This file is indexed.

/usr/share/pyshared/pebl/util.py is in python-pebl 1.0.2-2build1.

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
"""Miscellaneous utility functions."""

import numpy as N
import math
import os.path
from copy import copy
from collections import deque

def as_list(c):
    """Ensures that the result is a list.

    If input is a list/tuple/set, return it.
    If it's None, return empty list.
    Else, return a list with input as the only element.
    
    """

    if isinstance(c, (list,tuple,set)):
        return c
    elif c is None:
        return []
    else:
        return [c]


def cond(condition, expr1, expr2):
    """Marked for deletion.. Python2.5 provides this."""
    
    if condition:
        return expr1
    else:
        return expr2


def flatten(seq):
    """Given a nested datastructure, flatten it."""

    lst = []
    for el in seq:
        if type(el) in [list, tuple, set]:
            lst.extend(flatten(el))
        else:
            lst.append(el)
    return lst


def normalize(lst):
    """Normalizes a list of numbers (sets sum to 1.0)."""

    if not isinstance(lst, N.ndarray):
        lst = N.array(lst)

    return lst/lst.sum()

def rescale_logvalues(lst):
    """Rescales a list of log values by setting max value to 0.0

    This function is necessary when working with list of log values. Without
    it, we could have overflows. This is a lot faster than using arbitrary
    precision math libraries.
    
    """

    if not isinstance(lst, N.ndarray):
        lst = N.array(lst) 
    
    return lst - lst.max()

_LogZero = 1.0e-100
_MinLogExp = math.log(_LogZero);
def logadd(x, y):
    """Adds two log values.
    
    Ensures accuracy even when the difference between values is large.
    
    """
    if x < y:
        temp = x
        x = y
        y = temp

    z = math.exp(y - x)
    logProb = x + math.log(1.0 + z)
    if logProb < _MinLogExp:
        return _MinLogExp
    else:
        return logProb

def logsum(lst):
    """Sums a list of log values, ensuring accuracy."""
    
    if not isinstance(lst, N.ndarray):
        lst = N.array(lst)
    
    maxval = lst.max()
    lst = lst - maxval
    return reduce(logadd, lst) + maxval


## from webpy (webpy.org)
def autoassign(self, locals):
    """
    Automatically assigns local variables to `self`.
    Generally used in `__init__` methods, as in::

        def __init__(self, foo, bar, baz=1): 
            autoassign(self, locals())

    """
    for (key, value) in locals.iteritems():
        if key == 'self': 
            continue
        setattr(self, key, value)


def unzip(l, *jj):
    """Opposite of zip().

    jj is a tuple of list indexes (or keys) to extract or unzip. If not
    specified, all items are unzipped.

    """
	
    if jj==():
	    jj=range(len(l[0]))
    rl = [[li[j] for li in l] for j in jj] # a list of lists
    if len(rl)==1:
        rl=rl[0] #convert list of 1 list to a list
    return rl


def nestediter(lst1, lst2):
    """A syntactic shortform for doing nested loops."""
    for i in lst1:
        for j in lst2:
            yield (i,j)



def cartesian_product(list_of_lists):
    """Given n lists (or sets), generate all n-tuple combinations.

    >>> list(cartesian_product([[0,1], [0,1,"foo"]]))
    [(0, 0), (0, 1), (0, 'foo'), (1, 0), (1, 1), (1, 'foo')]

     >>> list(cartesian_product([[0,1], [0,1], [0,1]]))
     [(0, 0, 0), (0, 0, 1), (0, 1, 0), (0, 1, 1), (1, 0, 0), (1, 0, 1), (1, 1, 0), (1, 1, 1)]
    
    """

    head,rest = list_of_lists[0], list_of_lists[1:]
    if len(rest) is 0:
        for val in head:
            yield (val,)
    else:
        for val in head:
            for val2 in cartesian_product(rest):
                yield (val,) + val2


def probwheel(items, weights):
    """Randomly select an item from a weighted list of items."""
    
    # convert to numpy array and normalize
    weights = normalize(N.array(weights))

    # edges for bins
    binedges = weights.cumsum()
    randval = N.random.random()
    for item, edge in zip(items, binedges):
        if randval <= edge:
            return item
    
    # should never reach here.. but might due to rounding errors.
    return items[-1]


def logscale_probwheel(items, logweights):
    """Randomly select an item from a [log] weighted list of items.
    
    Fucntion just rescale logweights and exponentiates before calling
    probwheel. 
    
    """
    return probwheel(items, N.exp(rescale_logvalues(logweights)))


def entropy_of_list(lst):
    """Given a list of values, generate histogram and calculate the entropy."""

    unique_values = N.unique(lst)
    unique_counts = N.array([float(len([i for i in lst if i == unique_val])) for unique_val in unique_values])
    total = N.sum(unique_counts)
    probs = unique_counts/total

    # remove probabilities==0 because log(0) = -Inf and causes problems.
    # This is ok because p*log(p) == 0*log(0) == 0 so removing these doesn't affect the final sum.
    probs = probs[probs>0] 

    return sum(-probs*N.log(probs))


def edit_distance(network1, network2):
    """Returns the edit distance between two networks.
    
    This is a good (but not the only one) metric for determining similarity
    between two networks.  
    
    """

    def inverse(edge):
        return (edge[1], edge[0])

    edges1 = copy(list(network1.edges))
    edges2 = copy(list(network2.edges))

    # Calculating distance:
    #   Add 1 to distance for every edge in one network but not in the other,
    #   EXCEPT, if inverse of edge exists in the other network, distance is 
    #   1 not 2 (this is because edit operations include add, remove and reverse)
    dist = 0
    for edge in edges1:
        if edge in edges2:
            edges2.remove(edge)
        elif inverse(edge) in edges2:
            dist += 1
            edges2.remove(inverse(edge))
        else:
            dist += 1
    
    # edges2 now contains all edges not in edges1.
    dist += len(edges2)

    return dist

def levenshtein(a,b):
    """Calculates the Levenshtein distance between *strings* a and b.

    from http://hetland.org/coding/python/levenshtein.py

    """
    n, m = len(a), len(b)
    if n > m:
        # Make sure n <= m, to use O(min(n,m)) space
        a,b = b,a
        n,m = m,n
        
    current = range(n+1)
    for i in range(1,m+1):
        previous, current = current, [i]+[0]*n
        for j in range(1,n+1):
            add, delete = previous[j]+1, current[j-1]+1
            change = previous[j-1]
            if a[j-1] != b[i-1]:
                change = change + 1
            current[j] = min(add, delete, change)
            
    return current[n]


def extended_property(func):
  """Function decorator for defining property attributes

  The decorated function is expected to return a dictionary
  containing one or more of the following pairs:

      * fget - function for getting attribute value
      * fset - function for setting attribute value
      * fdel - function for deleting attribute

  """
  return property(doc=func.__doc__, **func())

def lru_cache(maxsize):
    '''Decorator applying a least-recently-used cache with the given maximum size.

    Arguments to the cached function must be hashable.
    Cache performance statistics stored in f.hits and f.misses.

    from http://code.activestate.com/recipes/498245/
    '''
    def decorating_function(f):
        cache = {}              # mapping of args to results
        queue = deque()         # order that keys have been accessed
        refcount = {}           # number of times each key is in the access queue
        def wrapper(*args):
            
            # localize variable access (ugly but fast)
            _cache=cache; _len=len; _refcount=refcount; _maxsize=maxsize
            queue_append=queue.append; queue_popleft = queue.popleft

            # get cache entry or compute if not found
            try:
                result = _cache[args]
                wrapper.hits += 1
            except KeyError:
                result = _cache[args] = f(*args)
                wrapper.misses += 1

            # record that this key was recently accessed
            queue_append(args)
            _refcount[args] = _refcount.get(args, 0) + 1

            # Purge least recently accessed cache contents
            while _len(_cache) > _maxsize:
                k = queue_popleft()
                _refcount[k] -= 1
                if not _refcount[k]:
                    del _cache[k]
                    del _refcount[k]
    
            # Periodically compact the queue by duplicate keys
            if _len(queue) > _maxsize * 4:
                for i in [None] * _len(queue):
                    k = queue_popleft()
                    if _refcount[k] == 1:
                        queue_append(k)
                    else:
                        _refcount[k] -= 1
                assert len(queue) == len(cache) == len(refcount) == sum(refcount.itervalues())

            return result
        wrapper.__doc__ = f.__doc__
        wrapper.__name__ = f.__name__
        wrapper.hits = wrapper.misses = 0
        return wrapper
    return decorating_function