Question for Unix guru's - diff command

[calbear2000]

Hi,
I have 2 gigantic files I'd like to diff. After diff runs for a few minutes, I get "diff: files too big, try -h"


I was thinking of dividing each file into 2 halves so that I can diff each half at a time. Is there a way to break up a file without manually going in with a text editor to cut and paste?

Thanks

 

[Bootprint]

cat and tail

or Split

 

[calbear2000]

Excellent... the split command. Thanks Bootprint

"Oh mighty unix, is there anything you can't do?"

 

[manly]

I'm surprised diff cares how long the files are. Another way of doing this would be to compute a hash of each file, and compare the hashes.

The command md5sum is common on many UNIX systems; MD5 is a de facto standard for computing cryptographic hashes (although it's somewhat being deprecated in favor of SHA).

Go Bears!

 

[ggavinmoss]

Originally posted by: manly
I'm surprised diff cares how long the files are. Another way of doing this would be to compute a hash of each file, and compare the hashes.

The command md5sum is common on many UNIX systems; MD5 is a de facto standard for computing cryptographic hashes (although it's somewhat being deprecated in favor of SHA).

Go Bears!

Wouldn't this only help if he wanted to distinguish if the files were different and not HOW they were different?

-geoff

 

[manly]

Originally posted by: ggavinmoss

Wouldn't this only help if he wanted to distinguish if the files were different and not HOW they were different?

-geoff

Yes, but most times people don't diff any text files larger than a few Kbytes in size for the actual deltas. Gargantuan files are typically binary of course.

Since he mentioned a text editor, I suppose it's (remotely) possible he's interested in the actual differences between two "gigantic" files.