[RESOLVED] BASH: read from end of file upwards

[AlbertGM]

Hi all,

I'm not sure if this is the right place to ask for a doubt in Bash programming. If it is not, please sorry.

I need to read a text file from last line to first line, ie in reverse order. To read I do:

cat $VAR | while read line; do  
	INPUT=$(echo ${line})
	echo "./dplt-catalog.sh undeploy-'$INPUT'"
done

where $VAR contains the name of the file. My first try was to read whole file counting the lines in file. Once I'd know the number of lines I'm able to read the until the last line and treat, then I read again to the second last and treat it, and so forth.
This looks terrible, right? Because in need to read file #lines + 1 times. I'm quite sure Linux/UNIX has a command or a way to do it much more simple. For example to list the contains of a directory in reverse alphabetical order:

ls -l | sort -r

So, I'm searching something like this. I hope you understand what I try.
Any hint, please?

Thanks a lot

Albert.

[AlbertGM]

I found a possible solution:

cat $VAR -n | sort -r -n | cut -f 2-

Anyway, any other hint or tip will be very welcome.

Albert.

[AmadeusAD]

There is a native coreutil:

tac <filename>

produces reversed output line by line, pretty like cat.

[AmadeusAD]

А на этом форуме пишут по-русски?)

[AlbertGM]

There is a native coreutil:

tac <filename>

produces reversed output line by line, pretty like cat.

That's exactly what I was searching.... I've been wasting my time for 2 or 3 hours... but I learned many things
Thanks a lot!!!!

Albert.

[AlbertGM]

А на этом форуме пишут по-русски?)

I already have many problems in English. Russian is impossible.
Well, if it's Russian... or is it Ukrainian?

Albert.

[AmadeusAD]

Sorry, I posted that occasionally, just for fun

Actually, I like the process of spending hours solving a problem too, that brings a lot to one's qualification. And was trying to construct something like the command line you showed above (of three commmands using pipes). Composed that quickly on perl, but in bash you were quicker Anyway, the string manipulation tasks, if they cannot be performed by a set of standard utils or it looks clumsy, have better to be performed with perl, awk, sed or something of that series (if there is not particular requirements).

[AmadeusAD]

That was Russian, you are right
The question asks whether one can post in Russian on this forum. Nevermind

[AlbertGM]

That was Russian, you are right
The question asks whether one can post in Russian on this forum. Nevermind

You can post... but I guess few people will be able to understand, and even less to answer

if they cannot be performed by a set of standard utils or it looks clumsy, have better to be performed with perl, awk, sed or something of that series (if there is not particular requirements).

No, there's no requirement, but I don't know nothing about perl, awk or sed. Well, in fact, I don't know almost nothing about BASH, but I need to make some scripts in Linux.... so I must learn

Thanks again

Albert.