sed replace - need help replacing text in bunch of files

Question

How to parse bunch of files and replace each occurance of __esct function?

__esct('Full name')   

convert to:

$this->escape($this->translate('Full name'))  

I'm confused as to - how to add that second brace after replaced text?

Answer 1

Try using this :

sed 's@__esct(\(.*)\)@$this->escape($this->translate(\1)@'  file.txt

Answer 2

Perl has nice extensions to do something like that:

perl -pi -e 's;__esct\(\'([a-zA-Z ]*)\'\);\$this->escape\(\$this->translate\(\'\1\'));g'  list of the files to mangle

It will leave the originals as <file>.bak. This is just a global replace on each line, just as you'd write it in vi(1). In Perl, () and such are metacharacters, to match them you need \( \).

[I'm not 100% sure of the handling of the ', need to check what your particular shell does in such a case.]

Answer 3

sed -e 's/__esct(\(.*\))/$this->escape($this->translate(\1))/'

Should get you a long way. It'll fail if you have foo(__esct('Some text'))

sed -e 's/__esct(\([^(]*\))/$this->escape($this->translate(\1))/'

This one will work for those, but fail for __esct(foo('Some text'))