summaryrefslogtreecommitdiff
path: root/Lib/pstats.py
blob: d861413d4195f726ae6e67542c5479ee12f9f966 (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
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
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
"""Class for printing reports on profiled python code."""

# Written by James Roskind
# Based on prior profile module by Sjoerd Mullender...
#   which was hacked somewhat by: Guido van Rossum

# Copyright Disney Enterprises, Inc.  All Rights Reserved.
# Licensed to PSF under a Contributor Agreement
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
# either express or implied.  See the License for the specific language
# governing permissions and limitations under the License.


import sys
import os
import time
import marshal
import re
from functools import cmp_to_key

__all__ = ["Stats"]

class Stats:
    """This class is used for creating reports from data generated by the
    Profile class.  It is a "friend" of that class, and imports data either
    by direct access to members of Profile class, or by reading in a dictionary
    that was emitted (via marshal) from the Profile class.

    The big change from the previous Profiler (in terms of raw functionality)
    is that an "add()" method has been provided to combine Stats from
    several distinct profile runs.  Both the constructor and the add()
    method now take arbitrarily many file names as arguments.

    All the print methods now take an argument that indicates how many lines
    to print.  If the arg is a floating point number between 0 and 1.0, then
    it is taken as a decimal percentage of the available lines to be printed
    (e.g., .1 means print 10% of all available lines).  If it is an integer,
    it is taken to mean the number of lines of data that you wish to have
    printed.

    The sort_stats() method now processes some additional options (i.e., in
    addition to the old -1, 0, 1, or 2).  It takes an arbitrary number of
    quoted strings to select the sort order.  For example sort_stats('time',
    'name') sorts on the major key of 'internal function time', and on the
    minor key of 'the name of the function'.  Look at the two tables in
    sort_stats() and get_sort_arg_defs(self) for more examples.

    All methods return self, so you can string together commands like:
        Stats('foo', 'goo').strip_dirs().sort_stats('calls').\
                            print_stats(5).print_callers(5)
    """

    def __init__(self, *args, stream=None):
        self.stream = stream or sys.stdout
        if not len(args):
            arg = None
        else:
            arg = args[0]
            args = args[1:]
        self.init(arg)
        self.add(*args)

    def init(self, arg):
        self.all_callees = None  # calc only if needed
        self.files = []
        self.fcn_list = None
        self.total_tt = 0
        self.total_calls = 0
        self.prim_calls = 0
        self.max_name_len = 0
        self.top_level = set()
        self.stats = {}
        self.sort_arg_dict = {}
        self.load_stats(arg)
        try:
            self.get_top_level_stats()
        except Exception:
            print("Invalid timing data %s" %
                  (self.files[-1] if self.files else ''), file=self.stream)
            raise

    def load_stats(self, arg):
        if arg is None:
            self.stats = {}
            return
        elif isinstance(arg, str):
            with open(arg, 'rb') as f:
                self.stats = marshal.load(f)
            try:
                file_stats = os.stat(arg)
                arg = time.ctime(file_stats.st_mtime) + "    " + arg
            except:  # in case this is not unix
                pass
            self.files = [arg]
        elif hasattr(arg, 'create_stats'):
            arg.create_stats()
            self.stats = arg.stats
            arg.stats = {}
        if not self.stats:
            raise TypeError("Cannot create or construct a %r object from %r"
                            % (self.__class__, arg))
        return

    def get_top_level_stats(self):
        for func, (cc, nc, tt, ct, callers) in self.stats.items():
            self.total_calls += nc
            self.prim_calls  += cc
            self.total_tt    += tt
            if ("jprofile", 0, "profiler") in callers:
                self.top_level.add(func)
            if len(func_std_string(func)) > self.max_name_len:
                self.max_name_len = len(func_std_string(func))

    def add(self, *arg_list):
        if not arg_list:
            return self
        for item in reversed(arg_list):
            if type(self) != type(item):
                item = Stats(item)
            self.files += item.files
            self.total_calls += item.total_calls
            self.prim_calls += item.prim_calls
            self.total_tt += item.total_tt
            for func in item.top_level:
                self.top_level.add(func)

            if self.max_name_len < item.max_name_len:
                self.max_name_len = item.max_name_len

            self.fcn_list = None

            for func, stat in item.stats.items():
                if func in self.stats:
                    old_func_stat = self.stats[func]
                else:
                    old_func_stat = (0, 0, 0, 0, {},)
                self.stats[func] = add_func_stats(old_func_stat, stat)
        return self

    def dump_stats(self, filename):
        """Write the profile data to a file we know how to load back."""
        with open(filename, 'wb') as f:
            marshal.dump(self.stats, f)

    # list the tuple indices and directions for sorting,
    # along with some printable description
    sort_arg_dict_default = {
              "calls"     : (((1,-1),              ), "call count"),
              "ncalls"    : (((1,-1),              ), "call count"),
              "cumtime"   : (((3,-1),              ), "cumulative time"),
              "cumulative": (((3,-1),              ), "cumulative time"),
              "file"      : (((4, 1),              ), "file name"),
              "filename"  : (((4, 1),              ), "file name"),
              "line"      : (((5, 1),              ), "line number"),
              "module"    : (((4, 1),              ), "file name"),
              "name"      : (((6, 1),              ), "function name"),
              "nfl"       : (((6, 1),(4, 1),(5, 1),), "name/file/line"),
              "pcalls"    : (((0,-1),              ), "primitive call count"),
              "stdname"   : (((7, 1),              ), "standard name"),
              "time"      : (((2,-1),              ), "internal time"),
              "tottime"   : (((2,-1),              ), "internal time"),
              }

    def get_sort_arg_defs(self):
        """Expand all abbreviations that are unique."""
        if not self.sort_arg_dict:
            self.sort_arg_dict = dict = {}
            bad_list = {}
            for word, tup in self.sort_arg_dict_default.items():
                fragment = word
                while fragment:
                    if not fragment:
                        break
                    if fragment in dict:
                        bad_list[fragment] = 0
                        break
                    dict[fragment] = tup
                    fragment = fragment[:-1]
            for word in bad_list:
                del dict[word]
        return self.sort_arg_dict

    def sort_stats(self, *field):
        if not field:
            self.fcn_list = 0
            return self
        if len(field) == 1 and isinstance(field[0], int):
            # Be compatible with old profiler
            field = [ {-1: "stdname",
                       0:  "calls",
                       1:  "time",
                       2:  "cumulative"}[field[0]] ]

        sort_arg_defs = self.get_sort_arg_defs()
        sort_tuple = ()
        self.sort_type = ""
        connector = ""
        for word in field:
            sort_tuple = sort_tuple + sort_arg_defs[word][0]
            self.sort_type += connector + sort_arg_defs[word][1]
            connector = ", "

        stats_list = []
        for func, (cc, nc, tt, ct, callers) in self.stats.items():
            stats_list.append((cc, nc, tt, ct) + func +
                              (func_std_string(func), func))

        stats_list.sort(key=cmp_to_key(TupleComp(sort_tuple).compare))

        self.fcn_list = fcn_list = []
        for tuple in stats_list:
            fcn_list.append(tuple[-1])
        return self

    def reverse_order(self):
        if self.fcn_list:
            self.fcn_list.reverse()
        return self

    def strip_dirs(self):
        oldstats = self.stats
        self.stats = newstats = {}
        max_name_len = 0
        for func, (cc, nc, tt, ct, callers) in oldstats.items():
            newfunc = func_strip_path(func)
            if len(func_std_string(newfunc)) > max_name_len:
                max_name_len = len(func_std_string(newfunc))
            newcallers = {}
            for func2, caller in callers.items():
                newcallers[func_strip_path(func2)] = caller

            if newfunc in newstats:
                newstats[newfunc] = add_func_stats(
                                        newstats[newfunc],
                                        (cc, nc, tt, ct, newcallers))
            else:
                newstats[newfunc] = (cc, nc, tt, ct, newcallers)
        old_top = self.top_level
        self.top_level = new_top = set()
        for func in old_top:
            new_top.add(func_strip_path(func))

        self.max_name_len = max_name_len

        self.fcn_list = None
        self.all_callees = None
        return self

    def calc_callees(self):
        if self.all_callees:
            return
        self.all_callees = all_callees = {}
        for func, (cc, nc, tt, ct, callers) in self.stats.items():
            if not func in all_callees:
                all_callees[func] = {}
            for func2, caller in callers.items():
                if not func2 in all_callees:
                    all_callees[func2] = {}
                all_callees[func2][func]  = caller
        return

    #******************************************************************
    # The following functions support actual printing of reports
    #******************************************************************

    # Optional "amount" is either a line count, or a percentage of lines.

    def eval_print_amount(self, sel, list, msg):
        new_list = list
        if isinstance(sel, str):
            try:
                rex = re.compile(sel)
            except re.error:
                msg += "   <Invalid regular expression %r>\n" % sel
                return new_list, msg
            new_list = []
            for func in list:
                if rex.search(func_std_string(func)):
                    new_list.append(func)
        else:
            count = len(list)
            if isinstance(sel, float) and 0.0 <= sel < 1.0:
                count = int(count * sel + .5)
                new_list = list[:count]
            elif isinstance(sel, int) and 0 <= sel < count:
                count = sel
                new_list = list[:count]
        if len(list) != len(new_list):
            msg += "   List reduced from %r to %r due to restriction <%r>\n" % (
                len(list), len(new_list), sel)

        return new_list, msg

    def get_print_list(self, sel_list):
        width = self.max_name_len
        if self.fcn_list:
            stat_list = self.fcn_list[:]
            msg = "   Ordered by: " + self.sort_type + '\n'
        else:
            stat_list = list(self.stats.keys())
            msg = "   Random listing order was used\n"

        for selection in sel_list:
            stat_list, msg = self.eval_print_amount(selection, stat_list, msg)

        count = len(stat_list)

        if not stat_list:
            return 0, stat_list
        print(msg, file=self.stream)
        if count < len(self.stats):
            width = 0
            for func in stat_list:
                if  len(func_std_string(func)) > width:
                    width = len(func_std_string(func))
        return width+2, stat_list

    def print_stats(self, *amount):
        for filename in self.files:
            print(filename, file=self.stream)
        if self.files:
            print(file=self.stream)
        indent = ' ' * 8
        for func in self.top_level:
            print(indent, func_get_function_name(func), file=self.stream)

        print(indent, self.total_calls, "function calls", end=' ', file=self.stream)
        if self.total_calls != self.prim_calls:
            print("(%d primitive calls)" % self.prim_calls, end=' ', file=self.stream)
        print("in %.3f seconds" % self.total_tt, file=self.stream)
        print(file=self.stream)
        width, list = self.get_print_list(amount)
        if list:
            self.print_title()
            for func in list:
                self.print_line(func)
            print(file=self.stream)
            print(file=self.stream)
        return self

    def print_callees(self, *amount):
        width, list = self.get_print_list(amount)
        if list:
            self.calc_callees()

            self.print_call_heading(width, "called...")
            for func in list:
                if func in self.all_callees:
                    self.print_call_line(width, func, self.all_callees[func])
                else:
                    self.print_call_line(width, func, {})
            print(file=self.stream)
            print(file=self.stream)
        return self

    def print_callers(self, *amount):
        width, list = self.get_print_list(amount)
        if list:
            self.print_call_heading(width, "was called by...")
            for func in list:
                cc, nc, tt, ct, callers = self.stats[func]
                self.print_call_line(width, func, callers, "<-")
            print(file=self.stream)
            print(file=self.stream)
        return self

    def print_call_heading(self, name_size, column_title):
        print("Function ".ljust(name_size) + column_title, file=self.stream)
        # print sub-header only if we have new-style callers
        subheader = False
        for cc, nc, tt, ct, callers in self.stats.values():
            if callers:
                value = next(iter(callers.values()))
                subheader = isinstance(value, tuple)
                break
        if subheader:
            print(" "*name_size + "    ncalls  tottime  cumtime", file=self.stream)

    def print_call_line(self, name_size, source, call_dict, arrow="->"):
        print(func_std_string(source).ljust(name_size) + arrow, end=' ', file=self.stream)
        if not call_dict:
            print(file=self.stream)
            return
        clist = sorted(call_dict.keys())
        indent = ""
        for func in clist:
            name = func_std_string(func)
            value = call_dict[func]
            if isinstance(value, tuple):
                nc, cc, tt, ct = value
                if nc != cc:
                    substats = '%d/%d' % (nc, cc)
                else:
                    substats = '%d' % (nc,)
                substats = '%s %s %s  %s' % (substats.rjust(7+2*len(indent)),
                                             f8(tt), f8(ct), name)
                left_width = name_size + 1
            else:
                substats = '%s(%r) %s' % (name, value, f8(self.stats[func][3]))
                left_width = name_size + 3
            print(indent*left_width + substats, file=self.stream)
            indent = " "

    def print_title(self):
        print('   ncalls  tottime  percall  cumtime  percall', end=' ', file=self.stream)
        print('filename:lineno(function)', file=self.stream)

    def print_line(self, func):  # hack: should print percentages
        cc, nc, tt, ct, callers = self.stats[func]
        c = str(nc)
        if nc != cc:
            c = c + '/' + str(cc)
        print(c.rjust(9), end=' ', file=self.stream)
        print(f8(tt), end=' ', file=self.stream)
        if nc == 0:
            print(' '*8, end=' ', file=self.stream)
        else:
            print(f8(tt/nc), end=' ', file=self.stream)
        print(f8(ct), end=' ', file=self.stream)
        if cc == 0:
            print(' '*8, end=' ', file=self.stream)
        else:
            print(f8(ct/cc), end=' ', file=self.stream)
        print(func_std_string(func), file=self.stream)

class TupleComp:
    """This class provides a generic function for comparing any two tuples.
    Each instance records a list of tuple-indices (from most significant
    to least significant), and sort direction (ascending or decending) for
    each tuple-index.  The compare functions can then be used as the function
    argument to the system sort() function when a list of tuples need to be
    sorted in the instances order."""

    def __init__(self, comp_select_list):
        self.comp_select_list = comp_select_list

    def compare (self, left, right):
        for index, direction in self.comp_select_list:
            l = left[index]
            r = right[index]
            if l < r:
                return -direction
            if l > r:
                return direction
        return 0


#**************************************************************************
# func_name is a triple (file:string, line:int, name:string)

def func_strip_path(func_name):
    filename, line, name = func_name
    return os.path.basename(filename), line, name

def func_get_function_name(func):
    return func[2]

def func_std_string(func_name): # match what old profile produced
    if func_name[:2] == ('~', 0):
        # special case for built-in functions
        name = func_name[2]
        if name.startswith('<') and name.endswith('>'):
            return '{%s}' % name[1:-1]
        else:
            return name
    else:
        return "%s:%d(%s)" % func_name

#**************************************************************************
# The following functions combine statists for pairs functions.
# The bulk of the processing involves correctly handling "call" lists,
# such as callers and callees.
#**************************************************************************

def add_func_stats(target, source):
    """Add together all the stats for two profile entries."""
    cc, nc, tt, ct, callers = source
    t_cc, t_nc, t_tt, t_ct, t_callers = target
    return (cc+t_cc, nc+t_nc, tt+t_tt, ct+t_ct,
              add_callers(t_callers, callers))

def add_callers(target, source):
    """Combine two caller lists in a single list."""
    new_callers = {}
    for func, caller in target.items():
        new_callers[func] = caller
    for func, caller in source.items():
        if func in new_callers:
            if isinstance(caller, tuple):
                # format used by cProfile
                new_callers[func] = tuple([i[0] + i[1] for i in
                                           zip(caller, new_callers[func])])
            else:
                # format used by profile
                new_callers[func] += caller
        else:
            new_callers[func] = caller
    return new_callers

def count_calls(callers):
    """Sum the caller statistics to get total number of calls received."""
    nc = 0
    for calls in callers.values():
        nc += calls
    return nc

#**************************************************************************
# The following functions support printing of reports
#**************************************************************************

def f8(x):
    return "%8.3f" % x

#**************************************************************************
# Statistics browser added by ESR, April 2001
#**************************************************************************

if __name__ == '__main__':
    import cmd
    try:
        import readline
    except ImportError:
        pass

    class ProfileBrowser(cmd.Cmd):
        def __init__(self, profile=None):
            cmd.Cmd.__init__(self)
            self.prompt = "% "
            self.stats = None
            self.stream = sys.stdout
            if profile is not None:
                self.do_read(profile)

        def generic(self, fn, line):
            args = line.split()
            processed = []
            for term in args:
                try:
                    processed.append(int(term))
                    continue
                except ValueError:
                    pass
                try:
                    frac = float(term)
                    if frac > 1 or frac < 0:
                        print("Fraction argument must be in [0, 1]", file=self.stream)
                        continue
                    processed.append(frac)
                    continue
                except ValueError:
                    pass
                processed.append(term)
            if self.stats:
                getattr(self.stats, fn)(*processed)
            else:
                print("No statistics object is loaded.", file=self.stream)
            return 0
        def generic_help(self):
            print("Arguments may be:", file=self.stream)
            print("* An integer maximum number of entries to print.", file=self.stream)
            print("* A decimal fractional number between 0 and 1, controlling", file=self.stream)
            print("  what fraction of selected entries to print.", file=self.stream)
            print("* A regular expression; only entries with function names", file=self.stream)
            print("  that match it are printed.", file=self.stream)

        def do_add(self, line):
            if self.stats:
                try:
                    self.stats.add(line)
                except IOError as e:
                    print("Failed to load statistics for %s: %s" % (line, e), file=self.stream)
            else:
                print("No statistics object is loaded.", file=self.stream)
            return 0
        def help_add(self):
            print("Add profile info from given file to current statistics object.", file=self.stream)

        def do_callees(self, line):
            return self.generic('print_callees', line)
        def help_callees(self):
            print("Print callees statistics from the current stat object.", file=self.stream)
            self.generic_help()

        def do_callers(self, line):
            return self.generic('print_callers', line)
        def help_callers(self):
            print("Print callers statistics from the current stat object.", file=self.stream)
            self.generic_help()

        def do_EOF(self, line):
            print("", file=self.stream)
            return 1
        def help_EOF(self):
            print("Leave the profile brower.", file=self.stream)

        def do_quit(self, line):
            return 1
        def help_quit(self):
            print("Leave the profile brower.", file=self.stream)

        def do_read(self, line):
            if line:
                try:
                    self.stats = Stats(line)
                except OSError as err:
                    print(err.args[1], file=self.stream)
                    return
                except Exception as err:
                    print(err.__class__.__name__ + ':', err, file=self.stream)
                    return
                self.prompt = line + "% "
            elif len(self.prompt) > 2:
                line = self.prompt[:-2]
                self.do_read(line)
            else:
                print("No statistics object is current -- cannot reload.", file=self.stream)
            return 0
        def help_read(self):
            print("Read in profile data from a specified file.", file=self.stream)
            print("Without argument, reload the current file.", file=self.stream)

        def do_reverse(self, line):
            if self.stats:
                self.stats.reverse_order()
            else:
                print("No statistics object is loaded.", file=self.stream)
            return 0
        def help_reverse(self):
            print("Reverse the sort order of the profiling report.", file=self.stream)

        def do_sort(self, line):
            if not self.stats:
                print("No statistics object is loaded.", file=self.stream)
                return
            abbrevs = self.stats.get_sort_arg_defs()
            if line and all((x in abbrevs) for x in line.split()):
                self.stats.sort_stats(*line.split())
            else:
                print("Valid sort keys (unique prefixes are accepted):", file=self.stream)
                for (key, value) in Stats.sort_arg_dict_default.items():
                    print("%s -- %s" % (key, value[1]), file=self.stream)
            return 0
        def help_sort(self):
            print("Sort profile data according to specified keys.", file=self.stream)
            print("(Typing `sort' without arguments lists valid keys.)", file=self.stream)
        def complete_sort(self, text, *args):
            return [a for a in Stats.sort_arg_dict_default if a.startswith(text)]

        def do_stats(self, line):
            return self.generic('print_stats', line)
        def help_stats(self):
            print("Print statistics from the current stat object.", file=self.stream)
            self.generic_help()

        def do_strip(self, line):
            if self.stats:
                self.stats.strip_dirs()
            else:
                print("No statistics object is loaded.", file=self.stream)
        def help_strip(self):
            print("Strip leading path information from filenames in the report.", file=self.stream)

        def help_help(self):
            print("Show help for a given command.", file=self.stream)

        def postcmd(self, stop, line):
            if stop:
                return stop
            return None

    if len(sys.argv) > 1:
        initprofile = sys.argv[1]
    else:
        initprofile = None
    try:
        browser = ProfileBrowser(initprofile)
        for profile in sys.argv[2:]:
            browser.do_add(profile)
        print("Welcome to the profile statistics browser.", file=browser.stream)
        browser.cmdloop()
        print("Goodbye.", file=browser.stream)
    except KeyboardInterrupt:
        pass

# That's all, folks.