findr is a ruby find and replace tool for mass editing of text files using (advanced) regular expressions.

Usage: findr [options] <search regex> [<replacement string>]
    -g, --glob FILE SEARCH GLOB      e.g. "*.{rb,erb}"; glob relative to the current directory
    -G, --global-glob GLOB           e.g. "/**/*.{rb,erb}"; glob may escape the current directory
    -i, --ignore-case                do a case-insensitive search
    -n, --number-of-context LINES    number of lines in context to print
    -x, --execute                    actually execute the replacement
    -c, --coding FILE CODING SYSTEM  e.g. "iso-8859-1"
    -s, --save                       saves your options to .findr-config for future use
    -C, --codings-list               list available encodings
    -v, --verbose                    verbose output

findr recursively searches for the given regular expression in all files matching the given glob, starting from the current subdirectory.

• To search the term 123.456 in all txt-files:

    findr -g '*.txt' 123\\.456
  

  The *.txt must be escaped (from the shell) by using apostrophes, otherwise the shell might expand it to the names of txt files in the current directory.

  The dot must be escaped (from the regexp parser), otherwise is would stand for any character. To escape the dot---on most shells---you could either write two backslashes to feed one backslash into the program, or put the command under single apostrophes, like

    findr '123\.456'
  

• To replace the term 123.456 by 33.0:

    findr '123\.456' 33.0
  

  The previous command shows a preview of what would happen. Adding -x executes the change and irreversibly alters your files:

    findr 123\.456 33.0 -x
  

• To do a case-insensitive search for word use

    findr '(?i:word)'
  

• A more complicated example: findr allows to use matches in the replacement string and also allows the use of advanced regular expression features like lazy matching. Consider the following input file example.txt:

    a = 99
    b=123;
    var  = 44
    d = 55 ;
    x=
  

  Say, you want to normalize it in a way that there is exactly one space before and after the equal sign and each line should end with an semicolon. One way to do it:

    findr -g example.txt '(.*?)\s*=\s*(.*?)[\s;]*$' '\1 = \2;' -x
  

  \s stands for any whitespace character. .*? lazily matches any character, until the following expression (here [\s;]*$) matches. The $ sign is necessary to make the second lazy matcher .*? not too lazy. The lazy matcher gives the minimal string possible to fulfill the regular expression--even when this means aborting prematurely. So, we need to take the line end matcher $ into the regexp.

  This changes example.txt to:

    a = 99;
    b = 123;
    var = 44;
    d = 55;
    x = ;