LinuxQuestions.org - How to remove the last \n from a file .

Page 1 of 2

Show 50 post(s) from this thread on one page

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - How to remove the last \n from a file . (https://www.linuxquestions.org/questions/linux-newbie-8/how-to-remove-the-last-%5Cn-from-a-file-794369/)

sumanch

03-10-2010 12:41 AM

How to remove the last \n from a file .

I have a file sample.xml . If I do "vi sample.xml" I can see the following

bash$ cat sample.xml
<content/>
bash$

but I don't want the last \n character so that it becomes following

bash$ cat sample.xml
<content/>bash$

Please let me know how do I achieve that ?

regards
Suman

vinaytp

03-10-2010 01:02 AM

Very simple

Code:

tr -d '\n'< sample.xml

kainosnous

03-10-2010 03:01 AM

Personally, I would just edit it in vi. Just type G and dd and you should be done. Was this a one time thing, or did you need a script to do this?

catkin

03-10-2010 03:05 AM

Quote:

Originally Posted by vinaytp (Post 3892497)

Very simple

Code:

tr -d '\n'< sample.xml

That would delete all line endings from sample.xml. Even if it were modified to remove only the/any line end from the end of the file it would not print the bash prompt after the file contents as the OP wants because bash starts a new line after command output, before displaying the prompt, as in this example:

Code:

c@CW8:~$ < /dev/null

c@CW8:~$

The only time I recall seeing the bash prompt immediately after command output is when the terminal's line end handling has been disabled, usually by accidentally sending the contents of a binary file to the terminal.

kainosnous

03-10-2010 03:21 AM

catkin is right. I missed that. It will still start the prompt on a new line. What were you trying to do? Perhaps there is another approach.

vinaytp

03-10-2010 04:21 AM

Quote:

Originally Posted by kainosnous (Post 3892589)

Personally, I would just edit it in vi. Just type G and dd and you should be done. Was this a one time thing, or did you need a script to do this?

Could you pleas tell me how this works ? Hope this deletes entire last line. Also even after deleting last line you will end up with \n in last but one line.

you may inspect this with Ocatal dump after G and dd

Code:

od -t c sample.xml

I thought \n has to be deleted entirely from the file. If you want to delete it only form the last line why can't you write a simple script to do that. Something similar to below will do

Code:

var=`wc sample.xml | tr -s ' ' | cut -d ' ' -f 2`

lines=`expr $var - 1`

head -n $lines sample.xml ; tail -1 sample.xml | tr -d '\n'

colucix

03-10-2010 04:58 AM

I have some doubts about bash printing a new prompt on a new line after command output. If you print out some text that misses the last newline, the prompt should actually be placed inline. For example, here is an alternative in gawk to remove the last newline (I think sed can do it better, but I cannot catch it, now...):

Code:

[colucix@ocean-4 ~]$ cat testfile

line one

line two

line three

[colucix@ocean-4 ~]$ awk 'NR==1{("cat " FILENAME " | wc -l") | getline NL} NR < NL; END{printf "%s", $0}' testfile

line one

line two

line three[colucix@ocean-4 ~]$

Furthermore, if I redirect the output to a file and print out the content using cat:

Code:

[colucix@ocean-4 ~]$ awk 'NR==1{("cat " FILENAME " | wc -l") | getline NL} NR < NL; END{printf "%s", $0}' testfile > testfile_modified

[colucix@ocean-4 ~]$ cat testfile_modified

line one

line two

line three[colucix@ocean-4 ~]$

In other words apparently bash puts the prompt on a new line when you type a command (that does not produce any output) and press enter: actually, it puts the prompt on the same line with... "nothing".

Kenhelm

03-10-2010 12:06 PM

Code:

# Remove the last byte in a file

head -c-1 sample.xml > newsample.xml



# Remove the last byte in a file only if it is a newline.

# If the last byte isn't a '\n' then grep adds one, head removes it.

grep '^' sample.xml | head -c-1 - > newsample.xml

colucix

03-10-2010 12:25 PM

Quote:

Originally Posted by Kenhelm (Post 3893146)

Code:

# Remove the last byte in a file

head -c-1 sample.xml > newsample.xml



# Remove the last byte in a file only if it is a newline.

# If the last byte isn't a '\n' then grep adds one, head removes it.

grep '^' sample.xml | head -c-1 - > newsample.xml

Nice. Good catch!

jamescondron

03-10-2010 12:32 PM

You're all missing the point; the last line break is supposed to be there, its pretty important that it is, though for the life of me I can't remember why; something to do with how the FS works.

Bugger, let me find a link

EDIT:
Thats why, you need it for opening files in append mode, for code (some compilers crap themselves without it), for importing text as input.

The line isn't a '\n' at all. Open it in a hex editor and see

sundialsvcs

03-10-2010 10:59 PM

chomp! :hattip:

vinaytp

03-10-2010 11:04 PM

Quote:

Originally Posted by Kenhelm (Post 3893146)

Code:

# Remove the last byte in a file

head -c-1 sample.xml > newsample.xml



# Remove the last byte in a file only if it is a newline.

# If the last byte isn't a '\n' then grep adds one, head removes it.

grep '^' sample.xml | head -c-1 - > newsample.xml

Yeah, that's better.
Good sense of switch usage.

ghostdog74

03-10-2010 11:27 PM

Code:

awk '{q=p;p=$0}NR>1{print q}END{ORS = ""; print p}' file

aggresss

01-01-2019 09:09 AM

Helpful idea

It works , thanks a lot !

Quote:

Originally Posted by colucix (Post 3892680)

Code:

[colucix@ocean-4 ~]$ cat testfile

line one

line two

line three

[colucix@ocean-4 ~]$ awk 'NR==1{("cat " FILENAME " | wc -l") | getline NL} NR < NL; END{printf "%s", $0}' testfile

line one

line two

line three[colucix@ocean-4 ~]$

Furthermore, if I redirect the output to a file and print out the content using cat:

Code:

[colucix@ocean-4 ~]$ awk 'NR==1{("cat " FILENAME " | wc -l") | getline NL} NR < NL; END{printf "%s", $0}' testfile > testfile_modified

[colucix@ocean-4 ~]$ cat testfile_modified

line one

line two

line three[colucix@ocean-4 ~]$

MadeInGermany

01-01-2019 12:52 PM

With shell builtins

Code:

nl=

while IFS= read line

do

  printf "$nl%s" "$line"

  nl='\n'

done < textfile > incompletefile

Note that the standard text processing tools (like sed and awk) might not properly handle an incomplete text file.
So more often you want to do the opposite

Code:

{

while IFS= read line

do

  echo "$line"

done

[ -n "$line" ] && echo "$line"

} < incompletefile > textfile

All times are GMT -5. The time now is 12:05 AM.

Page 1 of 2

Show 50 post(s) from this thread on one page