"The Unix Programming Environment" states that '$' used in regular expression in sed means end-of-the line which is fine for me, because
cat file | sed 's/$/\n/'
is interpretted as "add newline at the end of each line".
The question arises, when I try to use the command:
cat file | sed '$d'
Shouldn't this line remove each line instead of the last one? In this context, dollar sign means end of the LAST line. What am I getting wrong?
$ is treated as regex anchor when used in pattern in s command e.g.
s/$/\n
However in $d, $ is not a regex anchor, it is address notation that means the last line of the input, which is deleted using the d command.
Also note that cat is unnecessary in your last command. It can be used as:
sed '$d' file
In the second usage, there is no regular expression. The $ there is an address, meaning the last line.
Note that regex in sed must be inside the delimiters(;,:, ~, etc) other than quotes.
/regex/
ex:
sed '/foo/s/bar/bux/g' file
or
~regex~
ex:
sed 's~dd~s~' file
but not 'regex'. So $ in '$d' won't be considered as regex by sed. '$d' acts like an address which points out the last line.
Related
how to remove comment lines (as # bal bla ) and empty lines (lines without charecters) from file with one sed command?
THX
lidia
If you're worried about starting two sed processes in a pipeline for performance reasons, you probably shouldn't be, it's still very efficient. But based on your comment that you want to do in-place editing, you can still do that with distinct commands (sed commands rather than invocations of sed itself).
You can either use multiple -e arguments or separate commands with a semicolon, something like (just one of these, not both):
sed -i 's/#.*$//' -e '/^$/d' fileName
sed -i 's/#.*$//;/^$/d' fileName
The following transcript shows this in action:
pax> printf 'Line # with a comment\n\n# Line with only a comment\n' >file
pax> cat file
Line # with a comment
# Line with only a comment
pax> cp file filex ; sed -i 's/#.*$//;/^$/d' filex ; cat filex
Line
pax> cp file filex ; sed -i -e 's/#.*$//' -e '/^$/d' filex ; cat filex
Line
Note how the file is modified in-place even with two -e options. You can see that both commands are executed on each line. The line with a comment first has the comment removed then all is removed because it's empty.
In addition, the original empty line is also removed.
#paxdiablo has a good answer but it can be improved.
(1) The '/^$/d' clause only matches 100% blank lines.
If you want to also match lines that are entirely whitespace (spaces, tabs etc.) use this instead:
'/^\s*$/d'
(2) The 's/#.*$//' clause only matches lines that start with the # character in column 0.
If you want to also match lines that have only whitespace before the first # use this instead:
'/^\s*#.*$/d'
The above criteria may not be universal (e.g. within a HEREDOC block, or in a Python multi-line string the different approaches could be significant), but in many cases the conventional definition of "blank" lines include whitespace-only, and "comment" lines include whitespace-then-#.
(3) Lastly, on OSX at least, the #paxdiablo solution in which the first clause turns comment lines into blank lines, and the second clause strips blank lines (including what were originally comments) doesn't work. It seems to be more portable to make both clauses /d delete actions as I've done.
The revised command incorporating the above is:
sed -e '/^\s*#.*$/d' -e '/^\s*$/d' inputFile
This tiny jewel removes all # comments, no matter where they begin in a line (see caution below):
sed -e 's/\s*#.*$//'
Example:
text="
this is a # test
#this is a test
#this is a #test
this is # another #test
"
$echo "$text" | sed -e 's/\s*#.*$//'
this is a
this is
Next this removes any resulting blank lines:
$echo "$text" | sed -e 's/\s*#.*$//' | sed -e '/^\s*$/d'
Caution: Depending on the syntax and/or interpretation of the lines your processing, this might not be an appropriate solution, as it just stupidly removes end of lines, even if the '#' is part of your data or code. However, for use cases where you'll never use a hash except for as an end of line comment then it works fine. So just as with all coding, context must be taken into consideration.
Alternative variant, using grep:
cat file.txt | grep -Ev '(#.*$)|(^$)'
you can use awk
awk 'NF{gsub(/^[ \t]*#/,"");print}' file
First example(paxdiablo) is very good except its not change file, just output result. If you want to change it inline:
sudo sed -i 's/#.*$//;/^$/d' inputFile
On (one of) my linux boxes, sed understands extended regular expressions with the -r option, so:
sed -r '/(^\s*#)|(^\s*$)/d' squid.conf.installed
is very useful for showing all non-blank, non comment lines.
The regex matches either start of line followed by zero or more spaces or tabs followed by either a hash or end of line, and deletes those matching lines from the input.
I wanted to use grep to exclude words from $lastblock by using a pipeline, but I found that grep works only for files, not for stdout output.
So, here is what I'm using:
lastblock="./2.json"
echo $lastblock | sed '1,/firstmatch/d;/.json/,$d'
I want to exclude ./ and .json, keeping only what is between.
This sed command is correct for this purpose, but how to escape the ./ replacing firstmatch so it can work?
Thanks in advance!
Use bash's Parameter Substitution
lastblock="./2.json"
name="${lastblock##*/}" # strips from the beginning until last / -> 2.json
base="${name%.*}" # strips from the last . to the end -> 2
but I found that grep works only for files, not for stdout output.
here it is. (if your grep supports the -P flag.
lastblock="./2.json"
echo "$lastblock" | grep -Po '(?<=\./).*(?=\.)'
but how to escape the ./
With sed(1), escape it using a back slash \
lastblock="./2.json"
echo "$lastblock" | sed 's/^\.\///;s/\..*$//'
Or use a different delimiter like a pipe |
sed 's|^\./||;s|\..*$||'
with awk
lastblock="./2.json"
echo "$lastblock" | awk -F'[./]+' '{print $2}'
Starting from bashv3, regular expression pattern matching is supported using the =~ operator inside the [[ ... ]] keyword.
lastblock="./2.json"
regex='^\./([[:digit:]]+)\.json'
[[ $lastblock =~ $regex ]] && echo "${BASH_REMATCH[1]}"
Although a P.E. should suffice just for this purpose.
I wanted to use grep to exclude words from $lastblock by using a pipeline, but I found that grep works only for files, not for stdout output.
Nonsense. grep works the same for the same input, regardless of whether it is from a file or from the standard input.
So, here is what I'm using:
lastblock="./2.json"
echo $lastblock | sed '1,/firstmatch/d;/.json/,$d'
I want to exclude ./ and .json, keeping only what is between. This sed
command is correct for this purpose,
That sed command is nowhere near correct for the stated purpose. It has this effect:
delete every line from the very first one up to and including the next subsequent one that matches the regular expression /firstmatch/, AND
delete every line from the first one matching the regular expression /.json/ to the last one of the file (and note that . is a regex metacharacter).
To remove part of a line instead of deleting a whole line, use an s/// command instead of a d command. As for escaping, you can escape a character to sed by preceding it with a backslash (\), which itself must be quoted or escaped to protect it from interpretation by the shell. Additionally, most regex metacharacters lose their special significance when they appear inside a character class, which I find to be a more legible way to include them in a pattern as literals. For example:
lastblock="./2.json"
echo "$lastblock" | sed 's/^[.]\///; s/[.]json$//'
That says to remove the literal characters ./ appearing at the beginning of the (any) line, and, separately, to remove the literal characters .json appearing at the end of the line.
Alternatively, if you want to modify only those lines that both start with ./ and end with .json then you can use a single s command with a capturing group and a backreference:
lastblock="./2.json"
echo "$lastblock" | sed 's/^[.]\/\(.*\)[.]json$/\1/'
That says that on lines that start with ./ and end with .json, capture everything between those two and replace the whole line with the captured part alone.
You can use another character like '#' when you want to avoid slashes.
You can remember a part that matches and use it in the replacement.
Use [.] avoiding the dot to be any character.
echo "$lastblock" | sed -r 's#[.]/(.*)[.]json#\1#'
Solution!
Just discovered today the tr command thanks to this legendary, unrelated answer.
When searching all over Google for how to exclude "." and "/", 100% of StackOverflow answers didn't helped.
So, to escape characters from the output of a command, just append this pipe:
| tr -d "{character-emoji-anything-you-want-to-exclude}"
So, a full working and simple sample:
echo "./2.json" | tr -d "/" | tr -d "." | tr -d "json"
And done!
I can't figure how to tell sed dot match new line:
echo -e "one\ntwo\nthree" | sed 's/one.*two/one/m'
I expect to get:
one
three
instead I get original:
one
two
three
sed is line-based tool. I don't think these is an option.
You can use h/H(hold), g/G(get).
$ echo -e 'one\ntwo\nthree' | sed -n '1h;1!H;${g;s/one.*two/one/p}'
one
three
Maybe you should try vim
:%s/one\_.*two/one/g
If you use a GNU sed, you may match any character, including line break chars, with a mere ., see :
.
Matches any character, including newline.
All you need to use is a -z option:
echo -e "one\ntwo\nthree" | sed -z 's/one.*two/one/'
# => one
# three
See the online sed demo.
However, one.*two might not be what you need since * is always greedy in POSIX regex patterns. So, one.*two will match the leftmost one, then any 0 or more chars as many as possible, and then the rightmost two. If you need to remove one, then any 0+ chars as few as possible, and then the leftmost two, you will have to use perl:
perl -i -0 -pe 's/one.*?two//sg' file # Non-Unicode version
perl -i -CSD -Mutf8 -0 -pe 's/one.*?two//sg' file # S&R in a UTF8 file
The -0 option enables the slurp mode so that the file could be read as a whole and not line-by-line, -i will enable inline file modification, s will make . match any char including line break chars, and .*? will match any 0 or more chars as few as possible due to a non-greedy *?. The -CSD -Mutf8 part make sure your input is decoded and output re-encoded back correctly.
You can use python this way:
$ echo -e "one\ntwo\nthree" | python -c 'import re, sys; s=sys.stdin.read(); s=re.sub("(?s)one.*two", "one", s); print s,'
one
three
$
This reads the entire python's standard input (sys.stdin.read()), then substitutes "one" for "one.*two" with dot matches all setting enabled (using (?s) at the start of the regular expression) and then prints the modified string (the trailing comma in print is used to prevent print from adding an extra newline).
This might work for you:
<<<$'one\ntwo\nthree' sed '/two/d'
or
<<<$'one\ntwo\nthree' sed '2d'
or
<<<$'one\ntwo\nthree' sed 'n;d'
or
<<<$'one\ntwo\nthree' sed 'N;N;s/two.//'
Sed does match all characters (including the \n) using a dot . but usually it has already stripped the \n off, as part of the cycle, so it no longer present in the pattern space to be matched.
Only certain commands (N,H and G) preserve newlines in the pattern/hold space.
N appends a newline to the pattern space and then appends the next line.
H does exactly the same except it acts on the hold space.
G appends a newline to the pattern space and then appends whatever is in the hold space too.
The hold space is empty until you place something in it so:
sed G file
will insert an empty line after each line.
sed 'G;G' file
will insert 2 empty lines etc etc.
How about two sed calls:
(get rid of the 'two' first, then get rid of the blank line)
$ echo -e 'one\ntwo\nthree' | sed 's/two//' | sed '/^$/d'
one
three
Actually, I prefer Perl for one-liners over Python:
$ echo -e 'one\ntwo\nthree' | perl -pe 's/two\n//'
one
three
Below discussion is based on Gnu sed.
sed operates on a line by line manner. So it's not possible to tell it dot match newline. However, there are some tricks that can implement this. You can use a loop structure (kind of) to put all the text in the pattern space, and then do the operation.
To put everything in the pattern space, use:
:a;N;$!ba;
To make "dot match newline" indirectly, you use:
(\n|.)
So the result is:
root#u1804:~# echo -e "one\ntwo\nthree" | sed -r ':a;N;$!ba;s/one(\n|.)*two/one/'
one
three
root#u1804:~#
Note that in this case, (\n|.) matches newline and all characters. See below example:
root#u1804:~# echo -e "oneXXXXXX\nXXXXXXtwo\nthree" | sed -r ':a;N;$!ba;s/one(\n|.)*two/one/'
one
three
root#u1804:~#
What I have:
test
more text
#user653434 text and so
test
more text
#user9659333 text and so
I'd like to filter this text and finally get the following list as .txt file:
user653434
user9659333
It's important to get the names without "#" sign.
Thx for help ;)
Using grep -P (requires GNU grep):
$ grep -oP '(?<=#)\w+' File
user653434
user9659333
-o tells grep to print only the match.
-P tells grep to use Perl-style regular expressions.
(?<=#) tells sed that # must precede the match but the # is not included in the match.
\w+ matches one or more word characters. This is what grep will print.
To change the file in place with grep:
grep -oP '(?<=#)\w+' File >tmp && mv tmp File
Using sed
$ sed -En 's/^#([[:alnum:]]+).*/\1/p' File
user653434
user9659333
And, to change the file in place:
sed -En -i.bak 's/^#([[:alnum:]]+).*/\1/p' File
-E tells sed to use the extended form of regular expressions. This reduces the need to use escapes.
-n tells sed not to print anything unless we explicitly ask it to.
-i.bak tells sed to change the file in place while leaving a backup file with the extension .bak.
The leading s in s/^#([[:alnum:]]+).*/\1/p tells sed that we are using a substitute command. The command has the typical form s/old/new/ where old is a regular expression and sed replaces old with new. The trailing p is an option to the substitute command: the p tells sed to print the resulting line.
In our case, the old part is ^#([[:alnum:]]+).*. Starting from the beginning of the line, ^, this matches # followed by one or more alphanumeric characters, ([[:alnum:]]+), followed by anything at all, .*. Because the alphanumeric characters are placed in parens, this is saved as a group, denoted \1.
The new part of the substitute command is just \1, the alphanumeric characters from above which comprise the user name.
Here, the s indicates that we are using a sed substitute command. The usual form
With GNU grep:
grep -Po '^#\K[^ ]*' file
Output:
user653434
user9659333
See: The Stack Overflow Regular Expressions FAQ
How to insert a newline before a pattern within a line?
For example, this will insert a newline behind the regex pattern.
sed 's/regex/&\n/g'
How can I do the same but in front of the pattern?
Given this sample input file, the pattern to match on is the phone number.
some text (012)345-6789
Should become
some text
(012)345-6789
This works in bash and zsh, tested on Linux and OS X:
sed 's/regexp/\'$'\n/g'
In general, for $ followed by a string literal in single quotes bash performs C-style backslash substitution, e.g. $'\t' is translated to a literal tab. Plus, sed wants your newline literal to be escaped with a backslash, hence the \ before $. And finally, the dollar sign itself shouldn't be quoted so that it's interpreted by the shell, therefore we close the quote before the $ and then open it again.
Edit: As suggested in the comments by #mklement0, this works as well:
sed $'s/regexp/\\\n/g'
What happens here is: the entire sed command is now a C-style string, which means the backslash that sed requires to be placed before the new line literal should now be escaped with another backslash. Though more readable, in this case you won't be able to do shell string substitutions (without making it ugly again.)
Some of the other answers didn't work for my version of sed.
Switching the position of & and \n did work.
sed 's/regexp/\n&/g'
Edit: This doesn't seem to work on OS X, unless you install gnu-sed.
In sed, you can't add newlines in the output stream easily. You need to use a continuation line, which is awkward, but it works:
$ sed 's/regexp/\
&/'
Example:
$ echo foo | sed 's/.*/\
&/'
foo
See here for details. If you want something slightly less awkward you could try using perl -pe with match groups instead of sed:
$ echo foo | perl -pe 's/(.*)/\n$1/'
foo
$1 refers to the first matched group in the regular expression, where groups are in parentheses.
On my mac, the following inserts a single 'n' instead of newline:
sed 's/regexp/\n&/g'
This replaces with newline:
sed "s/regexp/\\`echo -e '\n\r'`/g"
echo one,two,three | sed 's/,/\
/g'
You can use perl one-liners much like you do with sed, with the advantage of full perl regular expression support (which is much more powerful than what you get with sed). There is also very little variation across *nix platforms - perl is generally perl. So you can stop worrying about how to make your particular system's version of sed do what you want.
In this case, you can do
perl -pe 's/(regex)/\n$1/'
-pe puts perl into a "execute and print" loop, much like sed's normal mode of operation.
' quotes everything else so the shell won't interfere
() surrounding the regex is a grouping operator. $1 on the right side of the substitution prints out whatever was matched inside these parens.
Finally, \n is a newline.
Regardless of whether you are using parentheses as a grouping operator, you have to escape any parentheses you are trying to match. So a regex to match the pattern you list above would be something like
\(\d\d\d\)\d\d\d-\d\d\d\d
\( or \) matches a literal paren, and \d matches a digit.
Better:
\(\d{3}\)\d{3}-\d{4}
I imagine you can figure out what the numbers in braces are doing.
Additionally, you can use delimiters other than / for your regex. So if you need to match / you won't need to escape it. Either of the below is equivalent to the regex at the beginning of my answer. In theory you can substitute any character for the standard /'s.
perl -pe 's#(regex)#\n$1#'
perl -pe 's{(regex)}{\n$1}'
A couple final thoughts.
using -ne instead of -pe acts similarly, but doesn't automatically print at the end. It can be handy if you want to print on your own. E.g., here's a grep-alike (m/foobar/ is a regex match):
perl -ne 'if (m/foobar/) {print}'
If you are finding dealing with newlines troublesome, and you want it to be magically handled for you, add -l. Not useful for the OP, who was working with newlines, though.
Bonus tip - if you have the pcre package installed, it comes with pcregrep, which uses full perl-compatible regexes.
In this case, I do not use sed. I use tr.
cat Somefile |tr ',' '\012'
This takes the comma and replaces it with the carriage return.
To insert a newline to output stream on Linux, I used:
sed -i "s/def/abc\\\ndef/" file1
Where file1 was:
def
Before the sed in-place replacement, and:
abc
def
After the sed in-place replacement. Please note the use of \\\n. If the patterns have a " inside it, escape using \".
Hmm, just escaped newlines seem to work in more recent versions of sed (I have GNU sed 4.2.1),
dev:~/pg/services/places> echo 'foobar' | sed -r 's/(bar)/\n\1/;'
foo
bar
echo pattern | sed -E -e $'s/^(pattern)/\\\n\\1/'
worked fine on El Captitan with () support
In my case the below method works.
sed -i 's/playstation/PS4/' input.txt
Can be written as:
sed -i 's/playstation/PS4\nplaystation/' input.txt
PS4
playstation
Consider using \\n while using it in a string literal.
sed : is stream editor
-i : Allows to edit the source file
+: Is delimiter.
I hope the above information works for you 😃.
in sed you can reference groups in your pattern with "\1", "\2", ....
so if the pattern you're looking for is "PATTERN", and you want to insert "BEFORE" in front of it, you can use, sans escaping
sed 's/(PATTERN)/BEFORE\1/g'
i.e.
sed 's/\(PATTERN\)/BEFORE\1/g'
You can also do this with awk, using -v to provide the pattern:
awk -v patt="pattern" '$0 ~ patt {gsub(patt, "\n"patt)}1' file
This checks if a line contains a given pattern. If so, it appends a new line to the beginning of it.
See a basic example:
$ cat file
hello
this is some pattern and we are going ahead
bye!
$ awk -v patt="pattern" '$0 ~ patt {gsub(patt, "\n"patt)}1' file
hello
this is some
pattern and we are going ahead
bye!
Note it will affect to all patterns in a line:
$ cat file
this pattern is some pattern and we are going ahead
$ awk -v patt="pattern" '$0 ~ patt {gsub(patt, "\n"patt)}1' d
this
pattern is some
pattern and we are going ahead
sed -e 's/regexp/\0\n/g'
\0 is the null, so your expression is replaced with null (nothing) and then...
\n is the new line
On some flavors of Unix doesn't work, but I think it's the solution to your problem.
echo "Hello" | sed -e 's/Hello/\0\ntmow/g'
Hello
tmow
This works in MAC for me
sed -i.bak -e 's/regex/xregex/g' input.txt sed -i.bak -e 's/qregex/\'$'\nregex/g' input.txt
Dono whether its perfect one...
After reading all the answers to this question, it still took me many attempts to get the correct syntax to the following example script:
#!/bin/bash
# script: add_domain
# using fixed values instead of command line parameters $1, $2
# to show typical variable values in this example
ipaddr="127.0.0.1"
domain="example.com"
# no need to escape $ipaddr and $domain values if we use separate quotes.
sudo sed -i '$a \\n'"$ipaddr www.$domain $domain" /etc/hosts
The script appends a newline \n followed by another line of text to the end of a file using a single sed command.
In vi on Red Hat, I was able to insert carriage returns using just the \r character. I believe this internally executes 'ex' instead of 'sed', but it's similar, and vi can be another way to do bulk edits such as code patches. For example. I am surrounding a search term with an if statement that insists on carriage returns after the braces:
:.,$s/\(my_function(.*)\)/if(!skip_option){\r\t\1\r\t}/
Note that I also had it insert some tabs to make things align better.
Just to add to the list of many ways to do this, here is a simple python alternative. You could of course use re.sub() if a regex were needed.
python -c 'print(open("./myfile.txt", "r").read().replace("String to match", "String to match\n"))' > myfile_lines.txt
sed 's/regexp/\'$'\n/g'
works as justified and detailed by mojuba in his answer .
However, this did not work:
sed 's/regexp/\\\n/g'
It added a new line, but at the end of the original line, a \n was added.