1 | #!/usr/bin/env python
|
---|
2 | # ******************************************************
|
---|
3 | # Michael Cohen <scudette@users.sourceforge.net>
|
---|
4 | #
|
---|
5 | # ******************************************************
|
---|
6 | # Version: FLAG $Version: 0.87-pre1 Date: Thu Jun 12 00:48:38 EST 2008$
|
---|
7 | # ******************************************************
|
---|
8 | #
|
---|
9 | # * This program is free software; you can redistribute it and/or
|
---|
10 | # * modify it under the terms of the GNU General Public License
|
---|
11 | # * as published by the Free Software Foundation; either version 2
|
---|
12 | # * of the License, or (at your option) any later version.
|
---|
13 | # *
|
---|
14 | # * This program is distributed in the hope that it will be useful,
|
---|
15 | # * but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
16 | # * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
17 | # * GNU General Public License for more details.
|
---|
18 | # *
|
---|
19 | # * You should have received a copy of the GNU General Public License
|
---|
20 | # * along with this program; if not, write to the Free Software
|
---|
21 | # * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
---|
22 | # ******************************************************
|
---|
23 | """ A simple feed lexer.
|
---|
24 | """
|
---|
25 |
|
---|
26 | import re,sys
|
---|
27 |
|
---|
28 | class Lexer:
|
---|
29 | """ A generic feed lexer """
|
---|
30 | ## The following is a description of the states we have and the
|
---|
31 | ## way we move through them: format is an array of
|
---|
32 | ## [ state_re, re, token/action, next state ]
|
---|
33 | tokens = []
|
---|
34 | state = "INITIAL"
|
---|
35 | buffer = ''
|
---|
36 | error = 0
|
---|
37 | verbose = 0
|
---|
38 | state_stack = []
|
---|
39 | processed = 0
|
---|
40 | processed_buffer = ''
|
---|
41 | saved_state = None
|
---|
42 | flags = 0
|
---|
43 |
|
---|
44 | def __init__(self, verbose=0, fd=None):
|
---|
45 | if not self.verbose:
|
---|
46 | self.verbose = verbose
|
---|
47 |
|
---|
48 | if len(self.tokens[0])==4:
|
---|
49 | for row in self.tokens:
|
---|
50 | row.append(re.compile(row[0], re.DOTALL))
|
---|
51 | row.append(re.compile(row[1], re.DOTALL | re.M | re.S | self.flags ))
|
---|
52 |
|
---|
53 | self.fd = fd
|
---|
54 |
|
---|
55 | def save_state(self, t=None, m=None):
|
---|
56 | """ Returns a dict which represents the current state of the lexer.
|
---|
57 |
|
---|
58 | When provided to restore_state, the lexer is guaranteed to be
|
---|
59 | in the same state as when the save_state was called.
|
---|
60 |
|
---|
61 | Note that derived classes may need to extend this.
|
---|
62 | """
|
---|
63 | ## Cant save our state if we have errors. We need to guarantee
|
---|
64 | ## that we rewind to a good part of the file.
|
---|
65 | if self.error: return
|
---|
66 | try:
|
---|
67 | end = m.end()
|
---|
68 | except: end = 0
|
---|
69 |
|
---|
70 | self.saved_state = dict(state_stack = self.state_stack[:],
|
---|
71 | processed = self.processed - end,
|
---|
72 | processed_buffer = self.processed_buffer,
|
---|
73 | readptr = self.fd.tell() - len(self.buffer) - end,
|
---|
74 | state = self.state,
|
---|
75 | objects = self.objects[:],
|
---|
76 | error = self.error,
|
---|
77 | )
|
---|
78 |
|
---|
79 | if self.verbose>1:
|
---|
80 | print "Saving state %s" % self.processed
|
---|
81 |
|
---|
82 | def restore_state(self):
|
---|
83 | state = self.saved_state
|
---|
84 | if not state: return
|
---|
85 |
|
---|
86 | self.state_stack = state['state_stack']
|
---|
87 | self.processed = state['processed']
|
---|
88 | self.processed_buffer = state['processed_buffer']
|
---|
89 | self.buffer = ''
|
---|
90 | self.fd.seek(state['readptr'])
|
---|
91 | self.state = state['state']
|
---|
92 | self.objects = state['objects']
|
---|
93 | self.error = state['error']
|
---|
94 | if self.verbose>1:
|
---|
95 | print "Restoring state to offset %s" % self.processed
|
---|
96 |
|
---|
97 | def next_token(self, end = True):
|
---|
98 | ## Now try to match any of the regexes in order:
|
---|
99 | current_state = self.state
|
---|
100 | for state_re, re_str, token, next, state, regex in self.tokens:
|
---|
101 | ## Does the rule apply for us now?
|
---|
102 | if state.match(current_state):
|
---|
103 | if self.verbose > 2:
|
---|
104 | print "%s: Trying to match %r with %r" % (self.state, self.buffer[:10], re_str)
|
---|
105 | m = regex.match(self.buffer)
|
---|
106 | if m:
|
---|
107 | if self.verbose > 3:
|
---|
108 | print "%s matched %s" % (re_str, m.group(0).encode("utf8"))
|
---|
109 | ## The match consumes the data off the buffer (the
|
---|
110 | ## handler can put it back if it likes)
|
---|
111 | self.processed_buffer += self.buffer[:m.end()]
|
---|
112 | self.buffer = self.buffer[m.end():]
|
---|
113 | self.processed += m.end()
|
---|
114 |
|
---|
115 | ## Try to iterate over all the callbacks specified:
|
---|
116 | for t in token.split(','):
|
---|
117 | try:
|
---|
118 | if self.verbose > 0:
|
---|
119 | print "0x%X: Calling %s %r" % (self.processed, t, m.group(0))
|
---|
120 | cb = getattr(self, t, self.default_handler)
|
---|
121 | except AttributeError:
|
---|
122 | continue
|
---|
123 |
|
---|
124 | ## Is there a callback to handle this action?
|
---|
125 | next_state = cb(t, m)
|
---|
126 | if next_state == "CONTINUE":
|
---|
127 | continue
|
---|
128 |
|
---|
129 | elif next_state:
|
---|
130 | next = next_state
|
---|
131 | self.state = next
|
---|
132 |
|
---|
133 |
|
---|
134 | if next:
|
---|
135 | self.state = next
|
---|
136 |
|
---|
137 | return token
|
---|
138 |
|
---|
139 | ## Check that we are making progress - if we are too full, we
|
---|
140 | ## assume we are stuck:
|
---|
141 | if end and len(self.buffer)>0 or len(self.buffer)>1024:
|
---|
142 | self.processed_buffer += self.buffer[:1]
|
---|
143 | self.buffer = self.buffer[1:]
|
---|
144 | self.ERROR("Lexer Stuck, discarding 1 byte (%r) - state %s" % (self.buffer[:10], self.state))
|
---|
145 | return "ERROR"
|
---|
146 |
|
---|
147 | ## No token were found
|
---|
148 | return None
|
---|
149 |
|
---|
150 | def feed(self, data):
|
---|
151 | self.buffer += data
|
---|
152 |
|
---|
153 | def empty(self):
|
---|
154 | return not len(self.buffer)
|
---|
155 |
|
---|
156 | def default_handler(self, token, match):
|
---|
157 | if self.verbose > 2:
|
---|
158 | print "Default handler: %s with %r" % (token,match.group(0))
|
---|
159 |
|
---|
160 | def ERROR(self, message = None, weight =1):
|
---|
161 | if self.verbose > 0 and message:
|
---|
162 | print "Error(%s): %s" % (weight,message)
|
---|
163 |
|
---|
164 | self.error += weight
|
---|
165 |
|
---|
166 | def PUSH_STATE(self, token = None, match = None):
|
---|
167 | if self.verbose > 1:
|
---|
168 | print "Storing state %s" % self.state
|
---|
169 | self.state_stack.append(self.state)
|
---|
170 |
|
---|
171 | def POP_STATE(self, token = None, match = None):
|
---|
172 | try:
|
---|
173 | state = self.state_stack.pop()
|
---|
174 | if self.verbose > 1:
|
---|
175 | print "Returned state to %s" % state
|
---|
176 |
|
---|
177 | return state
|
---|
178 | except IndexError:
|
---|
179 | print "Tried to pop the state but failed - possible recursion error"
|
---|
180 | return None
|
---|
181 |
|
---|
182 | def close(self):
|
---|
183 | """ Just a conveniece function to force us to parse all the data """
|
---|
184 | while self.next_token(): pass
|
---|
185 |
|
---|
186 | class SelfFeederMixIn(Lexer):
|
---|
187 | """ This mixin is used to make a lexer which feeds itself one
|
---|
188 | sector at the time.
|
---|
189 |
|
---|
190 | Note that self.fd must be the fd we read from.
|
---|
191 | """
|
---|
192 | def parse_fd(self, fd):
|
---|
193 | self.feed(fd.read())
|
---|
194 | while self.next_token(): pass
|
---|
195 |
|
---|