Any Help would be appreciated

buchman · Feb 24, 2005, 12:52 AM

I've got this question for some coursework and i'm stuck, didn't know which forum to post, but the linux one looked best as its unix commands

Q.
Write a pipeline that updates the content of a file (a file of your choice) by replacing all matches of the word ‘and’ with the symbol ‘&’ and by replacing the end of each line with “your name”. If the command completes successfully it should print ‘YES’, otherwise it should print ‘NO’.

I've got to write it using standard linux/unix commands and i'm stuck

As in the title any help would be appreciated.

also does anyone know what 'grep 1' does i've read the help, but still don't have a clue

UberLord · Feb 24, 2005, 10:30 AM

Code:

[ -f ./foo -a -s ./foo ] && sed -i 's/and/\&/g;s/$/'`whoami`'/g' ./foo && echo YES || echo NO

Now, lets see what we do
[ -f ./foo -a -s ./foo ] tests if the ./foo exists and is a regular file and is not zero length
&& means if the previous statement is true then
sed - we'll do that later
|| means if anything prior is not true then

So in simple terms
"If the file ./foo exists and has content then run our sed statement and echo YES otherwise echo NO"

We do this as your argument must append your name to the send of each line - so it always should print YES unless it's an empty file or does not exist

Now for our sed statement - sed is a stream editor which is perfect for our needs. However, sed is very simple and yet very comples - here's a quick rundown on what it does

-i means edit file
(we could use -e to output file with our changes instead of editing it for testing)
s/foo/bar/g - this is a command that tells sed to replace the word "foo" with the word "bar". However, if the source had barfoobar in then barbarbar would be the result. The g bit on the end means global (ie every occurance)
The & symbol needs to be escaped so we use a \ to escape it (other chars need to be escaped like this too, like quotes, brackets, etc)
$ means "end of line"
`whoami` is a unix command that returns the name of the current user which we've embedded into our command

Hopefully this is the answer you need

buchman · Feb 24, 2005, 10:34 AM

well, WOW, and the explanation helps a lot too

thank you gonna test it now.

UberLord · Feb 24, 2005, 11:14 AM

Uh - I forgot to say that the ; char is a sed command seperator - it seperates our two substitution commands.

buchman · Feb 25, 2005, 08:29 AM

Thanks for the help that worked perfectly, you don't happen to know what 'grep 1' would do?

md5 · Feb 25, 2005, 09:53 AM

Quote:

Originally Posted by buchman

you don't happen to know what 'grep 1' would do?

Search its man page (type man grep) or have a look here

You can search for many Linux man pages online here

In short, grep 1 will filter all the output you feed it and will return you all the lines that contain the number "1" in them

UberLord · Feb 25, 2005, 12:16 PM

BTW, what I did isn't really a pipeline. You could use pipes if you wanted too, but this way is much more efficient

Feb 24, 2005, 12:52 AM	#1
buchman DriverHeaven Junior Member Join Date: Mar 2003 Location: London, UK Posts: 44 Rep Power: 0	Any Help would be appreciated I've got this question for some coursework and i'm stuck, didn't know which forum to post, but the linux one looked best as its unix commands Q. Write a pipeline that updates the content of a file (a file of your choice) by replacing all matches of the word ‘and’ with the symbol ‘&’ and by replacing the end of each line with “your name”. If the command completes successfully it should print ‘YES’, otherwise it should print ‘NO’. I've got to write it using standard linux/unix commands and i'm stuck As in the title any help would be appreciated. also does anyone know what 'grep 1' does i've read the help, but still don't have a clue Last edited by buchman; Feb 24, 2005 at 01:10 AM.

Feb 24, 2005, 10:30 AM	#2
UberLord A Legend in Underwear Join Date: May 2002 Location: Unknown Posts: 5,255 Rep Power: 0	Code: [ -f ./foo -a -s ./foo ] && sed -i 's/and/\&/g;s/$/'`whoami`'/g' ./foo && echo YES \|\| echo NO Now, lets see what we do [ -f ./foo -a -s ./foo ] tests if the ./foo exists and is a regular file and is not zero length && means if the previous statement is true then sed - we'll do that later \|\| means if anything prior is not true then So in simple terms "If the file ./foo exists and has content then run our sed statement and echo YES otherwise echo NO" We do this as your argument must append your name to the send of each line - so it always should print YES unless it's an empty file or does not exist Now for our sed statement - sed is a stream editor which is perfect for our needs. However, sed is very simple and yet very comples - here's a quick rundown on what it does -i means edit file (we could use -e to output file with our changes instead of editing it for testing) s/foo/bar/g - this is a command that tells sed to replace the word "foo" with the word "bar". However, if the source had barfoobar in then barbarbar would be the result. The g bit on the end means global (ie every occurance) The & symbol needs to be escaped so we use a \ to escape it (other chars need to be escaped like this too, like quotes, brackets, etc) $ means "end of line" `whoami` is a unix command that returns the name of the current user which we've embedded into our command Hopefully this is the answer you need __________________ Gentoo Linux - Developer (baselayout) Read my blog "I contend that we are both atheists. I just believe in one fewer god than you do. When you understand why you dismiss all the other possible gods, you will understand why I dismiss yours." Stephen Roberts

Feb 24, 2005, 11:14 AM	#4
UberLord A Legend in Underwear Join Date: May 2002 Location: Unknown Posts: 5,255 Rep Power: 0	Uh - I forgot to say that the ; char is a sed command seperator - it seperates our two substitution commands. __________________ Gentoo Linux - Developer (baselayout) Read my blog "I contend that we are both atheists. I just believe in one fewer god than you do. When you understand why you dismiss all the other possible gods, you will understand why I dismiss yours." Stephen Roberts

Feb 25, 2005, 12:16 PM	#7
UberLord A Legend in Underwear Join Date: May 2002 Location: Unknown Posts: 5,255 Rep Power: 0	BTW, what I did isn't really a pipeline. You could use pipes if you wanted too, but this way is much more efficient __________________ Gentoo Linux - Developer (baselayout) Read my blog "I contend that we are both atheists. I just believe in one fewer god than you do. When you understand why you dismiss all the other possible gods, you will understand why I dismiss yours." Stephen Roberts

Feb 24, 2005, 10:34 AM	Thread Starter #3
buchman DriverHeaven Junior Member Join Date: Mar 2003 Location: London, UK Posts: 44 Rep Power: 0	well, WOW, and the explanation helps a lot too thank you gonna test it now.

Feb 25, 2005, 08:29 AM	Thread Starter #5
buchman DriverHeaven Junior Member Join Date: Mar 2003 Location: London, UK Posts: 44 Rep Power: 0	Thanks for the help that worked perfectly, you don't happen to know what 'grep 1' would do?