Simple string filtering algorithm

filter.py

# During data processing you will face a situation when you will
# need to filter out some data based on string comparison (e.g. process
# only file names which pass some condition). Things get more complicated
# if the condition is meant to be provided by user or as a CLI argument.
#
# Some of the popular methods are following:
#  - wildcard matching (e.g. "*.jpg" which will return only jpeg files)
#  - regex (e.g. ".*\.jpg")
#  - code (e.g. filename.substr(-4) == ".jpg")
#
# Wildcard matching is easy and efficient, but not suitable for disjunction
# (e.g. match all image types), regex is very flexible but difficult to write,
# custom code can handle any situation but not possible to pass as CLI
# argument.
#
# With all these pros and cons in mind I have decided to design this 
# simple yet flexible filtering algorithm. Filtering expression is 
# composed of keywords, symbols and variables - you can use parentheses 
# to form more complex expressions with logic operators (And, Or, Not)
# and anything else is considered as a variable.
# All variables will be substituted with True or False depending whether
# they are present in the tested string or not
#
# For example "sunny and not windy" will match all strings which contain
# word "sunny", but do not contain "windy". In genereal we can pass any
# python logic expression where all variables will turn True if their
# names is present in input string

def check(filter, tags):
  test = []
  filter = filter.replace("(", " ( ")
  filter = filter.replace(")", " ) ")
  filter = filter.replace("  ", " ")
  current = tags.split(" ")

  for token in filter.split(" "):
    if token in ["", "True", "False", "or", "and", "not", "(", ")"]:
      test.append(token)
    elif token in current:
      test.append("True")
    else:
      test.append("False")
     
  return eval(" ".join(test))

# simple python expression matcher
print(check("day", "sunny day")) # True
print(check("night", "sunny day")) # False
print(check("sunny or cloudy", "sunny day")) # True
print(check("rainy or windy", "sunny day")) # False
print(check("sunny and not windy", "sunny day")) # True