[SOLVED] vimdiff problem comparing two files

101b147 · 10-30-2009, 08:44 PM

hello guys. im kind of embarrassed for asking this, but its driving me nuts.
im trying to compare two files with vimdiff, and vimdiff places a lot of hiffens ( i think it means that its because these lines doesnt appear on the other file ) but the thing is : the files are too simple to have this problem.

im not sure if i made myself clear, so ill post my results:

Quote:

leecher@darkstar:~/prog/uva/103$ cat out
5
10 11 2 1 3
4
5 3 4 2

Quote:

leecher@darkstar:~/prog/uva/103$ cat saida_uva
6
10 9 4 13 17 3
5
6 8 3 4 2

when i use vimdiff (with -o), it shows me this:

--------------------(the hiffens i talked about)
--------------------
5
10 11 2 1 3
4
5 3 4 2

6
10 9 4 13 17 3
5
6 8 3 4 2
-----------------(damn hiffens here again)
------------------

sorry for such a question, and yes, i know, i could just compare them manually, but ill need vimdiff for larger files. these are just some tests, and they are causing me trouble. thx

in case someone is wondering, here is the output of diff:

Quote:

leecher@darkstar:~/prog/uva/103$ diff out saida_uva
0a1,2
> 6
> 10 9 4 13 17 3
2,4c4
< 10 11 2 1 3
< 4
< 5 3 4 2
---
> 6 8 3 4 2

aus9 · 10-30-2009, 10:06 PM

hi

I use xxdiff.....command is xxdiff file1 file2

its a lot easier to see the differences and so I hope gives you a better answer

Vrajgh · 10-31-2009, 05:22 AM

diff is mostly a line oriented comparison. In this case it sees the line containing only "5" and tries to line them up in both files. Having lined those up, the hyphens represent non-existent lines before one file and after the other.

It doesn't simply compare line 1 with line 1 etc because it needs to be able to detect where lines have been added or deleted.

diff produces:

Code:

$ diff -y out.txt out1.txt 
		   >	6
		   >	10 9 4 13 17 3
5			5
10 11 2 1 3	   |	6 8 3 4 2 
4		   <
5 3 4 2 	   <

<

The first and second line can be interpreted as existing only in the second file. The third line is the same in both. Fourth is a modified line. Last two lines exist only in the first file.

Basically, these files are too simple for diff to work out what to do with them.

I have never used xxdiff so can't say if it would produce better output.

aus9 · 10-31-2009, 08:54 AM

http://imagebin.org/69888

I deliberately resized xxdiff smallish so it fitted under command but IMHO vision is better no need to explain it

image disappears from server in 7 days

101b147 · 10-31-2009, 09:49 AM

Quote:

Originally Posted by Vrajgh

diff is mostly a line oriented comparison. In this case it sees the line containing only "5" and tries to line them up in both files. Having lined those up, the hyphens represent non-existent lines before one file and after the other.

It doesn't simply compare line 1 with line 1 etc because it needs to be able to detect where lines have been added or deleted.

diff produces:

Code:

$ diff -y out.txt out1.txt 
		   >	6
		   >	10 9 4 13 17 3
5			5
10 11 2 1 3	   |	6 8 3 4 2 
4		   <
5 3 4 2 	   <

<

The first and second line can be interpreted as existing only in the second file. The third line is the same in both. Fourth is a modified line. Last two lines exist only in the first file.

Basically, these files are too simple for diff to work out what to do with them.

I have never used xxdiff so can't say if it would produce better output.

thx for the explanation. is there a way to disable this feature, or another utility program that only compares line per line?

berbae · 11-02-2009, 07:41 AM

If all the lines are different, as shown in your example, I don't think that vimdiff, or some other diff program, is the right way to cope with your problem.
Be more clear about what may differ in the files you want to compare, and about what you want the comparison for.
What do you intend to do ?

chrism01 · 11-02-2009, 07:00 PM

Just fyi hiffens => hyphens ; your spelling threw me for a minute there

101b147 · 11-04-2009, 03:44 PM

Quote:

Originally Posted by berbae

If all the lines are different, as shown in your example, I don't think that vimdiff, or some other diff program, is the right way to cope with your problem.
Be more clear about what may differ in the files you want to compare, and about what you want the comparison for.
What do you intend to do ?

not all lines are different. only some of them. i need to comapre the output of my programs with the correct output. like i said, this example is silly.

Quote:

Just fyi hiffens => hyphens ; your spelling threw me for a minute there

don't remember very well, but i think i used the spell check b4 posting.

berbae · 11-05-2009, 04:37 AM

To visually check the differences, you can try :

Code:

pr -mT file1 file2
5                                   6
10 11 2 1 3                         10 9 4 13 17 3
4                                   5
5 3 4 2                             6 8 3 4 2

Is this of some help ?

Or something like that :

Code:

sed '{R file2
a \

}' file1
5
6

10 11 2 1 3
10 9 4 13 17 3

4
5

5 3 4 2
6 8 3 4 2

Or provide more precise example of the contents of the files.

101b147 · 11-13-2009, 07:09 PM

Quote:

Originally Posted by berbae

To visually check the differences, you can try :

Code:

pr -mT file1 file2
5                                   6
10 11 2 1 3                         10 9 4 13 17 3
4                                   5
5 3 4 2                             6 8 3 4 2

Is this of some help ?

Or something like that :

Code:

sed '{R file2
a \

}' file1
5
6

10 11 2 1 3
10 9 4 13 17 3

4
5

5 3 4 2
6 8 3 4 2

Or provide more precise example of the contents of the files.

sorry for the delay in my response. i tried pr and sed, but they show equal lines as well. i'll try to clarify my needs again.

basically, what i want is a program that compares line per line. if ith line from file1 differs from ith line from file2, print i(the line number, so i cant check them).

i checked a little bit the internet, and i found cmp and comm, but these don't do exactly what i want. cmp print differing bytes, and comm needs the lines to be sorted. like Vrajgh, diff doesn't work because it's a utility that is used mostly to compare the same file, but from different versions, with some modifications; it does this to generate a patch. diff works for some of my files, given their size and relatively small number of differences, but fails in the examples i gave. still, i'd like to have something that works always for my files.

the program i want is easy to implement, but i need to check files not only in my pc, so i was looking for some program that comes in most linux distros. sed seems to be enough, but i don't know how to use it( i'm n00b, remember), so if some1 can post a sed expression that does what i need, it would help me a lot.

besides what i said above, really thx in advance to all of you guys, who helped me 'till now, even in such a silly question.

berbae · 11-14-2009, 08:54 AM

Continuing to use the pr command :

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) print NR}'

only prints the line numbers of the different lines.
and

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) printf("%s\n%s\n\n",$1,$2)}'

prints the different lines together.

I used the ! as field separator, but you should use a character which doesn't appear in the files.

I tested it with :
file1

Code:

1
1 2
1 2 3
1 2 3 4
1 2 3 4 5
1 2 3 4 5 6
1 2 3 4 5 6 7
1 2 3 4 5 6 7 8
1 2 3 4 5 6 7 8 9

file2

Code:

1
1 2
1 3 2
1 2 3 4
1 2 3 4 5
1 2 4 3 5 6
1 2 3 4 5 6 7
1 2 3 4 5 7 6 8
1 2 3 4 5 6 7 8 9

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) print NR}'
3
6
8

and

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) printf("%s\n%s\n\n",$1,$2)}'
1 2 3
1 3 2

1 2 3 4 5 6
1 2 4 3 5 6

1 2 3 4 5 6 7 8
1 2 3 4 5 7 6 8

I don't know if that is the answer to your problem, but what I suggest gives what you tell you are looking for.
You didn't provide a real example of the structure of the files you want to compare, so I had to guess with what I present here.

101b147 · 12-04-2009, 03:22 PM

Quote:

Originally Posted by berbae

Continuing to use the pr command :

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) print NR}'

only prints the line numbers of the different lines.
and

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) printf("%s\n%s\n\n",$1,$2)}'

prints the different lines together.

I used the ! as field separator, but you should use a character which doesn't appear in the files.

I tested it with :
file1

Code:

1
1 2
1 2 3
1 2 3 4
1 2 3 4 5
1 2 3 4 5 6
1 2 3 4 5 6 7
1 2 3 4 5 6 7 8
1 2 3 4 5 6 7 8 9

file2

Code:

1
1 2
1 3 2
1 2 3 4
1 2 3 4 5
1 2 4 3 5 6
1 2 3 4 5 6 7
1 2 3 4 5 7 6 8
1 2 3 4 5 6 7 8 9

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) print NR}'
3
6
8

and

Code:

pr -mTs! file1 file2|awk -F! '{if ($1 != $2) printf("%s\n%s\n\n",$1,$2)}'
1 2 3
1 3 2

1 2 3 4 5 6
1 2 4 3 5 6

1 2 3 4 5 6 7 8
1 2 3 4 5 7 6 8

I don't know if that is the answer to your problem, but what I suggest gives what you tell you are looking for.
You didn't provide a real example of the structure of the files you want to compare, so I had to guess with what I present here.

hey guys, long time no see

. the above is exactly what i need! really thx to all of u and ur patience!!