isn't (just like the string "two" isn't equivalent to the numeric constant 2).

Answer 2

Read the file, use re.sub to substitute, then write the new contents back:

with open(filename) as f:
    text = f.read()

new_text = re.sub(r'#define EPI_VERSION_STR         "\d+\(?:.\d+\)+ \(TOB\) \(r\d+ ASSRT\)"',
                  '#define EPI_VERSION_STR         "9.130.27.50.1.2.3 (r749679 ASSRT)"',
                  text)

with open(filename, 'w') as f:
    f.write(new_text)