def build_regex_tree(regex: str) -> "RegexTree":
"""
Takes a valid string form regex, and returns the RegexTree that
corresponds to it.
>>> build_regex_tree('0')
Leaf('0')
>>> build_regex_tree('0*')
StarTree(Leaf('0'))
>>> build_regex_tree('(0.1)')
DotTree(Leaf('0'), Leaf('1'))
>>> build_regex_tree('(1|0)')
BarTree(Leaf('1'), Leaf('0'))
>>> build_regex_tree('(0*|1*)')
BarTree(StarTree(Leaf('0')), StarTree(Leaf('1')))
>>> build_regex_tree('((0.1).0)')
DotTree(DotTree(Leaf('0'), Leaf('1')), Leaf('0'))
"""
if len(regex) == 1:
return Leaf(regex)
else:
if regex[-1] == "*":
return StarTree(build_regex_tree(regex[:-1]))
else:
#this then builds either a bar or a star using the regex_split
#function used eariler in is_regex
spl = regex_split(regex)
if spl[1] == "|":
return BarTree(build_regex_tree(spl[0]),
build_regex_tree(spl[2]))
else:
return DotTree(build_regex_tree(spl[0]),
build_regex_tree(spl[2]))
def build_regex_tree(regex: str) -> "RegexTree":
"""Given regex, a valid regular expression regex, build the corresponding
regular expression tree and returns its root.
>>> regex = "(0.1)*"
>>> build_regex_tree(regex)
StarTree(DotTree(Leaf('0'), Leaf('1')))
>>> regex = "(0*|(1.2)*)"
>>> build_regex_tree(regex)
BarTree(StarTree(Leaf('0')), StarTree(DotTree(Leaf('1'), Leaf('2'))))
"""
if regex in ('0', '1', '2', 'e'):
return Leaf(regex)
elif regex[-1] is '*':
return StarTree(build_regex_tree(regex[:-1]))
elif (regex[0] is '(' and regex[-1] is ')' and len(regex) is 5
and ('.' in regex or '|' in regex)):
if '.' in regex:
return DotTree(build_regex_tree(regex[1:2]),
build_regex_tree(regex[3:4]))
elif '|' in regex:
return BarTree(build_regex_tree(regex[1:2]),
build_regex_tree(regex[3:4]))
elif regex[0] is '(' and regex[-1] is ')' and ('.' in regex
or '|' in regex):
if regex[1:-1][0] is '(':
index = index_of_left_closing_bracket(regex[1:-1])
index_of_separator = index + 1
if regex[1:-1][index_of_separator] is '*':
index_of_separator += 1
separator = regex[1:-1][index_of_separator]
elif regex[1:-1][-1] is ')':
index_of_first_open_bracket = regex[1:-1].find('(')
index_of_separator = index_of_first_open_bracket - 1
separator = regex[1:-1][index_of_separator]
else:
if '(' in regex[1:-1]:
index_of_first_open_bracket = regex[1:-1].find('(')
index_of_separator = index_of_first_open_bracket - 1
separator = regex[1:-1][index_of_separator]
else:
index_of_separator = max(regex[1:-1].find('.'),
regex[1:-1].find('|'))
separator = regex[1:-1][index_of_separator]
if separator is '.':
return DotTree(
build_regex_tree(regex[1:-1][:index_of_separator]),
build_regex_tree(regex[1:-1][index_of_separator + 1:]))
elif separator is '|':
return BarTree(
build_regex_tree(regex[1:-1][:index_of_separator]),
build_regex_tree(regex[1:-1][index_of_separator + 1:]))
def build_regex_tree(regex):
if regex in ('0', '1', '2', 'e'):
return Leaf(regex)
elif regex[-1] is '*':
return StarTree(build_regex_tree(regex[:-1]))
elif (regex[0] is '(' and regex[-1] is ')' and len(regex) is 5
and ('.' in regex or '|' in regex)):
if '.' in regex:
return DotTree(build_regex_tree(regex[1:2]),
build_regex_tree(regex[3:4]))
elif '|' in regex:
return BarTree(build_regex_tree(regex[1:2]),
build_regex_tree(regex[3:4]))
elif regex[0] is '(' and regex[-1] is ')' and ('.' in regex
or '|' in regex):
if regex[1:-1][0] is '(':
index = index_of_left_closing_bracket(regex[1:-1])
index_of_separator = index + 1
if regex[1:-1][index_of_separator] is '*':
index_of_separator += 1
separator = regex[1:-1][index_of_separator]
elif regex[1:-1][-1] is ')':
index_of_first_open_bracket = regex[1:-1].find('(')
index_of_separator = index_of_first_open_bracket - 1
separator = regex[1:-1][index_of_separator]
else:
if '(' in regex[1:-1]:
index_of_first_open_bracket = regex[1:-1].find('(')
index_of_separator = index_of_first_open_bracket - 1
separator = regex[1:-1][index_of_separator]
else:
index_of_separator = max(regex[1:-1].find('.'),
regex[1:-1].find('|'))
separator = regex[1:-1][index_of_separator]
if separator is '.':
return DotTree(
build_regex_tree(regex[1:-1][:index_of_separator]),
build_regex_tree(regex[1:-1][index_of_separator + 1:]))
elif separator is '|':
return BarTree(
build_regex_tree(regex[1:-1][:index_of_separator]),
build_regex_tree(regex[1:-1][index_of_separator + 1:]))
def tree_helper(regex_list, ternaries=['0', '1', '2', 'e']):
''' (list) -> RegexTree
Takes in a list form of the regex and creates a
tree recursively
>>> tree_helper(['e'])
Leaf('e')
>>> tree_helper(['1','.','0'])
DotTree(Leaf('1'), Leaf('0'))
>>> tree_helper(['1','|',['2','.','e'],'*'])
BarTree(Leaf('1'), StarTree(DotTree(Leaf('2'), Leaf('e'))))
'''
# base check if there are dots or bars in the
# regex
if '.' in regex_list:
# make a dot node
# everything left of the dot is left, right of
# dot is right
result = DotTree(tree_helper(regex_list[:regex_list.index('.')]),
tree_helper(regex_list[regex_list.index('.') + 1:]))
# check for bars
elif '|' in regex_list:
# make a bar node
# everything left of the bar is left, right of
# bar is right
result = BarTree(tree_helper(regex_list[:regex_list.index('|')]),
tree_helper(regex_list[regex_list.index('|') + 1:]))
# check for stars
elif '*' in regex_list:
# make a star node, everything to the left of it is its child
result = StarTree(tree_helper(regex_list[:regex_list.index('*')]))
# if we've gone through all the dots and bars
else:
# if its a list
if isinstance(regex_list[0], list):
# call the tree helper on that list
result = tree_helper(regex_list[0])
# if its a symbol
elif regex_list[0] in ternaries:
# make a new leaf node
result = Leaf(regex_list[0])
# return the result
return result
def build_regex_tree(regex):
'''(str) -> RegexTree
REQ: str must be a valid regex
Takes in a str that is a valid regex and creates a RegexTree for the regex
and returns the root of it
>>>build_regex_tree('1*')
StarTree(Leaf('1'))
>>>build_regex_tree('(1|0)*')
StarTree(BarTree(Leaf('1'), Leaf('0')))
>>>build_regex_tree('(2*.0)')
DotTree(StarTree(Leaf('2')), Leaf('0'))
>>>build_regex_tree('e*')
StarTree(Leaf('e'))
>>>build_regex_tree('((1.0)|(2.1))')
BarTree(DotTree(Leaf('1'), Leaf('0')), DotTree(Leaf('2'), Leaf('1')))
>>>build_regex_tree("((0|1*)|((2*.1*).(e*.0)))")
BarTree(BarTree(Leaf('0'), StarTree(Leaf('1'))),
DotTree(DotTree(StarTree(Leaf('2')), StarTree(Leaf('1'))),
DotTree(StarTree(Leaf('e')), Leaf('0'))))
'''
# Checks if the len is 0, 1, 2, or e
if len(regex) == 1:
# Calling Leaf function from RegexTree because regex has no children
return Leaf(regex)
# To check if the regex has a '*' at the end
elif regex[-1] == STAR:
# Calling StarTree function from RegexTree because Regex ends with '*'
return StarTree(build_regex_tree(regex[:len(regex)-1]))
else:
# Calling operationLocator helper function to determine what index
# the operation is located at
operation = operationLocator(regex)
# If the main operation is a bar, it will call BarTree from RegexTree
# because the index of the operation is a bar
if regex[operation] == BAR:
return BarTree(build_regex_tree(regex[1:operation]),
build_regex_tree(regex[operation+1:len(regex)-1]))
# This case checks if the operation is a dot, it will call DotTree
# from RegexTree because the index of the operation is a dot
else:
return DotTree(build_regex_tree(regex[1:operation]),
build_regex_tree(regex[operation+1:len(regex)-1]))
def build_regex_tree(regex):
'''(string) -> RegexTree
Builds a corresponding regex tree.
REQ regex is a valid regular expression
'''
if len(regex) == 1:
root = Leaf(regex)
elif regex[-1] == star:
child = build_regex_tree(regex[:-1])
root = StarTree(child)
else:
# remove the brackets!
nregex = regex[1:-1]
r1, r2, symbol = split(nregex)
childr1 = build_regex_tree(r1)
childr2 = build_regex_tree(r2)
if symbol == dot:
root = DotTree(childr1, childr2)
else:
root = BarTree(childr1, childr2)
return root
def build_regex_tree(regex):
'''(str) -> Tree
Given a regex inside a bracket, returns the node which represents that regex
'''
#while loop used for going through the regex
i = 0
#returns a list of list and str
ret = [[], '']
#iterates through the regex
while i < len(regex):
#a close bracket marks the end of the regex
if regex[i] == ')':
return ret
#an open bracket marks the start of the left child
elif regex[i] == '(':
ret[0] += build_regex_tree(regex[i + 1:])
#adds the number of items in the child + the number of brackets(2)
#and an operator(1)
i += len(ret[0][0]) + 3
#adds to the first sublist if it is a valid regex char
elif regex[i] in '012e':
ret[0] += regex[i]
i += 1
#sets type by looking at operator
elif regex[i] in '|.*':
ret[1] = regex[i]
i += 1
#bug, had to remove extra brackets
ret = ret[0]
#converting to Tree
if ret[1] == '|':
ret = BarTree(ret[0][0], ret[0][1])
elif ret[1] == '.':
ret = DotTree(ret[0][0], ret[0][1])
elif ret[1] == '*':
ret = StarTree(ret[0][0])
#returns Tree at the end
return ret
def build_regex_tree(regex):
'''(str) -> RegexTree
takes a string regex, build a regex tree, and returns the root of the tree
>>>build_regex_tree("(1.2)") == RegexTree('.', [Leaf('1'), Leaf('2')])
True
>>>build_regex_tree("(1*|e)") == RegexTree('|', [StarTree(Leaf('1')), Leaf(
'e')])
True
>>>build_regex_tree("(((e.1).2*)|((0***.e).(1|2*)))") == RegexTree('|', [
RegexTree('.', [RegexTree('.', [Leaf('e'), Leaf('1')]), StarTree(Leaf(
'2'))]), RegexTree('.', [RegexTree('.', [StarTree(StarTree(StarTree(Leaf(
'0')))), Leaf('e')]), RegexTree('|', [Leaf('1'), StarTree(Leaf('2'))])])])
True
'''
degree = 0
# base case if length is 1,regex is a leaf since it does not have any child
if len(regex) == 1:
return Leaf(regex)
# if it ends with a "*"
elif regex != "" and regex[-1] == '*':
# set the star to be the parent of a tree built with the rest of regex
return StarTree(build_regex_tree(regex[:-1]))
# if it is in brackets
elif regex != "" and regex[0] == "(" and regex[-1] == ")":
# loop and index of the dot or bar that's in the first brackets
for i in range(len(regex)):
if regex[i] == "(":
degree += 1
elif regex[i] == ")":
degree -= 1
if degree == 1 and (regex[i] == "." or regex[i] == '|'):
I = i
# recursion: remove the two brackets
# divide to 2 parts at the dot or bar found and build trees
# for each of the parts
return RegexTree(regex[I], [build_regex_tree(
regex[1:I]), build_regex_tree(regex[I+1:-1])])
def build_regex_tree(regex):
'''(str) -> RegexTree
Builds the RegexTree as specified by the valid regex parameter. Return
this tree's root
REQ: regex is a valid regex
'''
# Length 1 regex
if len(regex) == 1:
return Leaf(regex)
# '*' regex
elif regex[-1] == '*':
return StarTree(build_regex_tree(regex[:-1]))
# '.' or '|' regex
else:
# Find the '' or '|' associated with the outer brackets and build
# two children rooted to the '.' or '|' tree
# Traverse by index. Upon finding '.' or '|' split the string at that
# point and call the init with the two parts as parameters,
# excluding the most outer brackets. If '(' is found, continue until
# all '(' on the left side are found and an equal number of ')' are
# found; the '.' or '|' afterwards is the associated operation.
lr_bracket_balance = 0
count = 1
while count < len(regex) - 1:
if regex[count] == '.' and lr_bracket_balance == 0:
return DotTree(build_regex_tree(regex[1:count]),
build_regex_tree(regex[count + 1: -1]))
elif regex[count] == '|' and lr_bracket_balance == 0:
return BarTree(build_regex_tree(regex[1:count]),
build_regex_tree(regex[count + 1: -1]))
elif regex[count] == '(':
lr_bracket_balance += 1
elif regex[count] == ')':
lr_bracket_balance -= 1
count += 1
def build_regex_tree(regex):
"""
(str) -> object
This function takes in a regex and returns the root of the
tree node. It returns None if the regex is not valid.
>>>build_regex_tree("0")
Leaf('0')
>>>build_regex_tree("1")
Leaf('1')
>>>build_regex_tree("2")
Leaf('2')
>>>build_regex_tree("e")
Leaf('e')
>>>build_regex_tree("1*")
StarTree(Leaf('1'))
>>>build_regex_tree("(1|2)")
BarTree(Leaf('1'), Leaf('2'))
>>>build_regex_tree("(e.2)")
DotTree(Leaf('e'), Leaf('2'))
>>>build_regex_tree("(1|(0.2)*)")
BarTree(Leaf('1'), StarTree(DotTree(Leaf('0'), Leaf('2'))))
>>>build_regex_tree("(()")
None
>>>build_regex_tree("(1.(e*|(1*|2)))")
DotTree(Leaf('1'), BarTree(StarTree(Leaf('e')), BarTree(StarTree(Leaf('1')), Leaf('2'))))
"""
# If the regex is not correct return None
if (not (is_regex(regex))):
return None
# If the length is one then return the leaf of the particular SINGLE_CHAR
if (len(regex) == 1):
return Leaf(regex)
# If the last char is a Star, Create a star tree
elif (regex[-1] == '*'):
# Make the child equal to the rest of the regex
return StarTree(build_regex_tree(regex[:-1]))
# If regex has a bracket
else:
# Check if the expression in the bracket requires a Binary Tree
try:
# Get the left expression, operation and the right expression
(left, operation, right) = splicer(regex)
# If error raised check if it was a StarTree or a Leaf
except Exception:
# Remove Brackets
return build_regex_tree(regex[1:-1])
# Get the correct corresponding tree object
tree = BINARY[operation]
# Build the tree, also make the left and right children into trees
return tree(build_regex_tree(left), build_regex_tree(right))
def build_regex_tree(regex):
'''(str) -> Dot/Bar/Star/Leaf Tree
REQ: regex must be valid
This function will take the given valid regex string, then it will build
the appropriate tree and finally it will return its root
>>>build_regex_tree('1***')
StarTree(StarTree(StarTree(Leaf('1'))))
>>>build_regex_tree('(1|2)')
BarTree(Leaf('1'), Leaf('2'))
>>>build_regex_tree('(0.e)')
DotTree(Leaf('0'), Leaf('e'))
>>>build_regex_tree('(0.1)')
DotTree(Leaf('0'), Leaf('1'))
>>>build_regex_tree('((0.1).2)')
DotTree(DotTree(Leaf('0'), Leaf('1')), Leaf('2'))
>>>build_regex_tree('((0.1)|2)')
BarTree(DotTree(Leaf('0'), Leaf('1')), Leaf('2'))
>>>build_regex_tree('((0.1)|2*)')
BarTree(DotTree(Leaf('0'), Leaf('1')), StarTree(Leaf('2')))
>>>build_regex_tree('((0.1)*|2*)')
BarTree(StarTree(DotTree(Leaf('0'), Leaf('1'))), StarTree(Leaf('2')))
>>>build_regex_tree('((0.1)*|2*)*')
StarTree(BarTree(StarTree(DotTree(Leaf('0'), Leaf('1'))),
StarTree(Leaf('2'))))
>>>build_regex_tree("((0.1)|(2.1))")
BarTree(DotTree(Leaf('0'), Leaf('1')), DotTree(Leaf('2'), Leaf('1')))
>>>build_regex_tree("((0.1).(2.1))")
DotTree(DotTree(Leaf('0'), Leaf('1')), DotTree(Leaf('2'), Leaf('1')))
>>>build_regex_tree("((0|1).(2.1))")
DotTree(BarTree(Leaf('0'), Leaf('1')), DotTree(Leaf('2'), Leaf('1')))
>>>build_regex_tree("((0|1).(2.1))*")
StarTree(DotTree(BarTree(Leaf('0'), Leaf('1')), DotTree(Leaf('2'),
Leaf('1'))))
>>>build_regex_tree('(0.(1.(0|2)*))')
DotTree(Leaf('0'), DotTree(Leaf('1'), StarTree(BarTree(Leaf('0'),
Leaf('2')))))
>>>build_regex_tree('(((0***.1)|2)|2***)')
BarTree(BarTree(DotTree(StarTree(StarTree(StarTree(Leaf('0')))),
Leaf('1')), Leaf('2')), StarTree(StarTree(StarTree(Leaf('2')))))
>>>build_regex_tree("(1***.(0|(2.(1|e)*)**)***)*")
StarTree(DotTree(StarTree(StarTree(StarTree(Leaf('1')))),
StarTree(StarTree(StarTree(BarTree(Leaf('0'),
StarTree(StarTree(DotTree(Leaf('2'), StarTree(BarTree(Leaf('1'),
Leaf('e'))))))))))))
'''
# If we have a single letter then it has no children hence it will be
# a leaf
if len(regex) == 1:
return Leaf(regex)
# Otherwise we either have to create a Dot/Bar/Star Tree
else:
# If the last character is a star then we have to create a star tree
if regex[-1] is star:
return StarTree(build_regex_tree(regex[:-1]))
else:
# Remove the brackets
regex = regex[1:-1]
# If we have a left sided regex that means we have to divide the
# string from the right hand side
if regex[0] is left and regex[-1] is not right:
# Find the position of the dot and the bar
find_dot = regex.rfind(dot)
find_bar = regex.rfind(bar)
# If the dot is greater in index then partition it at the dot
if find_dot > find_bar:
s = regex.rpartition(dot)
# Since we partition it at the dot, build a DotTree
return (DotTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
else:
# Else build the BarTree
s = regex.rpartition(bar)
return (BarTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
# If we have a right sided regex that means we have to divide the
# string from the left hand side
elif regex[-1] is right and regex[0] is not left:
# Find the position of the dot and the bar
find_dot = regex.find(dot)
find_bar = regex.find(bar)
# If the dot is lower in index then partition it at the dot
if ((find_dot is not -1 and find_dot < find_bar) or
(find_bar is -1)):
s = regex.partition(dot)
# Since we partition it at the dot, build a DotTree
return (DotTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
else:
# Else build the BarTree
s = regex.partition(bar)
return (BarTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
# If we have an even regex
elif regex[0] is left and regex[-1] is right:
# Set the counter
i = 0
# Loop through the regex to find the position of either the
# bar of the dot
while i < len(regex):
if regex[i] is right and regex[i+2] is left:
index = i+1
# Once we get the index set i so we can exit loop
i = len(regex)
i += 1
# If there is bar at the given index then build a BarTree
if regex[index] is bar:
return (BarTree(build_regex_tree(regex[:index]),
build_regex_tree(regex[index+1:])))
# Otherwise just build a DotTree
else:
return (DotTree(build_regex_tree(regex[:index]),
build_regex_tree(regex[index+1:])))
# If either side do not contain a bracket
elif regex[0] is not left and regex[-1] is not right:
# Then check if the last char is a star
if regex[-1] is star:
# Find the dot and the bar since we will be spliting it
# from the left side
find_dot = regex.find(dot)
find_bar = regex.find(bar)
# If dot has lower index then bar
if ((find_dot is not -1 and find_dot < find_bar) or
(find_bar is -1)):
s = regex.partition(dot)
# Build a DotTree
return (DotTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
# Otherwise build a BarTree
else:
s = regex.partition(bar)
return (BarTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
# If the regex is of form (r1*.r2) or (r1*|r2)
elif regex[1] is star:
# Find the dot and the bar since we will be spliting it
# from the right side
find_dot = regex.rfind(dot)
find_bar = regex.rfind(bar)
# If the dot has higher index then bar
if find_dot > find_bar:
s = regex.rpartition(dot)
# Build the DotTree
return (DotTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
else:
s = regex.rpartition(bar)
# Otherwise build the BarTree
return (BarTree(build_regex_tree(s[0]),
build_regex_tree(s[2])))
# Otherwise if we have regex of form (r1|r2)
elif bar in regex:
# Build a BarTree
return (BarTree(build_regex_tree(regex[0]),
build_regex_tree(regex[2])))
# Otherwise we have regex of form (r1.r2)
elif dot in regex:
# Build a DotTree
return (DotTree(build_regex_tree(regex[0]),
build_regex_tree(regex[2])))
else:
# Look for the dot
if dot in regex:
# Seperate the string at the dot so we can recurse on r1
# and r2
s = regex.partition(dot)
# Create the Dot Tree
return DotTree(Leaf(s[0]), Leaf(s[2]))
# Look for the bar
elif bar in regex:
# Seperate the string at the bar so we can recurse on r1
# and r2
s = regex.partition(bar)
# Create the Bar Tree
return BarTree(Leaf(s[0]), Leaf(s[2]))
def build_regex_tree(regex):
''' (str) -> set(str)
The function takes a valid regex string and produces a regex tree in order
to return the root.
REQ: String must be a valid regex
>>> build_regex_tree("((1.(0|1)*).2)")
DotTree(DotTree(Leaf('1'), StarTree(BarTree(Leaf('0'), Leaf('1')))),
Leaf('2'))
>>> build_regex_tree("(1|2)")
BarTree(Leaf('1'), Leaf('2'))
>>> build_regex_tree("(1.2)")
DotTree(Leaf('1'), Leaf('2'))
>>>build_regex_tree("1****")
StarTree(StarTree(StarTree(StarTree(Leaf('1')))))
'''
# Base Case
if len(regex) <= 1:
return Leaf(regex)
# If Regex obtains a star in as the last element
elif regex[-1] == "*":
# Return the StarTree with the rest of the regex
# without the last element
return StarTree(build_regex_tree(regex[:-1]))
# If regex is in the form of ( + r1 + |/. + r2 + )
else:
# Create counter variables
break_location = 0
bracket_count = 0
length = len(regex)
# Find the location to split the regex into r1 and r2 components
# Search the elements in range of the regex
# regex[1:-1] Accommodates for the outter brackets
for element in range(len(regex[1:-1])):
# If element is a left bracket
if regex[1:-1][element] == bracket[0]:
bracket_count += 1
# If element is a right bracket
elif regex[1:-1][element] == bracket[1]:
bracket_count -= 1
# If there is a set of brackets followed by a |.
# Is the location where to split for components
elif (bracket_count == 0 and regex[1:-1][element] in duo_regex):
# accomadate for the outer bracket
break_location += element + 1
# If the node found after the set of brackets is a "|"
if regex[break_location] == duo_regex[0]:
# Return BarTree on the left side "r1" and the right side "r2"
return BarTree(
build_regex_tree(regex[1:break_location]),
build_regex_tree(regex[break_location + 1:length - 1]))
# If the node found after the set of brackets is a "."
elif regex[break_location] == duo_regex[1]:
# Return DotTree on the left side "r1" and the right side "r2"
return DotTree(
build_regex_tree(regex[1:break_location]),
build_regex_tree(regex[break_location + 1:length - 1]))
def build_regex_tree(regex):
'''
(str) -> RegexTree
This function takes a valid regex and build a tree correspoinding to its
given permutation.
REQ: regex must be string.
Example:
>>> build_regex_tree('(0*|1*)')
BarTree(StarTree(Leaf('0')), StarTree(Leaf('1')))
>>> build_regex_tree('((0.1).0)')
DotTree(DotTree(Leaf('0'), Leaf('1')), Leaf('0'))
>>> build_regex_tree('((1.(0|1)*).0)')
DotTree(DotTree(Leaf('1'), StarTree(BarTree(Leaf('0'), Leaf('1')))),
Leaf('0'))
'''
if len(regex) == 1:
if regex in '012e':
return Leaf(regex)
if regex[-1] == '*':
return StarTree(build_regex_tree(regex[:-1]))
if regex[0] == '(' and regex[-1] == ')':
num_left = 0
num_right = 0
o_index = 0
bar = 0
dot = 0
for i in range(1, len(regex)):
if regex[i] == '(':
num_left += 1
if regex[i] == ')':
num_right += 1
if num_left == num_right and num_left != 0 and i < len(regex):
if i > len(regex) - 4:
o_index = regex[1:].index('(')
elif regex[i + 1] == '*':
# the next char that is not a '*' must be a operator
for s in range(i + 2, len(regex)):
if regex[s] != '*':
o_index = s
break
else:
o_index = i + 1
break
if o_index == 0:
if '|' in regex:
bar = regex.index('|')
if '.' in regex:
dot = regex.index('.')
if not (bar == 0 and dot == 0) or (bar == 0 and dot == 0):
o_index = max(bar, dot)
# check left side and right side
if regex[o_index] == '.':
return DotTree(build_regex_tree(regex[1:o_index]),
build_regex_tree(regex[o_index + 1:-1]))
else:
return BarTree(build_regex_tree(regex[1:o_index]),
build_regex_tree(regex[o_index + 1:-1]))
def build_regex_tree(valid_r):
''' (str) -> RegexTree or subclass of RegexTree
Return the root of the tree which has been tranformed from the valid regex
expression valid_r.
REQ: valid_r must be valid regex expression.
>>> build_regex_tree('(1|(0*.e))')
BarTree(Leaf('1'), DotTree(StarTree(Leaf('0')), Leaf('e')))
>>> build_regex_tree('0***')
StarTree(StarTree(StarTree(Leaf('0'))))
'''
# base case
# if valid_r is the base regex which the length is one.
# the form of 'base_regex'
if valid_r in base_regex:
root = Leaf(valid_r)
return root
# if the valid_r is the base regex with star(s).
# the form of 'base_regex + star(s)'
elif valid_r[0] in base_regex:
# if there is with only one star
l1 = Leaf(valid_r[0])
root = StarTree(l1)
# if there are more than one stars
if len(valid_r) > 2:
for i in range(len(valid_r) - 2):
root = StarTree(root)
return root
# the form of '(regex)'
elif valid_r[0] == '(' and valid_r[-1] == ')':
# if valid_r takes bar operation with two leaves,
# the form of (base_regex + star(s) operation base_regex)
if valid_r[1] in base_regex and valid_r[-2] in base_regex:
r_c1 = Leaf(valid_r[-2])
l_c_r1 = build_regex_tree(valid_r[1:-3])
opr1 = valid_r[-3]
if opr1 == '.':
root = DotTree(l_c_r1, r_c1)
elif opr1 == '|':
root = BarTree(l_c_r1, r_c1)
return root
# if the left side of the bar is not in base regex and the right side
# of the bar is in base regex,
# the form of ((regex) + star(s) operation base_regex)
elif valid_r[1] == '(' and valid_r[-2] in base_regex:
r_c2 = Leaf(valid_r[-2])
l_c_r2 = build_regex_tree(valid_r[1:-3])
opr2 = valid_r[-3]
if opr2 == '.':
root = DotTree(l_c_r2, r_c2)
elif opr2 == '|':
root = BarTree(l_cr2, r_c2)
return root
# the form of ((regex) + star(s) operation (regex)/base_regex (star(s))
elif valid_r[1] == '(' and valid_r[-2] == star:
# find the operation and base regex on the left side
ind = 0
found = False
while ind < len(valid_r) - 1 and not found:
if valid_r[-ind - 2] in base_regex or valid_r[-ind - 2] == ')':
rr_index = -ind - 2
found = True
ind += 1
# the form of ((regex) operation(s) base_regex (star(s)))
if valid_r[rr_index] in base_regex:
r_c3 = build_regex_tree(valid_r[rr_index:-1])
l_c_r3 = build_regex_tree(valid_r[1:rr_index - 1])
opr3 = valid_r[rr_index - 1]
if opr3 == '.':
root = DotTree(l_c_r3, r_c3)
elif opr3 == '|':
root = BarTree(l_c_r3, r_c3)
return root
# the form of ((regex) operation(s) (r) + (star(s)))
if valid_r[rr_index] == ')':
seperated_list = seperate_parenthesis(valid_r[1:-1])
l_c_r4 = build_regex_tree(seperate_list[0] +
[seperate_list[1][:-1]])
r_c_r4 = build_regex_tree(seperate_list[2] + seperate_list[3])
opr4 = seperate_list[1][-1]
if opr4 == '.':
root = DotTree(l_c_r4, r_c_r4)
elif opr4 == '|':
root = BarTree(l_c_r4, r_c_r4)
return root
# if the right side of the bar is not in base regex and the left side
# of the bar is in base_regex.
# the form of (base_r operation(s) (r))
elif valid_r[1] in base_regex and valid_r[-2] == ')':
# find the parenthesis on the right side of the non star operation.
j = 2
found_right = False
while j < len(valid_r) - 2 and not found_right:
if valid_r[j] == '(':
found_right = True
right_ind = j
j += 1
l_c_r5 = build_regex_tree(valid_r[1:right_ind - 1])
r_c_r5 = build_regex_tree(valid_r[right_ind:-1])
opr5 = valid_r[right_ind - 1]
if opr5 == '.':
root = DotTree(l_c_r5, r_c_r5)
elif opr5 == '|':
root = BarTree(l_c_r5, r_c_r5)
return root
# the form of (base_r + operation(s) + (r)/base_r + star(s))
elif valid_r[1] in base_regex and valid_r[-2] == star:
k = 2
found_r = False
while k < len(valid_r) - 2 and not found_r:
if valid_r[k] in non_star_oprs:
opr6_ind = k
found_r = True
k += 1
opr6 = valid_r[opr6_ind]
right_r = valid_r[opr6_ind + 1]
l_c6 = build_regex_tree(valid_r[1:opr6_ind])
r_c6 = build_regex_tree(valid_r[opr6_ind + 1:-1])
if opr6 == '.':
root = DotTree(l_c6, r_c6)
elif opr6 == '|':
root = BarTree(l_c6, r_c6)
return root
# the form of ((r) + operations + (r))
elif valid_r[1] == '(' and valid_r[-2] == ')':
s_list = seperate_parenthesis(valid_r[1:-1])
opr7 = s_list[1][-1]
l_c_r7 = build_regex_tree(s_list[0])
r_c_r7 = build_regex_tree(s_list[2])
if opr7 == '.':
root = DotTree(l_c_r7, r_c_r7)
elif opr7 == '|':
root = BarTree(l_c_r7, r_c_r7)
return root
# the form of (r)*
elif valid_r[0] == '(' and valid_r[-1] == star:
for i in range(len(valid_r)):
if valid_r[-i - 1] == ')':
r_ind = -i - 1
c = build_regex_tree(valid_r[0:r_ind + 1])
root = StarTree(c)
if len(valid_r[r_ind:]) > 2:
for i in range(len(valid_r[r_ind:]) - 2):
root = StarTree(root)
return root
root = StarTree(c)
if len(valid_r[r_ind:]) > 2:
for i in range(len(valid_r[r_ind:]) - 2):
root = StarTree(root)
return root
if __name__ == '__main__':
br1 = RegexTree('0', [])
br2 = RegexTree('1', [])
br3 = RegexTree('2', [])
br4 = RegexTree('e', [])
dt = DotTree(br1, br2)
bt = BarTree(dt, br2)
dt2 = DotTree(br1, br2)
st = StarTree(dt)
print('THEY MATCH -> ' + str(regex_match(st, '1111')))
print('THEY MATCH -> ' + str(regex_match(dt, '01')))
print('THEY MATCH -> ' + str(regex_match(bt, '01')))
print('THEY MATCH -> ' + str(regex_match(dt2, '01')))
print('THEY MATCH -> ' + str(regex_match(bt, '1')))
#print('perm'+ str(all_regex_permutations('((1|0*).0)')))
a = build_regex_tree('1')
print(a)
b = build_regex_tree('((1.0)|(0*.1))')
print(b)
c = build_regex_tree('(1*.0)')
print(c)
print(c.get_left_child().get_child().get_symbol())
d = build_regex_tree('(1*|0***)')
print(d)