aboutsummaryrefslogtreecommitdiffstats
path: root/libmproxy/filt.py
blob: 7d2bd737f8d2269f1c0ef70e4129bc2a8d260e19 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
"""
    The following operators are understood:

        ~q          Request
        ~s          Response

    Headers:

        Patterns are matched against "name: value" strings. Field names are
        all-lowercase.

        ~a          Asset content-type in response. Asset content types are:
                        text/javascript
                        application/x-javascript
                        application/javascript
                        text/css
                        image/*
                        application/x-shockwave-flash
        ~h rex      Header line in either request or response
        ~hq rex     Header in request
        ~hs rex     Header in response

        ~b rex      Expression in the body of either request or response
        ~bq rex     Expression in the body of request
        ~bq rex     Expression in the body of response
        ~t rex      Shortcut for content-type header.

        ~d rex      Request domain
        ~m rex      Method
        ~u rex      URL
        ~c CODE     Response code.
        rex         Equivalent to ~u rex
"""
from __future__ import absolute_import
import re, sys
from .contrib import pyparsing as pp
from .protocol.http import decoded


class _Token:
    def dump(self, indent=0, fp=sys.stdout):
        print >> fp, "\t"*indent, self.__class__.__name__,
        if hasattr(self, "expr"):
            print >> fp, "(%s)"%self.expr,
        print >> fp


class _Action(_Token):
    @classmethod
    def make(klass, s, loc, toks):
        return klass(*toks[1:])


class FErr(_Action):
    code = "e"
    help = "Match error"
    def __call__(self, f):
        return True if f.error else False


class FReq(_Action):
    code = "q"
    help = "Match request with no response"
    def __call__(self, f):
        if not f.response:
            return True


class FResp(_Action):
    code = "s"
    help = "Match response"
    def __call__(self, f):
        return True if f.response else False


class _Rex(_Action):
    def __init__(self, expr):
        self.expr = expr
        try:
            self.re = re.compile(self.expr)
        except:
            raise ValueError, "Cannot compile expression."


def _check_content_type(expr, o):
    val = o.headers["content-type"]
    if val and re.search(expr, val[0]):
        return True
    return False


class FAsset(_Action):
    code = "a"
    help = "Match asset in response: CSS, Javascript, Flash, images."
    ASSET_TYPES = [
        "text/javascript",
        "application/x-javascript",
        "application/javascript",
        "text/css",
        "image/.*",
        "application/x-shockwave-flash"
    ]
    def __call__(self, f):
        if f.response:
            for i in self.ASSET_TYPES:
                if _check_content_type(i, f.response):
                    return True
        return False


class FContentType(_Rex):
    code = "t"
    help = "Content-type header"
    def __call__(self, f):
        if _check_content_type(self.expr, f.request):
            return True
        elif f.response and _check_content_type(self.expr, f.response):
            return True
        return False


class FRequestContentType(_Rex):
    code = "tq"
    help = "Request Content-Type header"
    def __call__(self, f):
        return _check_content_type(self.expr, f.request)


class FResponseContentType(_Rex):
    code = "ts"
    help = "Response Content-Type header"
    def __call__(self, f):
        if f.response:
            return _check_content_type(self.expr, f.response)
        return False


class FHead(_Rex):
    code = "h"
    help = "Header"
    def __call__(self, f):
        if f.request.headers.match_re(self.expr):
            return True
        elif f.response and f.response.headers.match_re(self.expr):
            return True
        return False


class FHeadRequest(_Rex):
    code = "hq"
    help = "Request header"
    def __call__(self, f):
        if f.request.headers.match_re(self.expr):
            return True


class FHeadResponse(_Rex):
    code = "hs"
    help = "Response header"
    def __call__(self, f):
        if f.response and f.response.headers.match_re(self.expr):
            return True


class FBod(_Rex):
    code = "b"
    help = "Body"
    def __call__(self, f):
        if f.request and f.request.content:
            with decoded(f.request):
                if re.search(self.expr, f.request.content):
                    return True
        if f.response and f.response.content:
            with decoded(f.response):
                if re.search(self.expr, f.response.content):
                    return True
        return False


class FBodRequest(_Rex):
    code = "bq"
    help = "Request body"
    def __call__(self, f):
        if f.request and f.request.content:
            with decoded(f.request):
                if re.search(self.expr, f.request.content):
                    return True


class FBodResponse(_Rex):
    code = "bs"
    help = "Response body"
    def __call__(self, f):
        if f.response and f.response.content:
            with decoded(f.response):
                if re.search(self.expr, f.response.content):
                    return True


class FMethod(_Rex):
    code = "m"
    help = "Method"
    def __call__(self, f):
        return bool(re.search(self.expr, f.request.method, re.IGNORECASE))


class FDomain(_Rex):
    code = "d"
    help = "Domain"
    def __call__(self, f):
        return bool(re.search(self.expr, f.request.host, re.IGNORECASE))


class FUrl(_Rex):
    code = "u"
    help = "URL"
    # FUrl is special, because it can be "naked".
    @classmethod
    def make(klass, s, loc, toks):
        if len(toks) > 1:
            toks = toks[1:]
        return klass(*toks)

    def __call__(self, f):
        return re.search(self.expr, f.request.url)


class _Int(_Action):
    def __init__(self, num):
        self.num = int(num)


class FCode(_Int):
    code = "c"
    help = "HTTP response code"
    def __call__(self, f):
        if f.response and f.response.code == self.num:
            return True


class FAnd(_Token):
    def __init__(self, lst):
        self.lst = lst

    def dump(self, indent=0, fp=sys.stdout):
        print >> fp, "\t"*indent, self.__class__.__name__
        for i in self.lst:
            i.dump(indent+1, fp)

    def __call__(self, f):
        return all(i(f) for i in self.lst)


class FOr(_Token):
    def __init__(self, lst):
        self.lst = lst

    def dump(self, indent=0, fp=sys.stdout):
        print >> fp, "\t"*indent, self.__class__.__name__
        for i in self.lst:
            i.dump(indent+1, fp)

    def __call__(self, f):
        return any(i(f) for i in self.lst)


class FNot(_Token):
    def __init__(self, itm):
        self.itm = itm[0]

    def dump(self, indent=0, fp=sys.stdout):
        print >> fp, "\t"*indent, self.__class__.__name__
        self.itm.dump(indent + 1, fp)

    def __call__(self, f):
        return not self.itm(f)


filt_unary = [
    FReq,
    FResp,
    FAsset,
    FErr
]
filt_rex = [
    FHeadRequest,
    FHeadResponse,
    FHead,
    FBodRequest,
    FBodResponse,
    FBod,
    FMethod,
    FDomain,
    FUrl,
    FRequestContentType,
    FResponseContentType,
    FContentType,
]
filt_int = [
    FCode
]
def _make():
    # Order is important - multi-char expressions need to come before narrow
    # ones.
    parts = []
    for klass in filt_unary:
        f = pp.Literal("~%s"%klass.code)
        f.setParseAction(klass.make)
        parts.append(f)

    simplerex = "".join(c for c in pp.printables if c not in  "()~'\"")
    rex = pp.Word(simplerex) |\
          pp.QuotedString("\"", escChar='\\') |\
          pp.QuotedString("'", escChar='\\')
    for klass in filt_rex:
        f = pp.Literal("~%s"%klass.code) + rex.copy()
        f.setParseAction(klass.make)
        parts.append(f)

    for klass in filt_int:
        f = pp.Literal("~%s"%klass.code) + pp.Word(pp.nums)
        f.setParseAction(klass.make)
        parts.append(f)

    # A naked rex is a URL rex:
    f = rex.copy()
    f.setParseAction(FUrl.make)
    parts.append(f)

    atom = pp.MatchFirst(parts)
    expr = pp.operatorPrecedence(
                atom,
                [
                    (pp.Literal("!").suppress(), 1, pp.opAssoc.RIGHT, lambda x: FNot(*x)),
                    (pp.Literal("&").suppress(), 2, pp.opAssoc.LEFT, lambda x: FAnd(*x)),
                    (pp.Literal("|").suppress(), 2, pp.opAssoc.LEFT, lambda x: FOr(*x)),
                ]
           )
    expr = pp.OneOrMore(expr)
    return expr.setParseAction(lambda x: FAnd(x) if len(x) != 1 else x)
bnf = _make()


def parse(s):
    try:
        return bnf.parseString(s, parseAll=True)[0]
    except pp.ParseException:
        return None
    except ValueError:
        return None