maybe I'm missing something in the OP's requirement...
Both sed and grep are line-oriented utilities, so you'd need to orient the input into 'records' for it to be of any use.
Each line in the data is a record of comma-delimited fields. I don't see the problem with that.
Quote:
Originally Posted by curleb
You'd need to specify a full line match with -x or use ^...$ anchors to prevent partial matches from falsely eliminating keepers. It would also be a good idea to use -F, forcing fixed patterns, and -f, so that egrep reads the fixed strings from a file (that way, we can avoid a potentially lengthy command line): egrep -vxFf listfile
However, this approach irreversibly modifies the data, since after tr it's not possible to distinguish between newlines that were newlines and newlines that were commas, unless some restrictive assumptions can be made regarding the data (such as forbidding blank lines or null comma-delimited fields). This may not be a problem; I'm just mentioning it in case it is relevant to the original poster.
If an output format of one number per line is acceptable, the following might suffice:
Regards,
Alister
---------- Post updated at 03:49 PM ---------- Previous update was at 03:42 PM ----------
If the data must be left intact, format unmodified except for the deleted numbers, the following script would do the job:
Important: I have not modified IFS in the read statement, nor set read to raw mode, nor quoted $n, nor bothered to escape the contents of $n before inserting it into the ed regular expressions because I am assuming that each line in the list file consists of nothing but numerical digits. In sed basic regular expression syntax, each line in the list file should match /^[0-9]\{1,\}$/. If the following doesn't yield the entire list file's contents, then don't use the above code:
Test run (in what follows, the script above is rmnum.sh):
If the list may contain floating point numbers, whose decimal point would be considered a wildcard, I would opt to use perl. I don't use it often, but, as I recall, it supports quoting variable text within regular expressions, so that metacharacters are treated literally.
Hi all, I am trying to use SED to input lines to a file in a specific place.
So far I have;
sed '/#NewEntry/ i\Insert this line' myfile.txt
The output is printing to screen correctly but what is the best way to write it back to the file without overwriting everything?
I tried adding >... (6 Replies)
Hello Im fairly familiar with using the sed command for substitution, however I have been passed a script which checks the logged on username and directory type with a sed section which I cant figure out. The sed function has me baffled and I cant find out from the man page what its trying to do... (1 Reply)
I have a sed query.
There is a line which has tilde and I want to separate this line using sed.
The line is:
ABP_ETC_ROOT=~xdmadm
The query to get this line is:
sed -n '/\(.*\)~\(.*\)/p' infile
I want to get xdmadm from this line and I am using this sed command:
sed -n... (5 Replies)
hi
i had posted this earlier.. got no reply !!
how to change assigned value in a file using sed
suppose the file contains
age = 30;
how to change it to
age =50;
i tried sed 's/^age*./age =50;' filename but i am getting the o/p as
age =50; 30;
plz hlp!! (4 Replies)
I have done a script using sed which replaces a string in file.
I face 2 problems, while using sed command.
1)last line of the file is missing.
2)if i am are using wild card character in the command, Its not taking the next matching pattern ie if I am giving the pattern abc*def to be replaced... (2 Replies)
Hi Penchal,
I would appreciate if you can provide me a brief explanation on what you are trying to do in the commnad below.
echo "6-9-2008" | sed 's/\(.\)-\(.\)-\(.*\)/\3-0\2-0\1/g'
Thanks
Amit (1 Reply)
Hi
I have a file which looks like this
//string = "abcd"; //info
//string = "*pqrs"; //add
string = "#123"; //sub
//string = "#1234567890"
data = check(string)
//string = "#1234567890"
I want to modify this as
string = "#987"; //mult
data = check(string)
How do i do that? (1 Reply)
I'm writing a script which word counts the number of lines in two files. If one file is bigger than the other I'd like to edit one of the files to delete some lines to make both the same. It does not matter where in the file the lines are deleted from. It's expected that this will be ran from... (3 Replies)
I have a series of folders /temp/a /temp/b /temp/c
In folders a, b, and c, I have files
a1.txt..........a20.txt
b1.txt..........b40.txt &
c1.txt..........c60.txt
Each file has the same data format :-
Line 1 AAAAA aaaa
Line 2 BBB bbbbbb
Line 3 CCCC cccccc
Etc etc
I need to write a... (13 Replies)
hi, I have an xml file and I need to replace the tags with different names all at a time here is what I have
<cevalue>
<cevalue1>
<cevalue2>
<cevalue3>
<cevalue4>
<cevalue5>
and I need these like these...
<cevalue>
<cevalue>
<cevalue>
<cevalue>
<cevalue>
<cevalue>
I tried a few but... (2 Replies)