A rewrite of cowsay in python. Allows for parsing of existing .cow files.

pip install python-cowsay

The classic cowsay can be generated by the cowsay or cowthink functions:

from cowsay import cowsay

message = """
The most remarkable thing about my mother is that for thirty years she served
the family nothing but leftovers.  The original meal has never been found.
		-- Calvin Trillin
""".strip()
print(cowsay(message))

Will yield:

 __________________________________________ 
/ The most remarkable thing about my       \
| mother is that for thirty years she      |
| served the family nothing but leftovers. |
| The original meal has never been found.  |
|                                          |
\ -- Calvin Trillin                        /
 ------------------------------------------ 
        \   ^__^
         \  (oo)\_______
            (__)\       )\/\
                ||----w |
                ||     ||

The parameters for these functions are:

• message – a string to wrap in the text bubble
• cow='default' – the name of the cow (valid names from list_cows)
• preset=None – the original cowsay presets: -bgpstwy
• eyes=Option.eyes – A custom eye string
• tongue=Option.tongue – A custom tongue string
• width=40 – The width of the text bubble
• wrap_text=True – Whether text should be wrapped in the bubble
• cowfile=None – A custom string representing a cow

The available builtin cows can be found with list_cows. A cow can be chosen randomly from this list with get_random_cow.

A custom .cow file can be parsed using the read_dot_cow function which takes a TextIO stream. I.e., You can either create a TextIO from a string or read a file.

The read_dot_cow will look for the first heredoc in the steam and extract the heredoc contents. If no heredoc exists, the whole stream is used instead. Escape characters are then escaped. The default escape characters can be changed by passing in an optional escape dictionary parameter mapping escape codes to their chars.

For example:

from io import StringIO

from cowsay import read_dot_cow, cowthink

cow = read_dot_cow(StringIO("""
$the_cow = <<EOC;
         $thoughts
          $thoughts
           ___
          (o o)
         (  V  )
        /--m-m-
EOC
"""))
message = """
Nothing is illegal if one hundred businessmen decide to do it.
        -- Andrew Young
""".strip()
print(cowthink(message, cowfile=cow))

Will yield:

 ___________________________________ 
( Nothing is illegal if one hundred )
( businessmen decide to do it.      )
(                                   )
( -- Andrew Young                   )
 ----------------------------------- 
         o
          o
           ___
          (o o)
         (  V  )
        /--m-m-

A bit of a hack at the moment, but if any full-width characters are found in the message string, all characters in the thought bubble are converted to full-width. For example:

　＿＿＿＿＿＿＿＿＿＿＿＿　
（　喵喵喵。我是一只猫。　）
　－－－－－－－－－－－－　
   o
    o

     |\_/|
     |o o|__
     --*--__\
     C_C_(___)

This works fine when all characters in the message are full-width, but does not work so well when there is a mix of full- and neutral-width characters:

　＿＿＿＿＿＿＿＿＿＿＿＿＿＿＿＿＿　
（　喵喵喵。Ｉ　ａｍ　ａ　ｃａｔ．　）
　－－－－－－－－－－－－－－－－－　
   o
    o

     |\_/|
     |o o|__
     --*--__\
     C_C_(___)

Each full-width character still only counts as one character when setting the text width. For example:

from cowsay import cowthink

message = "喵喵喵。我是一只猫。我想吃鱼和喝牛奶。"
print(cowthink(message, cow="kitten", width=10))

Will yield:

　＿＿＿＿＿＿＿＿＿＿＿＿　
（　喵喵喵。我是一只猫。　）
（　我想吃鱼和喝牛奶。　　）
　－－－－－－－－－－－－　
   o
    o

     |\_/|
     |o o|__
     --*--__\
     C_C_(___)