fakr reads a Jinja2 Template from stdin, renders it using a vocabulary and writes the result to stdout. It's as simple as that...

$ pip install fakr

$ fakr --help

Generate a simple csv file with 100k rows

$ echo "{{row}},{{row%100}},{{firstname}},{{lastname}},{{email}}" \
  | fakr --count 100000``

Use a file for complex template

$ cat examples/templates/vcard.tpl \
  | fakr --count 1000

Make a http request using curl:

$ echo "company={{company|urlencode}}&city={{city|urlencode}}"  \
  | fakr -c1 \
  | curl httpbin.org/post -d @-

Write data to a redis server:

$ echo 'firstname \"{{firstname}}\" lastname \"{{lastname}}\" email \"{{email}}\"' \
  | fakr \
  | xargs -i redis-cli HMSET {}``

The templates you use for data generation are plain Jinja2 Templates. See their reference for detailed information.

There are a few custom filters, functions and variables for use with fakr:

Custom filters:

• ascii: Converts the value to ascii (using Unidecode) (i. e. {{lastname|ascii}})
• shuffle: Shuffles the value randomly (i. e. {{lastname|shuffle}})
• chance: Gives the value a chance from 0.0 to 1.0 to be returned (i. e. {{firstname|chance(0.9)}} - firstname will be 90% returned, 10% empty)
• rjust: Right-justifies the value to the given with (i. e. {{company|rjust(40)}})
• ljust: Left-justifies the value to the given with (i. e. {{company|ljust(40)}})
• center: Centers the value in width (i. e. {{company|center(40)}})
• hash: Calculates the hash with a given hashing algo (i. e. {{email|hash('md5')}})

Custom functions:

• translate
• uuid4: Returns a new UUIDv4 on every call (i. e. {{uuid4()}})
• unixtime: Returns a the current unixtime as float in seconds, (i. e. {{unixtime()}})

Fixed (vocabulary independent) variables:

• row: The row (starting from 0) of the current dataset
• id: The id or sequence number of the current dataset
• guid: The representation of the id in a uuid-like fashion