A simple adaptive parser in Python

adaptive_parser.py

import copy
import sys
import subprocess
from tokenize import tokenize
from io import BytesIO

def tokenize_input(the_input):
	return ([a.string for a in list(tokenize(BytesIO(the_input.encode('utf-8')).readline)) if a.string not in ["\n","","utf-8"]])

def matches_pattern(string,pattern):
	if len(string) != len(pattern):
		return False
	for i in range(0,len(string)):
		if is_expr_(pattern[i]):
			for j in range(0,len(string)):
				if pattern[i] == pattern[j] and string[i] != string[j]:
					return False
		elif string[i] != pattern[i]:
			return False
	return True

def rewrite_rule(input_string,input_pattern,output_pattern):
	output_string = copy.deepcopy(output_pattern)
	if not matches_pattern(input_string,input_pattern):
		return False
	for i in range(0, len(input_pattern)):
		if input_pattern[i][0].islower() and input_pattern[i] != input_string[i]:
			return False
		else:
			for j in range(0,len(output_pattern)):
				if output_pattern[j] == input_pattern[i]:
					output_string[j] = input_string[i]
	return output_string

def replace_matching_subsequence(input_string,input_pattern,output_pattern):
	output_string = copy.deepcopy(input_string)
	for i in range(0,len(input_string)-len(input_pattern)+1):
		output_substring = rewrite_rule(input_string[i:len(input_pattern)+i],input_pattern,output_pattern)
		if output_substring != False:
			if type_of(output_substring) != None:
				output_string[i : i+len(input_pattern)] = [output_substring]
			else:
				output_string[i : i+len(input_pattern)] = output_substring
			return output_string
	return False

def matches_patterns(a,patterns):
	for pattern in patterns:
		if matches_pattern(a,pattern):
			return True
	return False


types = {}

def simplify_pattern(pattern):
	if type(pattern) == list:
		return pattern
	output = []
	for current in pattern.split(" "):
		if current == ":number":
			output += [is_number]
		elif current == ":string":
			output += [is_string]
		elif current[0] == ":" and (current[1:len(current)] in ["bool","set","list","imperative"]):
			output += [is_type_pattern(current[1:len(current)])]
		elif current == ":expr":
			output += [is_expr]
		else:
			output += [current]
	return output

def simplify_patterns(patterns):
	return [simplify_pattern(x) for x in [b.strip() for b in patterns.strip().split("\n")]]

def is_expr_(a):
	if type(a) == str and (a[0].isupper()):
		if not a in types:
			types[a] = "expr"
		return True
	return False

def is_expr(a):
	return type_of(a) != None

def matches_type_pattern(a,pattern):
	if(len(a) != len(pattern)):
		return False
	elif type(a) == str and a == pattern:
		return True
	for i in range(0,len(a)):
		if(type(pattern[i]) == str and pattern[i] != a[i] or callable(pattern[i]) and not pattern[i](a[i])):
			return False
	return True

global type_patterns
type_patterns = {}

def matches_type_patterns(a,patterns):
	for pattern in patterns:
		if matches_type_pattern(a,pattern):
			return True
	return False

def is_number(a):
	global type_patterns
	if (type(a) == str and (a.isdigit() or a[0].isupper())) or matches_type_patterns(a,type_patterns["number"]):
		if(type(a) == str):
			types[a] = "number"
		return True
	else:
		return False

def is_string(a):
	global type_patterns
	if (type(a) == str and (a[0] == "\"")) or matches_type_patterns(a,type_patterns["string"]):
		if(type(a) == str):
			types[a] = "string"
		return True
	else:
		return False

def is_type_pattern(the_type):
	def to_return(a):
		global type_patterns
		if (
			is_expr_(a)
			or matches_type_patterns(a,type_patterns[the_type])
		):
			if(type(a) == str):
				types[a] = the_type
			return True
		else:
			return False
	return to_return

type_patterns["list"] = simplify_patterns('''
:expr in :list
[ :expr for :expr in :list ]
[ :expr for :expr in :list if :bool ]
type of :expr
dot product of :list and :list
cross product of :list and :list
determinant of :list
largest number in :list
smallest number in :list
list ( itertools . permutations ( :list ) )
:list [ :: -1 ]
sorted ( :list )
:list + :list
''')

type_patterns["number"] = simplify_patterns('''
- :number
Math . PI
:number % :number
:number + :number
:number - :number
:number * :number
:number / :number
:number ** :number
maximize :number subject to :bool
absolute value of :number
sqrt ( :number )
cbrt :number
diff ( :number , :number )
integral of :number with respect to :number
integral of :number
integral of :number from :number to :number
integral of :number with respect to :number from :number to :number
greatest common factor of :number and :number
all solutions of :number
sin ( :number )
cos ( :number )
tan ( :number )
ln :number
limit of :number as :number -> :number
domain of the function :number
range of the function :number
a random number between :number and :number
distance from :list to :list
''')
type_patterns["bool"] = simplify_patterns('''
:string means :string
:number == :number
forall :expr ( :bool )
:bool == :bool
:list == :list
:set == :set
:number != :number
:bool != :bool
:list != :list
:set != :set
:number > :number
:number < :number
:number >= :number
:number <= :number
:number is a number
:number is a string
:number is a list
:bool or :bool
:bool and :bool
issubsetof ( :set , :set )
there exists :expr such that :bool
true
false
''')
type_patterns["set"] = simplify_patterns('''
intersection of :set and :set
union of :set and :set''')
type_patterns["string"] = simplify_patterns('''
substring of :string from :number to :number
longest match of the regex :string in the string :string
:string is a substring of :string
:string matches the regex :string
:replace :string in the string :string with :string''')

type_patterns["imperative"] = simplify_patterns('''
console . log ( :expr ) ;
while ( :bool ) { :imperative }
var :expr = :expr ;
:expr = :expr ;
input ( :string ) ;
return :expr ;
if :bool then :imperative
if :bool then :imperative else :imperative
''')


def pattern_to_string(pattern):
	str1 = ""
	num = 0
	for i in pattern:
		if(callable(i)):
			num += 1
			str1 += " A" + str(num)
		else:
			str1 += " " + i
	return str1 + " --> " + str1

rewrite_rule_string = ""
for current in type_patterns.values():
	for i in current:
		rewrite_rule_string += "\n" + pattern_to_string(i)
global rewrite_rules
global synonyms
synonyms = []

rewrite_rules = [[list(filter(None,i.split(" "))),list(filter(None,j.split(" ")))] for [i,j] in [a.split("-->") for a in [b.strip() for b in
	(rewrite_rule_string+'''
	first A letters of B --> substring of B from 0 to A
	A is a substring of B --> A is a substring of B
	A is a superset of B --> A is a subset of B
	A contains B --> B is in A
	the string A contains the string B --> B is a substring of A
	A is more than B --> A > B
	A and B --> A and B
	A squared --> A to the power of 2
	A cubed --> A to the power of 2
	although --> though
	greater --> more
	but --> and
	A though B --> A but B
	A is greater than or equal to B --> A >= B
	A is not equal to B --> A != B
	A is less than or equal to B --> A <= B
	A divided by B --> A / B
	A is less than B --> B > A
	A is C than B and D than E --> ( A is C than B ) and ( A is D than E )
	( A is C than B ) and D than E --> ( A is C than B ) and ( A is D than E )
	A minus B --> A - B
	sum of A and B --> A plus B
	A is not more than B --> A <= B
	A is not less than B --> A >= B
	A is not one of B --> A is not in B
	A is not in B --> (A is in B) == false
	( A ) --> A
	∞ --> infinity
	sum of A and B --> A + B
	A must be B --> A == B
	A must not be B --> A != B
	A cannot be B --> A != B
	the A --> A
	A is not B --> ( A != B )
	A is neither B nor C --> ( A is not B ) and ( A is not C )
	A is not a B --> ( A is a B ) == false
	A multiplied by B --> A * B
	A to the power of B --> A ** B
	[ A for A in B where D ] --> [ A for A in B if D ]
	[ A in B where D ] --> [ A for A in B where D ]
	every number in L that is greater than B --> [ a in B where a > B ]
	all permutations of X --> list ( itertools . permutations ( X ) )
	every permutation of X --> all permutations of X
	product of A and B --> A * B
	A mod B --> A % B
	A is divisible by B --> ( A % B ) == 0
	A is not X by Y --> ( A is X by Y ) == false
	A is not a number --> A is not a number
	if A then B otherwise C --> if A then B else C
	not A --> ( A == false )
	unless A then B otherwise C --> if ( not  A ) then B else C
	A is a match of the regex B --> A matches the regex B
	the string A matches the regex B --> A is a match of the regex B
	A are in B --> A is a subset of B
	plus --> +
	A is not a B of C --> ( A is a B of C ) == false
	none of A are in B --> A is not a subset of B
	derivative of A with respect to B --> diff ( A , B )
	A percent of B --> A * 0.10 * B
    A does not equal B --> A != B
    A ≠ B --> A != B
	A if B --> if B then A
	B implies A --> A if B
	A unless B --> if ( not A ) then B
	distance between A and B --> distance from A to B
	sin A --> sin ( A )
	cos A --> cos ( A )
	tan A --> tan ( A )
	sine of A --> sin A
	cosine of A --> cos A
	tangent of A --> tan A
	cotangent of A --> 1 / ( tan A )
	secant of A --> 1 / ( cos A )
	secant of A --> 1 / ( cos A )
	A = B --> A == B
	=/= --> !=
	A ÷ B --> A / B
	A % of B --> A percent of B
	quotient of A and B --> A / B
	A iff B --> ( A implies B ) and ( B implies A )
	⇔ --> iff
	[ A | B <- C ] --> [ A for B in C ]
	| A | --> absolute value of A
	square root of A --> sqrt ( A )
	cube root of A --> cbrt A
	^ --> **
	one --> 1
	two --> 2
	three --> 3
	four --> 4
	five --> 5
	six --> 6
	seven --> 7
	eight --> 8
	nine --> 9
	ten --> 10
	minimize A subject to B --> maximize A subject to - ( B )
	maximize A so that B --> maximize A subject to B
	maximize A such that B --> maximize A so that B
	maximize A where B --> maximize A so that B
	A ∪ B --> union of A and B
	A ∩ B --> intersection of A and B
	A || B --> A or B
	A && B --> A and B
	larger --> greater
	largest --> greatest
	maximum --> greatest
	minimum --> smallest
	natural logarithm of A --> ln A
	regex --> regular expression
	¬ A --> not A
	A ≥ B --> A >= B
	A ≤ B --> A <= B
	A ⊃ B --> A is a superset of B
	A ⊂ B --> A is a subset of B
	A is a subset of B --> issubsetof ( A , B )
	iff --> if and only if
	A is implied by B --> B implies A
	∀ A ( B ) --> forall A ( B )
	↔ --> iff
	A if B --> if B then A
	A ∨ B --> A or B
	∃ A ( B ) --> there exists A such that B
	if ( A ) { B } --> if A then B
	A is any of B --> A is in B
	A is in B --> A in B
	A equals B --> A == B
	A times B --> A * B
	A is a factor of B --> B is divisible by A
	longest substring of A that matches the regex B --> longest match of the regex B in the string A
	A is B --> A == B
	A reversed --> A [ :: -1 ]
	A spelled backwards --> A reversed
	A sorted --> sorted ( A )
	A in alphabetical order --> sorted ( A )
	until A { B } --> while ( A == false ) { B }
	A while B --> while ( B ) { A }
	! A --> A == false
	A := B --> A = B ;
	set A to B --> A := B
	initialize A to B --> var A = B ;
	A += B --> A := A + B
	A -= B --> A := A - B
	A *= B --> A := A * B
	A /= B --> A := A / B
	add A to B --> B += A
	divide B by A --> B /= A
	subtract B from A --> A -= B
	multiply A by B --> A *= B
	let A = B --> initialize A to B
	let A be B --> let A = B
	A until B --> until B { A }
	print A --> console . log ( A ) ;
	π --> Math . PI
	pi --> π
	if A { B } --> if A then B
	while A { B } --> while ( A ) { B }
	''').strip().split("\n")]]]
for current in rewrite_rules:
	if len(current[0]) == 1 and current[0] != current[1]:
		synonyms += [[current[0],current[1]]]
#print(synonyms)

def token_with_whitespace(a):
	return a[0] == "\"" and a[1].isalpha() or a[0].isalpha()

def join_tokens(token_list):
	result = ""
	for i in range(0,len(token_list)):
		if i == 0:
			result += token_list[i]
		elif token_with_whitespace(token_list[i]) and token_with_whitespace(token_list[i-1]):
			result += " __ " + token_list[i]
		else:
			result += " _ " + token_list[i]
	return result
			

def get_synonyms_list(synonym):
	global synonyms
	to_return = [[synonym]]
	changed = True
	while changed:
		changed = False
		for current in synonyms:
			if current[1] in to_return and not current[0] in to_return:
				to_return += [current[0]]
				changed = True
			elif current[0] in to_return and not current[1] in to_return:
				to_return += [current[1]]
				changed = True
	to_return = [["\""+b+"\"" for b in a] for a in to_return]
	if len(to_return) == 1:
		return to_return[0][0]
	#print(to_return)
	return "("+" | ".join([join_tokens(a) for a in to_return])+")"

#print(get_synonyms_list("though"))

def do_action(output):
	global rewrite_rules
	if len(output) == 3 and output[1] == "means":
		rewrite_rules += [[list(filter(None,output[0][1:len(output[0])-1].split(" "))),list(filter(None,output[2][1:len(output[2])-1].split(" ")))]]
		print([list(filter(None,output[0][1:len(output[0])-1].split(" "))),list(filter(None,output[2][1:len(output[2])-1].split(" ")))])
		print(rewrite_rules[0])
	else:
		print(str(output))

def recursive_rewrite(input_string):
	global rewrite_rules
	
	outputs = [input_string]

	changed = True
	while changed:
		changed = False
		for current in outputs:
			new_outputs = []
			for [a,b] in rewrite_rules:
				output = replace_matching_subsequence(current,a,b)
				if output != False and output not in outputs and output not in new_outputs:
					new_outputs += [output]
					changed = True
					if(len(output) == 1 and type_of(output[0]) != None):
						do_action(output[0])
						return [output]
			outputs = new_outputs
	return []

index_of = {}

def unparse(text):
	if type(text) == str:
		if text.isupper():
			return index_of[text]
		else:
			return "\""+text+"\"";
	if type(text) == list:
		return "("+" + \" \" + ".join([unparse(a) for a in text])+")"
	else:
		raise Exception("Cannot parse "+str(text))

def type_of(a):
	if is_expr_(a):
		return "expr"
	elif is_number(a):
		return "number"
	else:
		for current in type_patterns.keys():
			if not current in ["number","expr"] and is_type_pattern(current)(a):
				return current

thing = "hello"
print(thing and thing[0].isalpha())

#subprocess.run("python -m lark.tools.nearley parser.ne main nearley > parser.py".split(" "),shell=True)
#subprocess.run("nearleyc parser.ne -o parser.js".split(" "),shell=True)

import parser
#print(parser.parse("A is not less than B"))
print("type something to test the parser, or type exit to quit")
text_input = ""

for current in [tokenize_input(b.strip()) for b in ('''
	"while A do B end" means "while A { B }"
	"if A then B end" means "if A { B }"
	"A <- 1" means "A := 1"
	"while A < B do C endwhile" means "while A < B do C end"
	"A is not B" means "A != B"
	"A is not B" means "A == ( not B )"
	"A / B of C" means "C * ( A / B )"
	"half" means "1 / 2"
	"understand A as B" means "A means B"
	understand "twice A" as "A * 2"
	understand "the meaning of A is B" as "A means B"
	the meaning of "thrice A" is "A * 3"
	"twelve" means "12"
	"five" means "five"
	"nine" means "9"
	"A % of B" means "B * ( A / 100 )"
	"[every A in B such that D]" means "[ A for A in B where D ]"
'''.strip().split("\n"))]:
	recursive_rewrite(current)

while text_input != "exit":
	text_input = input()
	print(recursive_rewrite(tokenize_input(text_input)))