linux bash scripting to remove certain text

Question

i have a bash script like this:

#!/bin/bash
# ALL HTML FILES
FILES="*.html"
# for loop read each file
for f in $FILES
do
INF="$f"
OUTF="$f.out.tmp"
# replace javascript
sed '/<!--fff309/,/<!--\/fff309-->/d' $INF > $OUTF
/bin/cp $OUTF $INF
/bin/rm -f $OUTF
done

the html is look like this:

<html>
<body>
<div>some normal html code</div><!--fff309-->some javascript code goes here... <!--/fff309-->

<div>
some other html....
</div>
</body>
</html>

the bash script work, however it remove all the html part below the 

so it become:

<html>
<body>
<div>some normal html code</div>

is there anyway to do so that it only remove the part:

<!--fff309--> ...  <!--/fff309-->

Thanks

That sed script works fine on my system (with GNU sed version 4.2.1). What does sed --version report on your system? — ruakh
– ruakh, Commented Dec 4, 2012 at 6:25
Also OK on Mac OS X (BSD non-GNU, there is no -version or --version — Matteo
– Matteo, Commented Dec 4, 2012 at 6:27
this is my OS: Linux abc.com 2.6.32-279.14.1.el6.x86_64 #1 SMP Tue Nov 6 23:43:09 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux — Redbox
– Redbox, Commented Dec 4, 2012 at 6:30

Vijay · Accepted Answer · 2012-12-04 08:42:03Z

1

May be you can use perl inplace replacement for this as below: i had changed your script as below to make it more simple.

#!/bin/bash
# ALL HTML FILES
FILES="*.html"
# for loop read each file
for f in $FILES
do
perl -i -lne 'print unless(/\<\!--fff309--\>/.../\<\!--\/fff309--\>/)' $f
done

Main task is done by the line:

perl -i -lne 'print unless(/\<\!--fff309--\>/.../\<\!--\/fff309--\>/)'

This does change the file directly witout having a need to create a temporary file. I testted it and the command works for me.

Also if you insist on using sed the below works for me:

sed '/\<\!--fff309--\>/,/\<\!--\/fff309--\>/d' your_file

for your case it would be better to use -i option of sed which does in place replacement as below.

sed -i '/\<\!--fff309--\>/,/\<\!--\/fff309--\>/d' your_file

so using this your script will be changed to as below:

#!/bin/bash
# ALL HTML FILES
FILES="*.html"
# for loop read each file
for f in $FILES
do
sed '/\<\!--fff309--\>/,/\<\!--\/fff309--\>/d' $f
done

After looking at the HTML file:

this is the thing that you need:

perl -pi -e 's/\<\!--fff309--\>.*\<\!--\/fff309--\>//g' your_file

so your script becomes:

#!/bin/bash
# ALL HTML FILES
FILES="*.html"
# for loop read each file
for f in $FILES
do
perl -pi -e 's/\<\!--fff309--\>.*\<\!--\/fff309--\>//g' $f
done

edited Dec 4, 2012 at 8:42

answered Dec 4, 2012 at 6:57

Vijay

67.7k94 gold badges238 silver badges327 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

Redbox Over a year ago

same problem, it remove all the text below it.. check the original html here: pastebin.com/jpLf5L7j

Redbox Over a year ago

sed -i '/\<\!--fff309--\>/,/\<\!--\/fff309--\>/d' does not remove anynthing.. but this will: sed -i '/<\!--fff309-->/,/<\!--\/fff309-->/d' however it remove all part below it.

Vijay Over a year ago

I checked your html file.i see both FFF309's are in the same line.but your question says that they are on different lines.So change your question.i am downvoting your question since it is wrongly framed.And this is the reason why all the lines below the first find of FFF309 will be removed with the regex.

Redbox Over a year ago

sorry, i dont know about that.. what is the solution for that? i just want that part to be removed

Redbox Over a year ago

Bingo! now it work fine.. do you know how to loop all the subdirectory?

|

Rubens · Accepted Answer · 2012-12-04 07:17:07Z

1

The problem is not on your sed, but in your bash script. Remove the double quotes from "*.html".

Bash consider double quoted strings as an unique string, i. e., it won't separate the string using the value of IFS, neither expand your wildcards, avoiding, thus, the f variable on your script to be simply a <name>.html, and forcing it to be *.html.

If you echo the different results, you'll notice that, with double quotes, the variable f will be *.html, and when you echo it, the expansion will be performed, so that you'll have: <name_1>.html <name_2>.html ... <name_n>.html.

For example,

create two or more .html files in your directory;

execute the following script:

#!/bin/bash

echo "Incorrect:"    
for i in "*.html"; do
    echo $i;
done

echo "Correct:"
for i in *.html; do
    echo $i;
done

Try the following:

#!/bin/bash
# ALL HTML FILES
FILES=*.html
# for loop read each file
for f in $FILES
do
INF=$f
OUTF=$f.out.tmp
# replace javascript
sed '/<!--fff309/,/<!--\/fff309-->/d' $INF > $OUTF
/bin/cp $OUTF $INF
/bin/rm -f $OUTF
done

edited Dec 4, 2012 at 7:17

answered Dec 4, 2012 at 6:47

Rubens

14.9k11 gold badges68 silver badges94 bronze badges

5 Comments

Redbox Over a year ago

i removed. now become this: FILES=*.html, still got the same problem

Rubens Over a year ago

have you tried to use *.html directly in the script, just to check if the problem persists

Rubens Over a year ago

@Redbox I've tried to rewrite the .sh, with alterations; see my edit.

Redbox Over a year ago

your code work, but it still remove all part below the

Redbox Over a year ago

edited the question, the fff309 part is same line. sorry if that cause the problem?

Collectives™ on Stack Overflow

linux bash scripting to remove certain text

2 Answers 2

6 Comments

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

6 Comments

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related