WebKitTools/CodeCoverage/cov.py

   1 # Copyright (C) 2004, 2005, 2006 Nathaniel Smith
   2 # Copyright (C) 2006, 2007 Holger Hans Peter Freyther
   3 #
   4 # Redistribution and use in source and binary forms, with or without
   5 # modification, are permitted provided that the following conditions
   6 # are met:
   7 #
   8 # 1.  Redistributions of source code must retain the above copyright
   9 #     notice, this list of conditions and the following disclaimer.
  10 # 2.  Redistributions in binary form must reproduce the above copyright
  11 #     notice, this list of conditions and the following disclaimer in the
  12 #     documentation and/or other materials provided with the distribution.
  13 # 3.  Neither the name of Apple Computer, Inc. ("Apple") nor the names of
  14 #     its contributors may be used to endorse or promote products derived
  15 #     from this software without specific prior written permission.
  16 #
  17 # THIS SOFTWARE IS PROVIDED BY APPLE AND ITS CONTRIBUTORS "AS IS" AND ANY
  18 # EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
  19 # WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  20 # DISCLAIMED. IN NO EVENT SHALL APPLE OR ITS CONTRIBUTORS BE LIABLE FOR ANY
  21 # DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
  22 # (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
  23 # LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
  24 # ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  25 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  26 # THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  27
  28 import csv
  29 import time
  30 import os.path
  31 import shutil
  32
  33 def analyze_coverage(possible_gcov_files, source_files, runid, data_dir, base):
  34
  35     if not os.path.exists(data_dir):
  36         os.makedirs(data_dir)
  37
  38     output = open(os.path.join(data_dir, runid + ".csv"), "w")
  39     w = csv.writer(output)
  40     # First row: id and time
  41     w.writerow([runid, time.time()])
  42
  43     results = scan_gcov_files(possible_gcov_files, source_files)
  44     annotated_dir = os.path.join(data_dir, runid + ".annotated")
  45     if os.path.exists(annotated_dir):
  46         shutil.rmtree(annotated_dir)
  47
  48     keys = results.keys()
  49     keys.sort()
  50     for path in keys:
  51         (total, covered, annotated_data) = results[path]
  52         path = path[path.find(base)+len(base):]
  53         # Rest of the rows: filename, total_lines, covered_lines
  54         w.writerow([path, total, covered])
  55
  56         if path[:1] == "/":
  57             path = path[1:]
  58         annotated_path = os.path.join(annotated_dir, path)
  59         try:
  60             os.makedirs(os.path.dirname(annotated_path))
  61         except OSError:
  62             pass
  63         a = open(annotated_path, "w")
  64         a.write(annotated_data)
  65         a.close()
  66
  67
  68 # zecke's rewrite
  69 STATE_NOT_CODE = -1
  70 STATE_NOT_SEEN = -2
  71 STATE_TEST_CODE = -3
  72
  73 def find_gcov(f, possible_gcovs):
  74     """
  75     Find .gcov files that could be of interest for us
  76     """
  77     try:
  78         return possible_gcovs[f]
  79     except:
  80         return []
  81
  82
  83 def parse_source_file(file):
  84     """
  85     Parse one source file and return a list of lines
  86     """
  87     f_source_list = []
  88     init_state = STATE_NOT_SEEN
  89     in_test_code = False
  90     nesting = 0
  91
  92     for line in open(file, "r"):
  93         code = line.split(":", 2)[-1]
  94         if not in_test_code and code.startswith("#ifdef BUILD_UNIT_TESTS"):
  95             in_test_code = 1
  96         if in_test_code and code.startswith("#if"):
  97             nesting += 1
  98         if in_test_code and code.startswith("#endif"):
  99             nesting -= 1
 100             if not nesting:
 101                 in_test_code = True
 102         if in_test_code:
 103             init_state = STATE_TEST_CODE
 104         else:
 105             init_state = STATE_NOT_SEEN
 106         f_source_list.append([init_state, line.split(":", 1)[1]])
 107
 108     return f_source_list
 109
 110 # Runner-up, 3rd annual "write Python that looks like Perl" competition,
 111 # Well, not really.  It doesn't even use regexps.
 112 # He is right so I'm cleaning it up (zecke)
 113 def scan_gcov_files(possible_gcov_files, source_files):
 114     """Takes a list of gcov filenames and a list of source filenames.
 115
 116     The gcov files should have names of the form foo.o##foo.cc.gcov, as
 117     created by 'gcov -l'.
 118
 119     Returns a dict mapping source filenames to tuples
 120       (total_lines, tested_lines, gcov_annotated_source)
 121     which are a number, a number, and a very long string, respectively.
 122
 123     The fun bit is that we merge .gcov output generated by different object
 124     files; this way we can provide accurate information for header files and
 125     for monotone's current unit test system."""
 126     results = {}
 127     for f in source_files:
 128         possible_gcovs = find_gcov(f, possible_gcov_files)
 129         base_name = os.path.splitext(os.path.basename(f))[0]
 130         if len(possible_gcovs) == 0:
 131             print "No gcov files found for: '%s' but it was compiled" % f
 132             continue
 133
 134             (garbage,extension) = os.path.splitext(f)
 135             if extension  in [".cc", ".c", ".moc", ".cpp", ".cxx", ".m", ".mm"]:
 136                 lines = open(f, "r").readlines()
 137                 results[f] = (len(lines), 0, "".join(lines))
 138             continue
 139         elif len(possible_gcovs) > 1:
 140             print "More than one gcov file for %s %d" % (f,len(possible_gcovs))
 141         base_gcov_lines = parse_source_file(possible_gcovs[0])
 142
 143         # Now we will try hard to merge the results with others
 144         # Our requirement is that we have the same amount of lines as
 145         # as the original file
 146         for cov_file in possible_gcovs:
 147             lines = open(cov_file, "r").readlines()
 148
 149             # e.g. with phonon we have visualisation.h and we can not know
 150             # which header file (foldername) it is refering to. This is a gcov
 151             # limitation and i have no workaround yet. We just hope we will pick
 152             # the right header file...
 153             if len(lines) != len(base_gcov_lines):
 154                 print "Error Base: %s and Target: %s have different amount of lines" % (possible_gcovs[0],cov_file)
 155                 continue
 156
 157             # now do the merging of the file. If it has the same basename
 158             # and the same number of lines things might work out
 159             # In the future take a look at the header of the file
 160             i = 0
 161             for line in lines:
 162                 accumulator = base_gcov_lines[i]
 163                 if accumulator[0] != STATE_TEST_CODE:
 164                     info = line.split(":", 1)[0]
 165                     if info.endswith("-"):
 166                         if accumulator[0] == STATE_NOT_SEEN:
 167                             accumulator[0] = STATE_NOT_CODE
 168                     else:
 169                         if info.endswith("#"):
 170                             num = 0
 171                         else:
 172                             num = int(info)
 173                         if accumulator[0] in (STATE_NOT_SEEN, STATE_NOT_CODE):
 174                             accumulator[0] = 0
 175                         accumulator[0] += num
 176                 i += 1
 177
 178         # post processing of ths file
 179         (total_lines, total_covered) = (0, 0)
 180         annotated_lines = []
 181         for state, line in base_gcov_lines:
 182             if state == STATE_NOT_SEEN:
 183                 desc = "?????"
 184             elif state == STATE_TEST_CODE:
 185                 desc = "+"
 186             elif state == STATE_NOT_CODE:
 187                 desc = "-"
 188             elif state == 0:
 189                 desc = "#####"
 190                 total_lines += 1
 191             else:
 192                 desc = str(state)
 193                 total_lines += 1
 194                 total_covered += 1
 195             annotated_lines.append(":".join([desc.rjust(9), line]))
 196         results[f] = (total_lines, total_covered, "".join(annotated_lines))
 197     return results
 198
 199
 200
 201     return results