Search Replace Image Numbers

Dev Shed Forums Sponsor:

September 1st, 2012, 02:05 PM

tom_cos

Contributing User

Join Date: Jun 2007

Location: NJ, USA

Posts: 72

Time spent in forums: 1 Day 20 h 9 m 51 sec
Reputation Power: 47

Search Replace Image Numbers

I'm nearly totally unknowledgable of Python; but here is the scenario I am trying to solve. I have a txt file with several thousand lines of text in it. About 200 of the lines say "-----File: 001.png-----". I'd like to renumber (all but the first one) by incrementing each one (002.png, 003.png, etc.).

How can I accomplish this with a Python script?

The general flow would be something along the lines of:

Open "mytextfile.txt"
While (NOT EoF):
nmbr=2
Search: "-----File: %d.png-----"
Replace: -----File: %(number)03d.png-----
nmbr = nmbr + 1
Save "mytextfile.txt"

Any help is appreciated!

Thanks, Tom C.

September 1st, 2012, 05:04 PM

b49P23TIvg

Contributing User

Join Date: Aug 2011

Posts: 3,361

Time spent in forums: 1 Month 2 Weeks 3 Days 10 h 11 sec
Reputation Power: 383

Normally I'd use gawk for this sort of program. python's not too bad though.

Code:

'''
original mytextfile.txt content:
-----File: 001.png-----
-----File: 0771.png-----
-----File: not_a_number.png-----

final mytextfile.txt content:
-----File: 001.png-----
-----File: 002.png-----
-----File: 003.png-----
'''


import os
original = 'mytextfile.txt'
backup = original+'.bak'
os.rename(original,backup)           # preserve a copy of the original
with open(backup,'r') as inf:
    with open(original,'w') as ouf:
        n = 1                 # I assume the first match is numbered 1
        for line in inf:
            fields = line.split()
            if '-----File:' == fields[0]:
                of = fields[1].split('.')
                ouf.write('%s %03d.%s\n'%(fields[0],n,of[1]))
                n += 1

__________________
[code]Code tags[/code] are essential for python code!

September 4th, 2012, 04:14 PM

tom_cos

Contributing User

Join Date: Jun 2007

Location: NJ, USA

Posts: 72

Time spent in forums: 1 Day 20 h 9 m 51 sec
Reputation Power: 47

My in file is a bit more complex. The "-----File:" lines are at the page breaks for the OCRed output. So, each one represents a page break.

Code:

------File: 001x.png---------------------------------------------

FRANKLIN.

A SKETCH

JOHN BIGELOW.

Price 25 Cents]
------File: 001.png---------------------------------------------

------File: 001.png---------------------------------------------

FRANKLIN

A SKETCH.

BY JOHN BIGELOW.

BOSTON: LITTLE, BROWN, & CO.

1879.
------File: 001.png---------------------------------------------

Entered according to Act of Congress, in the year 1878, by

LITTLE, BROWN, & COMPANY,

in the office of the Librarian of Congress at Mashington.
------File: 001.png---------------------------------------------

FRANKLIN.

September 4th, 2012, 04:25 PM

b49P23TIvg

Contributing User

Join Date: Aug 2011

Posts: 3,361

Time spent in forums: 1 Month 2 Weeks 3 Days 10 h 11 sec
Reputation Power: 383

Way to specify the actual problem.

Code:

import os
import re
original = 'mytextfile.txt'
backup = original+'.bak'
os.rename(original,backup)           # preserve a copy of the original
match = re.compile('---+File:').match  # handle your carelessness about hyphen count
with open(backup,'r') as inf:
    with open(original,'w') as ouf:
        n = 1                 # I assume the first match is numbered 1
        for line in inf:
            fields = line.split()
            if not ((2 == len(fields)) and (match(fields[0]))):
                ouf.write(line)
            else:
                of = fields[1].split('.')
                ouf.write('%s %03d.%s\n'%(fields[0],n,of[1]))
                n += 1

ps. overwriting the input file is just about always foolish, stupid, and a pain of the arse. The backup I've made helps little. Suppose you test the program twice in a row forgetting to copy the backup file to the input. You've lost your input and the whole world sux.

Last edited by b49P23TIvg : September 4th, 2012 at 04:28 PM.

September 5th, 2012, 08:42 AM

tom_cos

Contributing User

Join Date: Jun 2007

Location: NJ, USA

Posts: 72

Time spent in forums: 1 Day 20 h 9 m 51 sec
Reputation Power: 47

Woo Hoo!

Yippie! That worked like a charm!

Excellent!!

Tom C.

Search Replace Image Numbers

Developer Shed Advertisers and Affiliates