cleaning & parser grammer input

Parsing.py

@@ -15,20 +15,18 @@ class Node():
         self.name     = name
         self.lex      = lex
 
-# A State has a list of rules, and a list of accepting tokens
-# accepting: string[]
-# rules:     string[]
+# A State has a list of rules, and a num indicating which state it is
+# num:   int
+# rules: string[]
 class State():
     def __init__(self):
-        self.accepting = []
-        self.rules     = []
+        self.num   = -1
+        self.rules = []
 
-    # addRule simply adds the string rule to rules, and adds the
-    #   accepting token to accepting
+    # addRule simply adds the string rule to rules
     # rule: string
-    def addRule(self, rule):
-        acceptingToken = rule[1]
-        self.accepting.append(acceptingToken)
+    def addRule(self, i, rule):
+        self.num = i
         self.rules.append(rule)
 
     # go to next state depending on what states you've already visited
@@ -50,7 +48,8 @@ class State():
                 break
 
         if not ruleToUse:
-            raise ValueError('No rule found') 
+            raise ValueError('No rule found for next=({}) in self.rules'.format(next),
+                self.rules, statesVisited, rulesToOutput)
 
         # what type of rule is it? shift or reduce?
         if (ruleToUse[2] == 'shift'):
@@ -80,12 +79,10 @@ class State():
             raise ValueError('rule neither shift nor reduce.')
 
     def output(self):
-        print('accepting: ['),
-        for i in self.accepting:
-            print(i + ', '),
-        print(']\nrules: ['),
-        for i in self.rules:
-            print(i + ', '),
+        print('state ' + self.num)
+        print('rules: ['),
+        for r in self.rules:
+            print(r + ', '),
         print(']')
 
 ##################### Globals ###############################
@@ -94,7 +91,7 @@ class State():
 cfgrules = [] # String[][]
 lr1trans = [] # String[][]
 theStack = [] # Node[]
-states   = [State() for _ in range(15)] # State[]
+states   = [] # State[]
 
 ##################### Main Functions ##########################################
 
@@ -161,28 +158,33 @@ def setUpGrammar():
     cfgrules = []
     lr1trans = []
     theStack = []
-    states   = [State() for _ in range(15)]
-
-    # TODO: is there an easier way to convert this?
-    #  '0 BOF shift 1\n1 procedures shift 2'
-    #  => [['0','BOF','shift','1'],['1','procedures','shift','2']]
-
+    states   = []
+
+    # one big file
+    with open('cfg/trans.txt', 'r') as f:
+        lines = f.read().splitlines()
+    
+    i  = int(lines[0]) + 1 # terminals
+    i += int(lines[i]) + 1 # non terminals
+    i += 1 # starting non terminal
+    
     # cfg rules
-    with open('cfgrules', 'r') as f:
-        cfgrules = f.read().splitlines()
-    for i, t in enumerate(cfgrules):
-        cfgrules[i] = splitString(t)
+    for r in range(int(lines[i])):
+        cfgrules.append(lines[i+r+1].split())
+        
+    # num of states
+    i     += int(lines[i]) + 1
+    states = [State() for _ in range(int(lines[i]))]
 
     # transitions
-    with open('lr1trans', 'r') as f:
-        lr1trans = f.read().splitlines()
-    for i, t in enumerate(lr1trans):
-        lr1trans[i] = splitString(t)
+    i += 1
+    for t in range(int(lines[i])):
+        lr1trans.append(lines[i+t+1].split())            
 
     # states
     for t in lr1trans:
         transState = int(t[0])
-        states[transState].addRule(t)
+        states[transState].addRule(transState, t)
 
 # main parse function
 # tokens: Token[]

cfgrules

-start BOF procedures EOF
-procedures CLASS ID LBRACK dcl RBRACK
-dcl INT ID ASSIGN ZERO SEMICO
-dcl
\ No newline at end of file

lr1trans

-0 BOF shift 1
-1 procedures shift 2
-1 CLASS shift 5
-2 EOF shift 15
-5 ID shift 6
-6 LBRACK shift 7
-7 INT shift 8
-7 dcl shift 13
-8 ID shift 9
-9 ASSIGN shift 10
-10 ZERO shift 11
-11 SEMICO shift 12
-12 RBRACK reduce 2
-13 RBRACK shift 14
-14 EOF reduce 1
\ No newline at end of file

utils.py

@@ -2,27 +2,6 @@ import string
 
 # ------- Utility functions
 
-# split string into array of strings
-# e.g. '0 BOF shift 80' => ['0', 'BOF', 'shift', '80']
-# s: string
-def splitString(s):
-    words = []
-    isword = 0
-    for c in s:
-        if c in " \r\n\t": # whitespace
-            isword = 0
-        elif not isword:
-            words = words + [c]
-            isword = 1
-        else:
-            words[-1] = words[-1] + c
-    return words
-
-# count number of words in a string
-# s: string
-def countWords(s):
-    return len(splitString(s))
-
 # a: stack
 def printStack(a):
     print('['),