This file is indexed.

/usr/lib/python2.7/dist-packages/whoosh/classify.py is in python-whoosh 2.7.0-2.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  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
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
# Copyright 2008 Matt Chaput. All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
#    1. Redistributions of source code must retain the above copyright notice,
#       this list of conditions and the following disclaimer.
#
#    2. Redistributions in binary form must reproduce the above copyright
#       notice, this list of conditions and the following disclaimer in the
#       documentation and/or other materials provided with the distribution.
#
# THIS SOFTWARE IS PROVIDED BY MATT CHAPUT ``AS IS'' AND ANY EXPRESS OR
# IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
# MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
# EVENT SHALL MATT CHAPUT OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
# INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA,
# OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
# LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
# NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE,
# EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#
# The views and conclusions contained in the software and documentation are
# those of the authors and should not be interpreted as representing official
# policies, either expressed or implied, of Matt Chaput.

"""Classes and functions for classifying and extracting information from
documents.
"""

from __future__ import division
import random
from collections import defaultdict
from math import log

from whoosh.compat import xrange, iteritems


# Expansion models

class ExpansionModel(object):
    def __init__(self, doc_count, field_length):
        self.N = doc_count
        self.collection_total = field_length

        if self.N:
            self.mean_length = self.collection_total / self.N
        else:
            self.mean_length = 0

    def normalizer(self, maxweight, top_total):
        raise NotImplementedError

    def score(self, weight_in_top, weight_in_collection, top_total):
        raise NotImplementedError


class Bo1Model(ExpansionModel):
    def normalizer(self, maxweight, top_total):
        f = maxweight / self.N
        return (maxweight * log((1.0 + f) / f) + log(1.0 + f)) / log(2.0)

    def score(self, weight_in_top, weight_in_collection, top_total):
        f = weight_in_collection / self.N
        return weight_in_top * log((1.0 + f) / f, 2) + log(1.0 + f, 2)


class Bo2Model(ExpansionModel):
    def normalizer(self, maxweight, top_total):
        f = maxweight * self.N / self.collection_total
        return maxweight * log((1.0 + f) / f, 2) + log(1.0 + f, 2)

    def score(self, weight_in_top, weight_in_collection, top_total):
        f = weight_in_top * top_total / self.collection_total
        return weight_in_top * log((1.0 + f) / f, 2) + log(1.0 + f, 2)


class KLModel(ExpansionModel):
    def normalizer(self, maxweight, top_total):
        return (maxweight * log(self.collection_total / top_total) / log(2.0)
                * top_total)

    def score(self, weight_in_top, weight_in_collection, top_total):
        wit_over_tt = weight_in_top / top_total
        wic_over_ct = weight_in_collection / self.collection_total

        if wit_over_tt < wic_over_ct:
            return 0
        else:
            return wit_over_tt * log(wit_over_tt
                                     / (weight_in_top / self.collection_total),
                                     2)


class Expander(object):
    """Uses an ExpansionModel to expand the set of query terms based on the top
    N result documents.
    """

    def __init__(self, ixreader, fieldname, model=Bo1Model):
        """
        :param reader: A :class:whoosh.reading.IndexReader object.
        :param fieldname: The name of the field in which to search.
        :param model: (classify.ExpansionModel) The model to use for expanding
            the query terms. If you omit this parameter, the expander uses
            :class:`Bo1Model` by default.
        """

        self.ixreader = ixreader
        self.fieldname = fieldname
        doccount =  self.ixreader.doc_count_all()
        fieldlen = self.ixreader.field_length(fieldname)

        if type(model) is type:
            model = model(doccount, fieldlen)
        self.model = model

        # Maps words to their weight in the top N documents.
        self.topN_weight = defaultdict(float)

        # Total weight of all terms in the top N documents.
        self.top_total = 0

    def add(self, vector):
        """Adds forward-index information about one of the "top N" documents.

        :param vector: A series of (text, weight) tuples, such as is
            returned by Reader.vector_as("weight", docnum, fieldname).
        """

        total_weight = 0
        topN_weight = self.topN_weight

        for word, weight in vector:
            total_weight += weight
            topN_weight[word] += weight

        self.top_total += total_weight

    def add_document(self, docnum):
        ixreader = self.ixreader
        if self.ixreader.has_vector(docnum, self.fieldname):
            self.add(ixreader.vector_as("weight", docnum, self.fieldname))
        elif self.ixreader.schema[self.fieldname].stored:
            self.add_text(ixreader.stored_fields(docnum).get(self.fieldname))
        else:
            raise Exception("Field %r in document %s is not vectored or stored"
                            % (self.fieldname, docnum))

    def add_text(self, string):
        # Unfortunately since field.index() yields bytes texts, and we want
        # unicode, we end up encoding and decoding unnecessarily.
        #
        # TODO: Find a way around this

        field = self.ixreader.schema[self.fieldname]
        from_bytes = field.from_bytes
        self.add((from_bytes(text), weight) for text, _, weight, _
                 in field.index(string))

    def expanded_terms(self, number, normalize=True):
        """Returns the N most important terms in the vectors added so far.

        :param number: The number of terms to return.
        :param normalize: Whether to normalize the weights.
        :returns: A list of ("term", weight) tuples.
        """

        model = self.model
        fieldname = self.fieldname
        ixreader = self.ixreader
        field = ixreader.schema[fieldname]
        tlist = []
        maxweight = 0

        # If no terms have been added, return an empty list
        if not self.topN_weight:
            return []

        for word, weight in iteritems(self.topN_weight):
            btext = field.to_bytes(word)
            if (fieldname, btext) in ixreader:
                cf = ixreader.frequency(fieldname, btext)
                score = model.score(weight, cf, self.top_total)
                if score > maxweight:
                    maxweight = score
                tlist.append((score, word))

        if normalize:
            norm = model.normalizer(maxweight, self.top_total)
        else:
            norm = maxweight
        tlist = [(weight / norm, t) for weight, t in tlist]
        tlist.sort(key=lambda x: (0 - x[0], x[1]))

        return [(t, weight) for weight, t in tlist[:number]]


# Similarity functions

def shingles(input, size=2):
    d = defaultdict(int)
    for shingle in (input[i:i + size]
                    for i in xrange(len(input) - (size - 1))):
        d[shingle] += 1
    return iteritems(d)


def simhash(features, hashbits=32):
    if hashbits == 32:
        hashfn = hash
    else:
        hashfn = lambda s: _hash(s, hashbits)

    vs = [0] * hashbits
    for feature, weight in features:
        h = hashfn(feature)
        for i in xrange(hashbits):
            if h & (1 << i):
                vs[i] += weight
            else:
                vs[i] -= weight

    out = 0
    for i, v in enumerate(vs):
        if v > 0:
            out |= 1 << i
    return out


def _hash(s, hashbits):
    # A variable-length version of Python's builtin hash
    if s == "":
        return 0
    else:
        x = ord(s[0]) << 7
        m = 1000003
        mask = 2 ** hashbits - 1
        for c in s:
            x = ((x * m) ^ ord(c)) & mask
        x ^= len(s)
        if x == -1:
            x = -2
        return x


def hamming_distance(first_hash, other_hash, hashbits=32):
    x = (first_hash ^ other_hash) & ((1 << hashbits) - 1)
    tot = 0
    while x:
        tot += 1
        x &= x - 1
    return tot


# Clustering

def kmeans(data, k, t=0.0001, distfun=None, maxiter=50, centers=None):
    """
    One-dimensional K-means clustering function.

    :param data: list of data points.
    :param k: number of clusters.
    :param t: tolerance; stop if changes between iterations are smaller than
        this value.
    :param distfun: a distance function.
    :param centers: a list of centroids to start with.
    :param maxiter: maximum number of iterations to run.
    """

    # Adapted from a C version by Roger Zhang, <rogerz@cs.dal.ca>
    # http://cs.smu.ca/~r_zhang/code/kmeans.c

    DOUBLE_MAX = 1.797693e308
    n = len(data)

    error = DOUBLE_MAX  # sum of squared euclidean distance

    counts = [0] * k  # size of each cluster
    labels = [0] * n  # output cluster label for each data point

    # c1 is an array of len k of the temp centroids
    c1 = [0] * k

    # choose k initial centroids
    if centers:
        c = centers
    else:
        c = random.sample(data, k)

    niter = 0
    # main loop
    while True:
        # save error from last step
        old_error = error
        error = 0

        # clear old counts and temp centroids
        for i in xrange(k):
            counts[i] = 0
            c1[i] = 0

        for h in xrange(n):
            # identify the closest cluster
            min_distance = DOUBLE_MAX
            for i in xrange(k):
                distance = (data[h] - c[i]) ** 2
                if distance < min_distance:
                    labels[h] = i
                    min_distance = distance

            # update size and temp centroid of the destination cluster
            c1[labels[h]] += data[h]
            counts[labels[h]] += 1
            # update standard error
            error += min_distance

        for i in xrange(k):  # update all centroids
            c[i] = c1[i] / counts[i] if counts[i] else c1[i]

        niter += 1
        if (abs(error - old_error) < t) or (niter > maxiter):
            break

    return labels, c


# Sliding window clusters

def two_pass_variance(data):
    n = 0
    sum1 = 0
    sum2 = 0

    for x in data:
        n += 1
        sum1 = sum1 + x

    mean = sum1 / n

    for x in data:
        sum2 += (x - mean) * (x - mean)

    variance = sum2 / (n - 1)
    return variance


def weighted_incremental_variance(data_weight_pairs):
    mean = 0
    S = 0
    sumweight = 0
    for x, weight in data_weight_pairs:
        temp = weight + sumweight
        Q = x - mean
        R = Q * weight / temp
        S += sumweight * Q * R
        mean += R
        sumweight = temp
    Variance = S / (sumweight - 1)  # if sample is the population, omit -1
    return Variance


def swin(data, size):
    clusters = []
    for i, left in enumerate(data):
        j = i
        right = data[j]
        while j < len(data) - 1 and right - left < size:
            j += 1
            right = data[j]
        v = 99999
        if j - i > 1:
            v = two_pass_variance(data[i:j + 1])
        clusters.append((left, right, j - i, v))
    clusters.sort(key=lambda x: (0 - x[2], x[3]))
    return clusters