noscripter / atmark

Awk+Sed for humans

The Atmark

The Atmark -- Awk+sed for humans.

Do this: :

# Rename a files in current directory (file-name -> file_name.jpg)
$ ls | @ sub - _ "mv # @.jpg" | sh

Except this: :

# Rename a files in current directory (file-name -> file_name.jpg)
$ ls | awk '{print $1 $1}' | sed s/"-"/"_"/ | awk '{print "mv", $2, $1, ".jpg"}' | sh

More deep: :

$ ls | @  sub  -  _  "mv # @.jpg" | sh
          ---  |  |  ------------
           |   |  |      \_ format string (# - link on first state,
           |   |  |                        @ - link on current state (after replace))
           |   |  |
           |   |   \_ second replace param (to replace)
           |   |
           |    \_ first replace param (what replace)
           |
            \_ function name (substitute)

More examples:

Change file's extension .html > .php :

# Atmark
$ ls | @ split . head "mv # @.php"

# Awk/Sed
$ ls | awk '{printf "mv "$0; sub(/html/,"php"); print " "$0}' | sh

Print all but the first three columns :

# Atmark (\t means tab)
$ ls -la | @ split_ drop 3 join \t

# Awk/Sed
$ ls -la | awk '{for(i=1;i<4;i++) $i="";print}'

Kill process by name :

# Atmark
$ ps aux | @ grep sysmond$ index 2 "kill @" | sh 

# Awk/Sed
$ ps aux | grep [s]ysmond | awk '{print "kill "$2}' | sh

And more, more, more.

Docs are available at https://atmark.readthedocs.org/. Pull requests with documentation enhancements and/or fixes are awesome and most welcome.

Requirements

• python >= 2.6

Installation

The Atmark should be installed using pip: :

pip install atmark

Bash completion

Atmark supports bash completion. Just add this lines to your `.bashrc`: :

_atmark_complete() {
    COMPREPLY=( $( COMP_WORDS="${COMP_WORDS[*]}" \\
                COMP_CWORD=$COMP_CWORD \\
                _ATMARK_COMPLETE=complete $1 ) )
    return 0
}

complete -F _atmark_complete -o default @ @@;

You can easy do it with command: :

@ -bs >> ~/.bashrc

Usage

Get help

$ @ -h

Atmark (@) -- is a command line utility for parsing text input and generating output.

You can pipe data within a Atmark (@) statement using standard unix style pipes ("|").
Provide for Atmark function composition and let them work for you.

Example. Replace "_" with "-" in files in current dir and change the files extensions to jpg:

    $ ls | @ replace _ -  split . "mv # @.jpg"

It is mean:

    $ ls > replace($LINE, "_", "-") > split($RESULT, ".") > format($RESULT, "mv $LINE $RESULT.jpg")

You can use "@ --debug ARGS" for debug Armark commands.

===================================================================================
LIST OF THE BUILT IN FUNCTIONS


format FORMAT_STRING     -- format and print a string.

    Symbol '@' in FORMAT_STRING represents the current value in process of composition of fuctions.
    Symbol '#' in FORMAT_STRING represents the history state.
        Where   # or #0 -- first state, #<n> (#1, #2) -- state with number n

    Synonyms: You can drop `format` function name. This lines are equalent:

        $ ls | @ upper format "@.BAK"
        $ ls | @ upper "@.BAK"

capitalize/cap   -- capitalize the string.

drop N       -- drop N elements from list/string.

equal/== PATTERN     -- return None if arg is not equal to PATTERN.

filter/if        -- filter results by value has length

head/h       -- extract the first element/character of a list/string

index/ix N           -- get the N-th element/character from list/string.

join/j SEPARATOR     -- concatenate a list/string with intervening occurrences of SEPARATOR

join_/j_         -- same as join but SEPARATOR set as ' '

kill REGEXP      -- replace in a string/list REGEXP to ''.

last         -- get last element/character of incoming list/string.

length/len       -- return length of list/string.

lower/l          -- make the string is lowercase

map FUNCTION     -- apply the following function to each element/character in list/string.

nocolor/nc       -- Remove ansi colors from string.

notequal/!= PATTERN  -- return None if arg is equal to PATTERN.

notgrep/ng REGEXP    -- filter results by REGEXP. Leave ungrepped

replace/r/sub FROM TO -- replace in a string/list FROM to TO.

reverse/rev      -- reverse list/string.

rstrip/rs/rtrim PATTERN -- return the string with trailing PATTERN removed.

grep/g REGEXP    -- filter results by REGEXP

sort         -- sort list/string.

split/sp SEPARATOR   -- return a list of the substrings of the string splited by SEPARATOR

split_/sp_       -- same as split by splited a string by whitespace characters

strip/s/trim PATTERN -- return the string with leading and trailing PATTERN removed.

strip_/s_/trim_      -- same as strip but trims a whitespaces.

tail/t       -- extract the elements after the head of a list

take N       -- take N elements from list/string.

upper/u          -- make the string is uppercase.

Bug tracker

If you have any suggestions, bug reports or annoyances please report them to the issue tracker at https://github.com/klen/atmark/issues

Contributing

Development of starter happens at github: https://github.com/klen/atmark

Contributors

• klen (Kirill Klenov)

License

Licensed under a BSD license.