def if_then_else_demo():
"""
Demo if-then-else grammar
"""
from nltk.grammar import Nonterminal, Production, ContextFreeGrammar
nonterminals = 'E E1 PLUS T T1 TIMES F LPAREN RPAREN ID'
(E, E1, PLUS, T, T1, TIMES, F, LPAREN, RPAREN,
ID) = [Nonterminal(s) for s in nonterminals.split()]
productions = (
Production(E, [T, E1]),
Production(E1, [PLUS, T, E1]),
Production(E1, []),
Production(T, [F, T1]),
Production(T1, [TIMES, F, T1]),
Production(T1, []),
Production(F, [LPAREN, E, RPAREN]),
Production(F, [ID]),
Production(PLUS, ['+']),
Production(TIMES, ['*']),
Production(LPAREN, ['(']),
Production(RPAREN, [')']),
Production(ID, ['a']),
Production(ID, ['b']),
Production(ID, ['c']),
)
grammar = ContextFreeGrammar(E, productions)
text = "a * b + c".split()
RecursiveDescentApp(grammar, text).mainloop()
def demo2():
from nltk import Nonterminal, Production, ContextFreeGrammar
nonterminals = 'S VP NP PP P N Name V Det'
(S, VP, NP, PP, P, N, Name, V,
Det) = [Nonterminal(s) for s in nonterminals.split()]
productions = (
# Syntactic Productions
Production(S, [NP, VP]),
Production(NP, [Det, N]),
Production(NP, [NP, PP]),
Production(VP, [VP, PP]),
Production(VP, [V, NP, PP]),
Production(VP, [V, NP]),
Production(PP, [P, NP]),
Production(PP, []),
Production(PP, ['up', 'over', NP]),
# Lexical Productions
Production(NP, ['I']),
Production(Det, ['the']),
Production(Det, ['a']),
Production(N, ['man']),
Production(V, ['saw']),
Production(P, ['in']),
Production(P, ['with']),
Production(N, ['park']),
Production(N, ['dog']),
Production(N, ['statue']),
Production(Det, ['my']),
)
grammar = ContextFreeGrammar(S, productions)
text = 'I saw a man in the park'.split()
d = CFGDemo(grammar, text)
d.mainloop()
def demo():
from nltk import Nonterminal, ContextFreeGrammar
nonterminals = 'S VP NP PP P N Name V Det'
(S, VP, NP, PP, P, N, Name, V, Det) = [Nonterminal(s)
for s in nonterminals.split()]
grammar = ContextFreeGrammar.fromstring("""
S -> NP VP
PP -> P NP
NP -> Det N
NP -> NP PP
VP -> V NP
VP -> VP PP
Det -> 'a'
Det -> 'the'
Det -> 'my'
NP -> 'I'
N -> 'dog'
N -> 'man'
N -> 'park'
N -> 'statue'
V -> 'saw'
P -> 'in'
P -> 'up'
P -> 'over'
P -> 'with'
""")
def cb(grammar): print(grammar)
top = Tk()
editor = CFGEditor(top, grammar, cb)
Label(top, text='\nTesting CFG Editor\n').pack()
Button(top, text='Quit', command=top.destroy).pack()
top.mainloop()
def demo(N=23):
from nltk.grammar import ContextFreeGrammar
print('Generating the first %d sentences for demo grammar:' % (N, ))
print(demo_grammar)
grammar = ContextFreeGrammar.fromstring(demo_grammar)
for n, sent in enumerate(generate(grammar, n=N), 1):
print('%3d. %s' % (n, ' '.join(sent)))
def add_new_vocab_rule(self, rule):
"""
Adds a new vocabulary rule to the set of rules, and
recreates self.cfg and self.parser.
"""
self.rules.append(Production(NT(rule[0]), rule[1]))
self.cfg = ContextFreeGrammar(NT("S"), self.rules)
self.parser = EarleyChartParser(self.cfg, trace=0)
def set_grammar(self, grammar):
"""
Asign a new grammar to the parser
Args:
- parser_args: needs grammar of type :class:`ContextFreeGrammar`
"""
self._grammar = ContextFreeGrammar(Nonterminal('S'), grammar)
def __init__(self, parent, cfg=None, set_cfg_callback=None):
self._parent = parent
if cfg is not None: self._cfg = cfg
else: self._cfg = ContextFreeGrammar(Nonterminal('S'), [])
self._set_cfg_callback = set_cfg_callback
self._highlight_matching_nonterminals = 1
# Create the top-level window.
self._top = Toplevel(parent)
self._init_bindings()
self._init_startframe()
self._startframe.pack(side='top', fill='x', expand=0)
self._init_prodframe()
self._prodframe.pack(side='top', fill='both', expand=1)
self._init_buttons()
self._buttonframe.pack(side='bottom', fill='x', expand=0)
self._textwidget.focus()
def parse_NP(self, sen):
"""
Parses a partial sentence (that is, usually a noun phrase.
Returns the parse, or returns a tuple.
"""
try:
cfg_temp = ContextFreeGrammar(NT("NP"), self.rules)
parser_temp = EarleyChartParser(cfg_temp, trace=0)
parse = parser_temp.nbest_parse(sen.strip().split(" "), trace=0)
except:
print traceback.format_exc()
else:
if parse:
return parse[0]
print "failure"
return None
def fail_demo():
"""
Demo grammar that should not work with backtracking for all inputs
"""
from nltk.grammar import Nonterminal, Production, ContextFreeGrammar
S = Nonterminal('S')
A = Nonterminal('A')
productions = (
Production(S, [ A, S, A ]),
Production(S, [ A, A ]),
Production(A, [ 'a' ]),
)
grammar = ContextFreeGrammar(S, productions)
text = "a a a a a a".split()
#text = "a a a a".split()
RecursiveDescentApp(grammar, text).mainloop()
def __init__(self, parent, cfg=None, set_cfg_callback=None):
self._parent = parent
if cfg is not None: self._cfg = cfg
else: self._cfg = ContextFreeGrammar(Nonterminal('S'), [])
self._set_cfg_callback = set_cfg_callback
self._highlight_matching_nonterminals = 1
# Create the top-level window.
self._top = Toplevel(parent)
self._init_bindings()
self._init_startframe()
self._startframe.pack(side='top', fill='x', expand=0)
self._init_prodframe()
self._prodframe.pack(side='top', fill='both', expand=1)
self._init_buttons()
self._buttonframe.pack(side='bottom', fill='x', expand=0)
self._textwidget.focus()
def app():
"""
Create a shift reduce parser app, using a simple grammar and
text.
"""
from nltk.grammar import Nonterminal, Production, ContextFreeGrammar
nonterminals = 'S VP NP PP P N Name V Det'
(S, VP, NP, PP, P, N, Name, V,
Det) = [Nonterminal(s) for s in nonterminals.split()]
productions = (
# Syntactic Productions
Production(S, [NP, VP]),
Production(NP, [Det, N]),
Production(NP, [NP, PP]),
Production(VP, [VP, PP]),
Production(VP, [V, NP, PP]),
Production(VP, [V, NP]),
Production(PP, [P, NP]),
# Lexical Productions
Production(NP, ['I']),
Production(Det, ['the']),
Production(Det, ['a']),
Production(N, ['man']),
Production(V, ['saw']),
Production(P, ['in']),
Production(P, ['with']),
Production(N, ['park']),
Production(N, ['dog']),
Production(N, ['statue']),
Production(Det, ['my']),
)
grammar = ContextFreeGrammar(S, productions)
# tokenize the sentence
sent = 'my dog saw a man in the park with a statue'.split()
ShiftReduceApp(grammar, sent).mainloop()
def __init__(self, rules_file="rules.gr", vocab_file="vocabulary.gr"):
"""
Reads in grammar rules (from rules_file) and vocab rules (from
vocab_file) and creates self.cfg (a ContextFreeGrammar) and
self.parser (a EarleyChartParser).
"""
self.rules = []
test_sentences = []
# get the rules from rules_file
grammar = open(rules_file, "r")
line = grammar.readline()
while line:
if line.strip() != "" and not line.strip().startswith("#"):
line = line[2:]
parts = line.partition("\t")
lhs = parts[0].strip()
rhs = [NT(x) for x in parts[2].strip().split(" ")]
self.rules.append(Production(NT(lhs), rhs))
line = grammar.readline()
grammar.close()
# get the rules from vocab_file
vocab = open(vocab_file, "r")
line = vocab.readline()
while line:
if line.strip() != "" and not line.strip().startswith("#"):
line = line[2:]
parts = line.partition("\t")
lhs = parts[0].strip()
rhs = parts[2].strip().lower().split(" ")
self.rules.append(Production(NT(lhs), rhs))
line = vocab.readline()
vocab.close()
# create the grammar and parser
self.cfg = ContextFreeGrammar(NT("S"), self.rules)
self.parser = EarleyChartParser(self.cfg, trace=0)
def app():
"""
Create a recursive descent parser demo, using a simple grammar and
text.
"""
from nltk.grammar import ContextFreeGrammar
grammar = ContextFreeGrammar.fromstring("""
# Grammatical productions.
S -> NP VP
NP -> Det N PP | Det N
VP -> V NP PP | V NP | V
PP -> P NP
# Lexical productions.
NP -> 'I'
Det -> 'the' | 'a'
N -> 'man' | 'park' | 'dog' | 'telescope'
V -> 'ate' | 'saw'
P -> 'in' | 'under' | 'with'
""")
sent = 'the dog saw a man in the park'.split()
RecursiveDescentApp(grammar, sent).mainloop()
def demo():
from nltk import Nonterminal, ContextFreeGrammar
nonterminals = 'S VP NP PP P N Name V Det'
(S, VP, NP, PP, P, N, Name, V,
Det) = [Nonterminal(s) for s in nonterminals.split()]
grammar = ContextFreeGrammar.fromstring("""
S -> NP VP
PP -> P NP
NP -> Det N
NP -> NP PP
VP -> V NP
VP -> VP PP
Det -> 'a'
Det -> 'the'
Det -> 'my'
NP -> 'I'
N -> 'dog'
N -> 'man'
N -> 'park'
N -> 'statue'
V -> 'saw'
P -> 'in'
P -> 'up'
P -> 'over'
P -> 'with'
""")
def cb(grammar):
print(grammar)
top = Tk()
editor = CFGEditor(top, grammar, cb)
Label(top, text='\nTesting CFG Editor\n').pack()
Button(top, text='Quit', command=top.destroy).pack()
top.mainloop()
class CFGEditor(object):
"""
A dialog window for creating and editing context free grammars.
``CFGEditor`` imposes the following restrictions:
- All nonterminals must be strings consisting of word
characters.
- All terminals must be strings consisting of word characters
and space characters.
"""
# Regular expressions used by _analyze_line. Precompile them, so
# we can process the text faster.
ARROW = SymbolWidget.SYMBOLS['rightarrow']
_LHS_RE = re.compile(r"(^\s*\w+\s*)(->|("+ARROW+"))")
_ARROW_RE = re.compile("\s*(->|("+ARROW+"))\s*")
_PRODUCTION_RE = re.compile(r"(^\s*\w+\s*)" + # LHS
"(->|("+ARROW+"))\s*" + # arrow
r"((\w+|'[\w ]*'|\"[\w ]*\"|\|)\s*)*$") # RHS
_TOKEN_RE = re.compile("\\w+|->|'[\\w ]+'|\"[\\w ]+\"|("+ARROW+")")
_BOLD = ('helvetica', -12, 'bold')
def __init__(self, parent, cfg=None, set_cfg_callback=None):
self._parent = parent
if cfg is not None: self._cfg = cfg
else: self._cfg = ContextFreeGrammar(Nonterminal('S'), [])
self._set_cfg_callback = set_cfg_callback
self._highlight_matching_nonterminals = 1
# Create the top-level window.
self._top = Toplevel(parent)
self._init_bindings()
self._init_startframe()
self._startframe.pack(side='top', fill='x', expand=0)
self._init_prodframe()
self._prodframe.pack(side='top', fill='both', expand=1)
self._init_buttons()
self._buttonframe.pack(side='bottom', fill='x', expand=0)
self._textwidget.focus()
def _init_startframe(self):
frame = self._startframe = Frame(self._top)
self._start = Entry(frame)
self._start.pack(side='right')
Label(frame, text='Start Symbol:').pack(side='right')
Label(frame, text='Productions:').pack(side='left')
self._start.insert(0, self._cfg.start().symbol())
def _init_buttons(self):
frame = self._buttonframe = Frame(self._top)
Button(frame, text='Ok', command=self._ok,
underline=0, takefocus=0).pack(side='left')
Button(frame, text='Apply', command=self._apply,
underline=0, takefocus=0).pack(side='left')
Button(frame, text='Reset', command=self._reset,
underline=0, takefocus=0,).pack(side='left')
Button(frame, text='Cancel', command=self._cancel,
underline=0, takefocus=0).pack(side='left')
Button(frame, text='Help', command=self._help,
underline=0, takefocus=0).pack(side='right')
def _init_bindings(self):
self._top.title('CFG Editor')
self._top.bind('<Control-q>', self._cancel)
self._top.bind('<Alt-q>', self._cancel)
self._top.bind('<Control-d>', self._cancel)
#self._top.bind('<Control-x>', self._cancel)
self._top.bind('<Alt-x>', self._cancel)
self._top.bind('<Escape>', self._cancel)
#self._top.bind('<Control-c>', self._cancel)
self._top.bind('<Alt-c>', self._cancel)
self._top.bind('<Control-o>', self._ok)
self._top.bind('<Alt-o>', self._ok)
self._top.bind('<Control-a>', self._apply)
self._top.bind('<Alt-a>', self._apply)
self._top.bind('<Control-r>', self._reset)
self._top.bind('<Alt-r>', self._reset)
self._top.bind('<Control-h>', self._help)
self._top.bind('<Alt-h>', self._help)
self._top.bind('<F1>', self._help)
def _init_prodframe(self):
self._prodframe = Frame(self._top)
# Create the basic Text widget & scrollbar.
self._textwidget = Text(self._prodframe, background='#e0e0e0',
exportselection=1)
self._textscroll = Scrollbar(self._prodframe, takefocus=0,
orient='vertical')
self._textwidget.config(yscrollcommand = self._textscroll.set)
self._textscroll.config(command=self._textwidget.yview)
self._textscroll.pack(side='right', fill='y')
self._textwidget.pack(expand=1, fill='both', side='left')
# Initialize the colorization tags. Each nonterminal gets its
# own tag, so they aren't listed here.
self._textwidget.tag_config('terminal', foreground='#006000')
self._textwidget.tag_config('arrow', font='symbol')
self._textwidget.tag_config('error', background='red')
# Keep track of what line they're on. We use that to remember
# to re-analyze a line whenever they leave it.
self._linenum = 0
# Expand "->" to an arrow.
self._top.bind('>', self._replace_arrows)
# Re-colorize lines when appropriate.
self._top.bind('<<Paste>>', self._analyze)
self._top.bind('<KeyPress>', self._check_analyze)
self._top.bind('<ButtonPress>', self._check_analyze)
# Tab cycles focus. (why doesn't this work??)
def cycle(e, textwidget=self._textwidget):
textwidget.tk_focusNext().focus()
self._textwidget.bind('<Tab>', cycle)
prod_tuples = [(p.lhs(),[p.rhs()]) for p in self._cfg.productions()]
for i in range(len(prod_tuples)-1,0,-1):
if (prod_tuples[i][0] == prod_tuples[i-1][0]):
if () in prod_tuples[i][1]: continue
if () in prod_tuples[i-1][1]: continue
print(prod_tuples[i-1][1])
print(prod_tuples[i][1])
prod_tuples[i-1][1].extend(prod_tuples[i][1])
del prod_tuples[i]
for lhs, rhss in prod_tuples:
print(lhs, rhss)
s = '%s ->' % lhs
for rhs in rhss:
for elt in rhs:
if isinstance(elt, Nonterminal): s += ' %s' % elt
else: s += ' %r' % elt
s += ' |'
s = s[:-2] + '\n'
self._textwidget.insert('end', s)
self._analyze()
# # Add the producitons to the text widget, and colorize them.
# prod_by_lhs = {}
# for prod in self._cfg.productions():
# if len(prod.rhs()) > 0:
# prod_by_lhs.setdefault(prod.lhs(),[]).append(prod)
# for (lhs, prods) in prod_by_lhs.items():
# self._textwidget.insert('end', '%s ->' % lhs)
# self._textwidget.insert('end', self._rhs(prods[0]))
# for prod in prods[1:]:
# print '\t|'+self._rhs(prod),
# self._textwidget.insert('end', '\t|'+self._rhs(prod))
# print
# self._textwidget.insert('end', '\n')
# for prod in self._cfg.productions():
# if len(prod.rhs()) == 0:
# self._textwidget.insert('end', '%s' % prod)
# self._analyze()
# def _rhs(self, prod):
# s = ''
# for elt in prod.rhs():
# if isinstance(elt, Nonterminal): s += ' %s' % elt.symbol()
# else: s += ' %r' % elt
# return s
def _clear_tags(self, linenum):
"""
Remove all tags (except ``arrow`` and ``sel``) from the given
line of the text widget used for editing the productions.
"""
start = '%d.0'%linenum
end = '%d.end'%linenum
for tag in self._textwidget.tag_names():
if tag not in ('arrow', 'sel'):
self._textwidget.tag_remove(tag, start, end)
def _check_analyze(self, *e):
"""
Check if we've moved to a new line. If we have, then remove
all colorization from the line we moved to, and re-colorize
the line that we moved from.
"""
linenum = int(self._textwidget.index('insert').split('.')[0])
if linenum != self._linenum:
self._clear_tags(linenum)
self._analyze_line(self._linenum)
self._linenum = linenum
def _replace_arrows(self, *e):
"""
Replace any ``'->'`` text strings with arrows (char \\256, in
symbol font). This searches the whole buffer, but is fast
enough to be done anytime they press '>'.
"""
arrow = '1.0'
while True:
arrow = self._textwidget.search('->', arrow, 'end+1char')
if arrow == '': break
self._textwidget.delete(arrow, arrow+'+2char')
self._textwidget.insert(arrow, self.ARROW, 'arrow')
self._textwidget.insert(arrow, '\t')
arrow = '1.0'
while True:
arrow = self._textwidget.search(self.ARROW, arrow+'+1char',
'end+1char')
if arrow == '': break
self._textwidget.tag_add('arrow', arrow, arrow+'+1char')
def _analyze_token(self, match, linenum):
"""
Given a line number and a regexp match for a token on that
line, colorize the token. Note that the regexp match gives us
the token's text, start index (on the line), and end index (on
the line).
"""
# What type of token is it?
if match.group()[0] in "'\"": tag = 'terminal'
elif match.group() in ('->', self.ARROW): tag = 'arrow'
else:
# If it's a nonterminal, then set up new bindings, so we
# can highlight all instances of that nonterminal when we
# put the mouse over it.
tag = 'nonterminal_'+match.group()
if tag not in self._textwidget.tag_names():
self._init_nonterminal_tag(tag)
start = '%d.%d' % (linenum, match.start())
end = '%d.%d' % (linenum, match.end())
self._textwidget.tag_add(tag, start, end)
def _init_nonterminal_tag(self, tag, foreground='blue'):
self._textwidget.tag_config(tag, foreground=foreground,
font=CFGEditor._BOLD)
if not self._highlight_matching_nonterminals:
return
def enter(e, textwidget=self._textwidget, tag=tag):
textwidget.tag_config(tag, background='#80ff80')
def leave(e, textwidget=self._textwidget, tag=tag):
textwidget.tag_config(tag, background='')
self._textwidget.tag_bind(tag, '<Enter>', enter)
self._textwidget.tag_bind(tag, '<Leave>', leave)
def _analyze_line(self, linenum):
"""
Colorize a given line.
"""
# Get rid of any tags that were previously on the line.
self._clear_tags(linenum)
# Get the line line's text string.
line = self._textwidget.get(repr(linenum)+'.0', repr(linenum)+'.end')
# If it's a valid production, then colorize each token.
if CFGEditor._PRODUCTION_RE.match(line):
# It's valid; Use _TOKEN_RE to tokenize the production,
# and call analyze_token on each token.
def analyze_token(match, self=self, linenum=linenum):
self._analyze_token(match, linenum)
return ''
CFGEditor._TOKEN_RE.sub(analyze_token, line)
elif line.strip() != '':
# It's invalid; show the user where the error is.
self._mark_error(linenum, line)
def _mark_error(self, linenum, line):
"""
Mark the location of an error in a line.
"""
arrowmatch = CFGEditor._ARROW_RE.search(line)
if not arrowmatch:
# If there's no arrow at all, highlight the whole line.
start = '%d.0' % linenum
end = '%d.end' % linenum
elif not CFGEditor._LHS_RE.match(line):
# Otherwise, if the LHS is bad, highlight it.
start = '%d.0' % linenum
end = '%d.%d' % (linenum, arrowmatch.start())
else:
# Otherwise, highlight the RHS.
start = '%d.%d' % (linenum, arrowmatch.end())
end = '%d.end' % linenum
# If we're highlighting 0 chars, highlight the whole line.
if self._textwidget.compare(start, '==', end):
start = '%d.0' % linenum
end = '%d.end' % linenum
self._textwidget.tag_add('error', start, end)
def _analyze(self, *e):
"""
Replace ``->`` with arrows, and colorize the entire buffer.
"""
self._replace_arrows()
numlines = int(self._textwidget.index('end').split('.')[0])
for linenum in range(1, numlines+1): # line numbers start at 1.
self._analyze_line(linenum)
def _parse_productions(self):
"""
Parse the current contents of the textwidget buffer, to create
a list of productions.
"""
productions = []
# Get the text, normalize it, and split it into lines.
text = self._textwidget.get('1.0', 'end')
text = re.sub(self.ARROW, '->', text)
text = re.sub('\t', ' ', text)
lines = text.split('\n')
# Convert each line to a CFG production
for line in lines:
line = line.strip()
if line=='': continue
productions += parse_cfg_production(line)
#if line.strip() == '': continue
#if not CFGEditor._PRODUCTION_RE.match(line):
# raise ValueError('Bad production string %r' % line)
#
#(lhs_str, rhs_str) = line.split('->')
#lhs = Nonterminal(lhs_str.strip())
#rhs = []
#def parse_token(match, rhs=rhs):
# token = match.group()
# if token[0] in "'\"": rhs.append(token[1:-1])
# else: rhs.append(Nonterminal(token))
# return ''
#CFGEditor._TOKEN_RE.sub(parse_token, rhs_str)
#
#productions.append(Production(lhs, *rhs))
return productions
def _destroy(self, *e):
if self._top is None: return
self._top.destroy()
self._top = None
def _ok(self, *e):
self._apply()
self._destroy()
def _apply(self, *e):
productions = self._parse_productions()
start = Nonterminal(self._start.get())
cfg = ContextFreeGrammar(start, productions)
if self._set_cfg_callback is not None:
self._set_cfg_callback(cfg)
def _reset(self, *e):
self._textwidget.delete('1.0', 'end')
for production in self._cfg.productions():
self._textwidget.insert('end', '%s\n' % production)
self._analyze()
if self._set_cfg_callback is not None:
self._set_cfg_callback(self._cfg)
def _cancel(self, *e):
try: self._reset()
except: pass
self._destroy()
def _help(self, *e):
# The default font's not very legible; try using 'fixed' instead.
try:
ShowText(self._parent, 'Help: Chart Parser Demo',
(_CFGEditor_HELP).strip(), width=75, font='fixed')
except:
ShowText(self._parent, 'Help: Chart Parser Demo',
(_CFGEditor_HELP).strip(), width=75)
def _apply(self, *e):
productions = self._parse_productions()
start = Nonterminal(self._start.get())
cfg = ContextFreeGrammar(start, productions)
if self._set_cfg_callback is not None:
self._set_cfg_callback(cfg)
class CFGEditor(object):
"""
A dialog window for creating and editing context free grammars.
C{CFGEditor} places the following restrictions on what C{CFG}s can
be edited:
- All nonterminals must be strings consisting of word
characters.
- All terminals must be strings consisting of word characters
and space characters.
"""
# Regular expressions used by _analyze_line. Precompile them, so
# we can process the text faster.
ARROW = SymbolWidget.SYMBOLS['rightarrow']
_LHS_RE = re.compile(r"(^\s*\w+\s*)(->|(" + ARROW + "))")
_ARROW_RE = re.compile("\s*(->|(" + ARROW + "))\s*")
_PRODUCTION_RE = re.compile(r"(^\s*\w+\s*)" + # LHS
"(->|(" + ARROW + "))\s*" + # arrow
r"((\w+|'[\w ]*'|\"[\w ]*\"|\|)\s*)*$") # RHS
_TOKEN_RE = re.compile("\\w+|->|'[\\w ]+'|\"[\\w ]+\"|(" + ARROW + ")")
_BOLD = ('helvetica', -12, 'bold')
def __init__(self, parent, cfg=None, set_cfg_callback=None):
self._parent = parent
if cfg is not None: self._cfg = cfg
else: self._cfg = ContextFreeGrammar(Nonterminal('S'), [])
self._set_cfg_callback = set_cfg_callback
self._highlight_matching_nonterminals = 1
# Create the top-level window.
self._top = Toplevel(parent)
self._init_bindings()
self._init_startframe()
self._startframe.pack(side='top', fill='x', expand=0)
self._init_prodframe()
self._prodframe.pack(side='top', fill='both', expand=1)
self._init_buttons()
self._buttonframe.pack(side='bottom', fill='x', expand=0)
self._textwidget.focus()
def _init_startframe(self):
frame = self._startframe = Frame(self._top)
self._start = Entry(frame)
self._start.pack(side='right')
Label(frame, text='Start Symbol:').pack(side='right')
Label(frame, text='Productions:').pack(side='left')
self._start.insert(0, self._cfg.start().symbol())
def _init_buttons(self):
frame = self._buttonframe = Frame(self._top)
Button(frame, text='Ok', command=self._ok, underline=0,
takefocus=0).pack(side='left')
Button(frame,
text='Apply',
command=self._apply,
underline=0,
takefocus=0).pack(side='left')
Button(
frame,
text='Reset',
command=self._reset,
underline=0,
takefocus=0,
).pack(side='left')
Button(frame,
text='Cancel',
command=self._cancel,
underline=0,
takefocus=0).pack(side='left')
Button(frame,
text='Help',
command=self._help,
underline=0,
takefocus=0).pack(side='right')
def _init_bindings(self):
self._top.title('CFG Editor')
self._top.bind('<Control-q>', self._cancel)
self._top.bind('<Alt-q>', self._cancel)
self._top.bind('<Control-d>', self._cancel)
#self._top.bind('<Control-x>', self._cancel)
self._top.bind('<Alt-x>', self._cancel)
self._top.bind('<Escape>', self._cancel)
#self._top.bind('<Control-c>', self._cancel)
self._top.bind('<Alt-c>', self._cancel)
self._top.bind('<Control-o>', self._ok)
self._top.bind('<Alt-o>', self._ok)
self._top.bind('<Control-a>', self._apply)
self._top.bind('<Alt-a>', self._apply)
self._top.bind('<Control-r>', self._reset)
self._top.bind('<Alt-r>', self._reset)
self._top.bind('<Control-h>', self._help)
self._top.bind('<Alt-h>', self._help)
self._top.bind('<F1>', self._help)
def _init_prodframe(self):
self._prodframe = Frame(self._top)
# Create the basic Text widget & scrollbar.
self._textwidget = Text(self._prodframe,
background='#e0e0e0',
exportselection=1)
self._textscroll = Scrollbar(self._prodframe,
takefocus=0,
orient='vertical')
self._textwidget.config(yscrollcommand=self._textscroll.set)
self._textscroll.config(command=self._textwidget.yview)
self._textscroll.pack(side='right', fill='y')
self._textwidget.pack(expand=1, fill='both', side='left')
# Initialize the colorization tags. Each nonterminal gets its
# own tag, so they aren't listed here.
self._textwidget.tag_config('terminal', foreground='#006000')
self._textwidget.tag_config('arrow', font='symbol')
self._textwidget.tag_config('error', background='red')
# Keep track of what line they're on. We use that to remember
# to re-analyze a line whenever they leave it.
self._linenum = 0
# Expand "->" to an arrow.
self._top.bind('>', self._replace_arrows)
# Re-colorize lines when appropriate.
self._top.bind('<<Paste>>', self._analyze)
self._top.bind('<KeyPress>', self._check_analyze)
self._top.bind('<ButtonPress>', self._check_analyze)
# Tab cycles focus. (why doesn't this work??)
def cycle(e, textwidget=self._textwidget):
textwidget.tk_focusNext().focus()
self._textwidget.bind('<Tab>', cycle)
prod_tuples = [(p.lhs(), [p.rhs()]) for p in self._cfg.productions()]
for i in range(len(prod_tuples) - 1, 0, -1):
if (prod_tuples[i][0] == prod_tuples[i - 1][0]):
if () in prod_tuples[i][1]: continue
if () in prod_tuples[i - 1][1]: continue
print prod_tuples[i - 1][1]
print prod_tuples[i][1]
prod_tuples[i - 1][1].extend(prod_tuples[i][1])
del prod_tuples[i]
for lhs, rhss in prod_tuples:
print lhs, rhss
s = '%s ->' % lhs
for rhs in rhss:
for elt in rhs:
if isinstance(elt, Nonterminal): s += ' %s' % elt
else: s += ' %r' % elt
s += ' |'
s = s[:-2] + '\n'
self._textwidget.insert('end', s)
self._analyze()
# # Add the producitons to the text widget, and colorize them.
# prod_by_lhs = {}
# for prod in self._cfg.productions():
# if len(prod.rhs()) > 0:
# prod_by_lhs.setdefault(prod.lhs(),[]).append(prod)
# for (lhs, prods) in prod_by_lhs.items():
# self._textwidget.insert('end', '%s ->' % lhs)
# self._textwidget.insert('end', self._rhs(prods[0]))
# for prod in prods[1:]:
# print '\t|'+self._rhs(prod),
# self._textwidget.insert('end', '\t|'+self._rhs(prod))
# print
# self._textwidget.insert('end', '\n')
# for prod in self._cfg.productions():
# if len(prod.rhs()) == 0:
# self._textwidget.insert('end', '%s' % prod)
# self._analyze()
# def _rhs(self, prod):
# s = ''
# for elt in prod.rhs():
# if isinstance(elt, Nonterminal): s += ' %s' % elt.symbol()
# else: s += ' %r' % elt
# return s
def _clear_tags(self, linenum):
"""
Remove all tags (except C{arrow} and C{sel}) from the given
line of the text widget used for editing the productions.
"""
start = '%d.0' % linenum
end = '%d.end' % linenum
for tag in self._textwidget.tag_names():
if tag not in ('arrow', 'sel'):
self._textwidget.tag_remove(tag, start, end)
def _check_analyze(self, *e):
"""
Check if we've moved to a new line. If we have, then remove
all colorization from the line we moved to, and re-colorize
the line that we moved from.
"""
linenum = int(self._textwidget.index('insert').split('.')[0])
if linenum != self._linenum:
self._clear_tags(linenum)
self._analyze_line(self._linenum)
self._linenum = linenum
def _replace_arrows(self, *e):
"""
Replace any C{'->'} text strings with arrows (char \\256, in
symbol font). This searches the whole buffer, but is fast
enough to be done anytime they press '>'.
"""
arrow = '1.0'
while 1:
arrow = self._textwidget.search('->', arrow, 'end+1char')
if arrow == '': break
self._textwidget.delete(arrow, arrow + '+2char')
self._textwidget.insert(arrow, self.ARROW, 'arrow')
self._textwidget.insert(arrow, '\t')
arrow = '1.0'
while 1:
arrow = self._textwidget.search(self.ARROW, arrow + '+1char',
'end+1char')
if arrow == '': break
self._textwidget.tag_add('arrow', arrow, arrow + '+1char')
def _analyze_token(self, match, linenum):
"""
Given a line number and a regexp match for a token on that
line, colorize the token. Note that the regexp match gives us
the token's text, start index (on the line), and end index (on
the line).
"""
# What type of token is it?
if match.group()[0] in "'\"": tag = 'terminal'
elif match.group() in ('->', self.ARROW): tag = 'arrow'
else:
# If it's a nonterminal, then set up new bindings, so we
# can highlight all instances of that nonterminal when we
# put the mouse over it.
tag = 'nonterminal_' + match.group()
if tag not in self._textwidget.tag_names():
self._init_nonterminal_tag(tag)
start = '%d.%d' % (linenum, match.start())
end = '%d.%d' % (linenum, match.end())
self._textwidget.tag_add(tag, start, end)
def _init_nonterminal_tag(self, tag, foreground='blue'):
self._textwidget.tag_config(tag,
foreground=foreground,
font=CFGEditor._BOLD)
if not self._highlight_matching_nonterminals:
return
def enter(e, textwidget=self._textwidget, tag=tag):
textwidget.tag_config(tag, background='#80ff80')
def leave(e, textwidget=self._textwidget, tag=tag):
textwidget.tag_config(tag, background='')
self._textwidget.tag_bind(tag, '<Enter>', enter)
self._textwidget.tag_bind(tag, '<Leave>', leave)
def _analyze_line(self, linenum):
"""
Colorize a given line.
"""
# Get rid of any tags that were previously on the line.
self._clear_tags(linenum)
# Get the line line's text string.
line = self._textwidget.get( ` linenum ` + '.0', ` linenum ` + '.end')
# If it's a valid production, then colorize each token.
if CFGEditor._PRODUCTION_RE.match(line):
# It's valid; Use _TOKEN_RE to tokenize the production,
# and call analyze_token on each token.
def analyze_token(match, self=self, linenum=linenum):
self._analyze_token(match, linenum)
return ''
CFGEditor._TOKEN_RE.sub(analyze_token, line)
elif line.strip() != '':
# It's invalid; show the user where the error is.
self._mark_error(linenum, line)
def _mark_error(self, linenum, line):
"""
Mark the location of an error in a line.
"""
arrowmatch = CFGEditor._ARROW_RE.search(line)
if not arrowmatch:
# If there's no arrow at all, highlight the whole line.
start = '%d.0' % linenum
end = '%d.end' % linenum
elif not CFGEditor._LHS_RE.match(line):
# Otherwise, if the LHS is bad, highlight it.
start = '%d.0' % linenum
end = '%d.%d' % (linenum, arrowmatch.start())
else:
# Otherwise, highlight the RHS.
start = '%d.%d' % (linenum, arrowmatch.end())
end = '%d.end' % linenum
# If we're highlighting 0 chars, highlight the whole line.
if self._textwidget.compare(start, '==', end):
start = '%d.0' % linenum
end = '%d.end' % linenum
self._textwidget.tag_add('error', start, end)
def _analyze(self, *e):
"""
Replace C{->} with arrows, and colorize the entire buffer.
"""
self._replace_arrows()
numlines = int(self._textwidget.index('end').split('.')[0])
for linenum in range(1, numlines + 1): # line numbers start at 1.
self._analyze_line(linenum)
def _parse_productions(self):
"""
Parse the current contents of the textwidget buffer, to create
a list of productions.
"""
productions = []
# Get the text, normalize it, and split it into lines.
text = self._textwidget.get('1.0', 'end')
text = re.sub(self.ARROW, '->', text)
text = re.sub('\t', ' ', text)
lines = text.split('\n')
# Convert each line to a CFG production
for line in lines:
line = line.strip()
if line == '': continue
productions += parse_cfg_production(line)
#if line.strip() == '': continue
#if not CFGEditor._PRODUCTION_RE.match(line):
# raise ValueError('Bad production string %r' % line)
#
#(lhs_str, rhs_str) = line.split('->')
#lhs = Nonterminal(lhs_str.strip())
#rhs = []
#def parse_token(match, rhs=rhs):
# token = match.group()
# if token[0] in "'\"": rhs.append(token[1:-1])
# else: rhs.append(Nonterminal(token))
# return ''
#CFGEditor._TOKEN_RE.sub(parse_token, rhs_str)
#
#productions.append(Production(lhs, *rhs))
return productions
def _destroy(self, *e):
if self._top is None: return
self._top.destroy()
self._top = None
def _ok(self, *e):
self._apply()
self._destroy()
def _apply(self, *e):
productions = self._parse_productions()
start = Nonterminal(self._start.get())
cfg = ContextFreeGrammar(start, productions)
if self._set_cfg_callback is not None:
self._set_cfg_callback(cfg)
def _reset(self, *e):
self._textwidget.delete('1.0', 'end')
for production in self._cfg.productions():
self._textwidget.insert('end', '%s\n' % production)
self._analyze()
if self._set_cfg_callback is not None:
self._set_cfg_callback(self._cfg)
def _cancel(self, *e):
try:
self._reset()
except:
pass
self._destroy()
def _help(self, *e):
# The default font's not very legible; try using 'fixed' instead.
try:
ShowText(self._parent,
'Help: Chart Parser Demo', (_CFGEditor_HELP).strip(),
width=75,
font='fixed')
except:
ShowText(self._parent,
'Help: Chart Parser Demo', (_CFGEditor_HELP).strip(),
width=75)
#
# Creazione di una PCFG estratta dal treebank Penn
#
import nltk
from nltk.corpus import treebank
from nltk.grammar import ContextFreeGrammar, Nonterminal, Production
#estrazione delle produzioni dal treebank
production_list = list(production for sent in treebank.parsed_sents()
for production in sent.productions())
tbank_productions = set(production_list)
tbank_grammar = ContextFreeGrammar(Nonterminal('S'), list(tbank_productions))
grammar_productions = tbank_grammar.productions()
lhsCount={}
prodCount={}
probs={}
#preparo le strutture dati: count(a->b)/count(a) con a non terminale
for production in production_list:
if production.lhs() in lhsCount:
lhsCount[production.lhs()] = lhsCount[production.lhs()] + 1
else:
lhsCount[production.lhs()] = 1
if production in prodCount:
prodCount[production] = prodCount[production] + 1
else:
prodCount[production] = 1
def parse(self, p_string):
"""
Parses a string and stores the resulting hierarchy of "domains"
"hierarchies" and "tables"
For the sake of NLP I've parsed the string using the nltk
context free grammar library.
A query is a "sentence" and can either be a domain, hierarchy or a table.
A domain is simply a word.
A hierarchy is expressed as "domain/domain"
A table is exressed as "table(sentence, sentence, sentence)"
Internally the query is represented as a nltk.parse.tree
Process:
1. string is tokenized
2. develop a context free grammar
3. parse
4. convert to a tree representation
"""
self.nltktree = None
# Store the query string
self.string = p_string
# Tokenize the query string, allowing only strings, parentheses,
# forward slashes and commas.
re_all = r'table[(]|\,|[)]|[/]|\w+'
data_tokens = tokenize.regexp_tokenize(self.string, re_all)
# Develop a context free grammar
# S = sentence, T = table, H = hierarchy, D = domain
O, T, H, D = nonterminals('O, T, H, D')
# Specify the grammar
productions = (
# A sentence can be either a table, hierarchy or domain
Production(O, [D]),
Production(O, [H]),
Production(O, [T]),
# A table must be the following sequence:
# "table(", sentence, comma, sentence, comma, sentence, ")"
Production(T, ['table(', O, ',', O, ',', O, ')']),
# A hierarchy must be the following sequence:
# domain, forward slash, domain
Production(H, [D, '/', D]),
# domain, forward slash, another operator
Production(H, [D, '/', O]))
# Add domains to the cfg productions
# A domain is a token that is entirely word chars
re_domain = compile(r'^\w+$')
# Try every token and add if it matches the above regular expression
for tok in data_tokens:
if re_domain.match(tok):
prod = Production(D, [tok]),
productions = productions + prod
# Make a grammar out of our productions
grammar = ContextFreeGrammar(O, productions)
rd_parser = parse.RecursiveDescentParser(grammar)
# Tokens need to be redefined.
# It disappears after first use, and I don't know why.
tokens = tokenize.regexp_tokenize(self.string, re_all)
toklist = list(tokens)
# Store the parsing.
# Only the first one, as the grammar should be completely nonambiguous.
try:
self.parseList = rd_parser.get_parse_list(toklist)[0]
except IndexError:
print "Could not parse query."
return
# Set the nltk.parse.tree tree for this query to the global sentence
string = str(self.parseList)
string2 = string.replace(":", "").replace("')'", "").replace(
"table(", "").replace("','", "").replace("'", "").replace("/", "")
self.nltktree = parse.tree.bracket_parse(string2)
# Store the resulting nltk.parse.tree tree
self.parseTree = QuerySentence(self.nltktree)
self.xml = self.parseTree.toXML()
import nltk
from nltk.corpus import treebank
from nltk.grammar import ContextFreeGrammar, Nonterminal
tbank_productions = set(production for sent in treebank.parsed_sents()
for production in sent.productions())
tbank_grammar = ContextFreeGrammar(Nonterminal('S'), list(tbank_productions))
print tbank_grammar
