Python text randomizer

Generate random text based on template

Example

from text_randomizer import TextRandomizer

template = '{Brown|Red} fox {jump over|dig under} [lazy|crazy] dog'
text_rnd = TextRandomizer(template)
print(text_rnd.get_text())

You will get random text like 'Red fox dig under crazy lazy dog'.

Template string

Template is a usual short or long text with special randomize commands:

• 'Synonyms' command - {variant1 | variant2 | variant3} - insert one of the variants to the result string
• If you want to omit text - use 'empty' variant - {|variant}
• Mixin command = [ text 1 | text 2 | text 3] - will mix this variants randomly
• You can use separator in mixin - [+,+text 1|text2 ] - you will get text2,text1. Separator can by any symbol or set of symbols: [+==+ a|b] - a==b or b==a
• If you want to get spesial symbol in your result ('{', '}', '[', ']', '|', '+') - use backslash for it - {, }, [, ], |, +
• All this commands can be mixed and nested in all combinations: 'start {aa|bb|{cc1|cc2}} or [a1|{word1|word2}|a3| [aa1|aa2|aa3]]'
• You can use special predefined random functions in templates - {random integer = $RANDINT(1,10), uuid = $UUID, now = $NOW(%Y-%M-%d)}. Result will be = 'random integer = 4, uuid = 8ae6bdf4-d321-40f6-8c3c-81d20b158acb, now = 2017-08-01' You can define your own randomization functions and use it in templates. Read about functions further.

API and more examples

Pass template string to TextRandomizer constructor. It will be parsed at once. After creating randomizer object, you can use its get_text() method.

This behaviour can be changed - if you pass parse=False parameter to constructor: TextRandomizer(templ_str, parse=False) - template will not be parsed before you run TextRandomizer.parse() method. Only after this method you can use get_text() method. This can be usefull is several cases. Lets see some examples.

from text_randomizer import TextRandomizer

# parse template at once
text_rnd1 = TextRandomizer('randomize {me|it|him|her} please')
for i in range(0,10):
    print(text_rnd1.get_text())

# Do not parse template. Parse when needed. This can save you time and memory
text_rnd2 = TextRandomizer('randomize {me|it|him|her} please $YOUR_FUNC(5)', parse=False)
# you can register your own randomization functions before parsing
# run parse() before get_text() call. Otherwise you will get exception
text_rnd2.parse()
for i in range(0,10):
    print(text_rnd2.get_text())

You can get number of all possible variants by calling TextRandomizer.variants_number() method. But if you use randomize functions - this value will be imprecise.

For example - 'start {word1 | word2} end' - will give you 2 variants. And 'start {word1 | word2} $UUID end' will give you infinite variants.

Next example shows how to register your own function. Do it before parse() call

text_rnd = TextRandomizer('$YOUR_FUNC1(5, 10) and $FUNC2(param1, param2)', parse=False)

# use simple callable for function
text_rnd.add_function('FUNC2', lambda p1, p2: '{} and {}'.format(p1, p2))

# use dictionary for more complicated cases.
text_rnd.add_function('YOUR_FUNC1', {'callable': lambda x, y: randint(x,y), 'coerce': int})

text_rnd.parse()
print(text_rnd.get_text())