gitstats

   1 #!/usr/bin/env python
   2 # Copyright (c) 2007-2012 Heikki Hokkanen <hoxu@users.sf.net> & others (see doc/author.txt)
   3 # GPLv2 / GPLv3
   4 import datetime
   5 import getopt
   6 import glob
   7 import os
   8 import pickle
   9 import platform
  10 import re
  11 import shutil
  12 import subprocess
  13 import sys
  14 import time
  15 import zlib
  16
  17 os.environ['LC_ALL'] = 'C'
  18
  19 GNUPLOT_COMMON = 'set terminal png transparent size 640,240\nset size 1.0,1.0\n'
  20 ON_LINUX = (platform.system() == 'Linux')
  21 WEEKDAYS = ('Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun')
  22
  23 exectime_internal = 0.0
  24 exectime_external = 0.0
  25 time_start = time.time()
  26
  27 # By default, gnuplot is searched from path, but can be overridden with the
  28 # environment variable "GNUPLOT"
  29 gnuplot_cmd = 'gnuplot'
  30 if 'GNUPLOT' in os.environ:
  31         gnuplot_cmd = os.environ['GNUPLOT']
  32
  33 conf = {
  34         'max_domains': 10,
  35         'max_ext_length': 10,
  36         'style': 'gitstats.css',
  37         'max_authors': 20,
  38         'authors_top': 5,
  39         'commit_begin': '',
  40         'commit_end': 'HEAD',
  41         'linear_linestats': 1,
  42         'project_name': '',
  43 }
  44
  45 def getpipeoutput(cmds, quiet = False):
  46         global exectime_external
  47         start = time.time()
  48         if not quiet and ON_LINUX and os.isatty(1):
  49                 print '>> ' + ' | '.join(cmds),
  50                 sys.stdout.flush()
  51         p0 = subprocess.Popen(cmds[0], stdout = subprocess.PIPE, shell = True)
  52         p = p0
  53         for x in cmds[1:]:
  54                 p = subprocess.Popen(x, stdin = p0.stdout, stdout = subprocess.PIPE, shell = True)
  55                 p0 = p
  56         output = p.communicate()[0]
  57         end = time.time()
  58         if not quiet:
  59                 if ON_LINUX and os.isatty(1):
  60                         print '\r',
  61                 print '[%.5f] >> %s' % (end - start, ' | '.join(cmds))
  62         exectime_external += (end - start)
  63         return output.rstrip('\n')
  64
  65 def getcommitrange(defaultrange = 'HEAD', end_only = False):
  66         if len(conf['commit_end']) > 0:
  67                 if end_only or len(conf['commit_begin']) == 0:
  68                         return conf['commit_end']
  69                 return '%s..%s' % (conf['commit_begin'], conf['commit_end'])
  70         return defaultrange
  71
  72 def getkeyssortedbyvalues(dict):
  73         return map(lambda el : el[1], sorted(map(lambda el : (el[1], el[0]), dict.items())))
  74
  75 # dict['author'] = { 'commits': 512 } - ...key(dict, 'commits')
  76 def getkeyssortedbyvaluekey(d, key):
  77         return map(lambda el : el[1], sorted(map(lambda el : (d[el][key], el), d.keys())))
  78
  79 def getstatsummarycounts(line):
  80         numbers = re.findall('\d+', line)
  81         if   len(numbers) == 1:
  82                 # neither insertions nor deletions: may probably only happen for "0 files changed"
  83                 numbers.append(0);
  84                 numbers.append(0);
  85         elif len(numbers) == 2 and line.find('(+)') != -1:
  86                 numbers.append(0);    # only insertions were printed on line
  87         elif len(numbers) == 2 and line.find('(-)') != -1:
  88                 numbers.insert(1, 0); # only deletions were printed on line
  89         return numbers
  90
  91 VERSION = 0
  92 def getversion():
  93         global VERSION
  94         if VERSION == 0:
  95                 VERSION = getpipeoutput(["git rev-parse --short %s" % getcommitrange('HEAD')]).split('\n')[0]
  96         return VERSION
  97
  98 def getgitversion():
  99         return getpipeoutput(['git --version']).split('\n')[0]
 100
 101 def getgnuplotversion():
 102         return getpipeoutput(['gnuplot --version']).split('\n')[0]
 103
 104 class DataCollector:
 105         """Manages data collection from a revision control repository."""
 106         def __init__(self):
 107                 self.stamp_created = time.time()
 108                 self.cache = {}
 109                 self.total_authors = 0
 110                 self.activity_by_hour_of_day = {} # hour -> commits
 111                 self.activity_by_day_of_week = {} # day -> commits
 112                 self.activity_by_month_of_year = {} # month [1-12] -> commits
 113                 self.activity_by_hour_of_week = {} # weekday -> hour -> commits
 114                 self.activity_by_hour_of_day_busiest = 0
 115                 self.activity_by_hour_of_week_busiest = 0
 116                 self.activity_by_year_week = {} # yy_wNN -> commits
 117                 self.activity_by_year_week_peak = 0
 118
 119                 self.authors = {} # name -> {commits, first_commit_stamp, last_commit_stamp, last_active_day, active_days, lines_added, lines_removed}
 120
 121                 self.total_commits = 0
 122                 self.total_files = 0
 123                 self.authors_by_commits = 0
 124
 125                 # domains
 126                 self.domains = {} # domain -> commits
 127
 128                 # author of the month
 129                 self.author_of_month = {} # month -> author -> commits
 130                 self.author_of_year = {} # year -> author -> commits
 131                 self.commits_by_month = {} # month -> commits
 132                 self.commits_by_year = {} # year -> commits
 133                 self.lines_added_by_month = {} # month -> lines added
 134                 self.lines_added_by_year = {} # year -> lines added
 135                 self.lines_removed_by_month = {} # month -> lines removed
 136                 self.lines_removed_by_year = {} # year -> lines removed
 137                 self.first_commit_stamp = 0
 138                 self.last_commit_stamp = 0
 139                 self.last_active_day = None
 140                 self.active_days = set()
 141
 142                 # lines
 143                 self.total_lines = 0
 144                 self.total_lines_added = 0
 145                 self.total_lines_removed = 0
 146
 147                 # size
 148                 self.total_size = 0
 149
 150                 # timezone
 151                 self.commits_by_timezone = {} # timezone -> commits
 152
 153                 # tags
 154                 self.tags = {}
 155
 156                 self.files_by_stamp = {} # stamp -> files
 157
 158                 # extensions
 159                 self.extensions = {} # extension -> files, lines
 160
 161                 # line statistics
 162                 self.changes_by_date = {} # stamp -> { files, ins, del }
 163
 164         ##
 165         # This should be the main function to extract data from the repository.
 166         def collect(self, dir):
 167                 self.dir = dir
 168                 if len(conf['project_name']) == 0:
 169                         self.projectname = os.path.basename(os.path.abspath(dir))
 170                 else:
 171                         self.projectname = conf['project_name']
 172
 173         ##
 174         # Load cacheable data
 175         def loadCache(self, cachefile):
 176                 if not os.path.exists(cachefile):
 177                         return
 178                 print 'Loading cache...'
 179                 f = open(cachefile, 'rb')
 180                 try:
 181                         self.cache = pickle.loads(zlib.decompress(f.read()))
 182                 except:
 183                         # temporary hack to upgrade non-compressed caches
 184                         f.seek(0)
 185                         self.cache = pickle.load(f)
 186                 f.close()
 187
 188         ##
 189         # Produce any additional statistics from the extracted data.
 190         def refine(self):
 191                 pass
 192
 193         ##
 194         # : get a dictionary of author
 195         def getAuthorInfo(self, author):
 196                 return None
 197
 198         def getActivityByDayOfWeek(self):
 199                 return {}
 200
 201         def getActivityByHourOfDay(self):
 202                 return {}
 203
 204         # : get a dictionary of domains
 205         def getDomainInfo(self, domain):
 206                 return None
 207
 208         ##
 209         # Get a list of authors
 210         def getAuthors(self):
 211                 return []
 212
 213         def getFirstCommitDate(self):
 214                 return datetime.datetime.now()
 215
 216         def getLastCommitDate(self):
 217                 return datetime.datetime.now()
 218
 219         def getStampCreated(self):
 220                 return self.stamp_created
 221
 222         def getTags(self):
 223                 return []
 224
 225         def getTotalAuthors(self):
 226                 return -1
 227
 228         def getTotalCommits(self):
 229                 return -1
 230
 231         def getTotalFiles(self):
 232                 return -1
 233
 234         def getTotalLOC(self):
 235                 return -1
 236
 237         ##
 238         # Save cacheable data
 239         def saveCache(self, cachefile):
 240                 print 'Saving cache...'
 241                 f = open(cachefile, 'wb')
 242                 #pickle.dump(self.cache, f)
 243                 data = zlib.compress(pickle.dumps(self.cache))
 244                 f.write(data)
 245                 f.close()
 246
 247 class GitDataCollector(DataCollector):
 248         def collect(self, dir):
 249                 DataCollector.collect(self, dir)
 250
 251                 self.total_authors += int(getpipeoutput(['git shortlog -s %s' % getcommitrange(), 'wc -l']))
 252                 #self.total_lines = int(getoutput('git-ls-files -z |xargs -0 cat |wc -l'))
 253
 254                 # tags
 255                 lines = getpipeoutput(['git show-ref --tags']).split('\n')
 256                 for line in lines:
 257                         if len(line) == 0:
 258                                 continue
 259                         (hash, tag) = line.split(' ')
 260
 261                         tag = tag.replace('refs/tags/', '')
 262                         output = getpipeoutput(['git log "%s" --pretty=format:"%%at %%aN" -n 1' % hash])
 263                         if len(output) > 0:
 264                                 parts = output.split(' ')
 265                                 stamp = 0
 266                                 try:
 267                                         stamp = int(parts[0])
 268                                 except ValueError:
 269                                         stamp = 0
 270                                 self.tags[tag] = { 'stamp': stamp, 'hash' : hash, 'date' : datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), 'commits': 0, 'authors': {} }
 271
 272                 # collect info on tags, starting from latest
 273                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), self.tags.items()))))
 274                 prev = None
 275                 for tag in reversed(tags_sorted_by_date_desc):
 276                         cmd = 'git shortlog -s "%s"' % tag
 277                         if prev != None:
 278                                 cmd += ' "^%s"' % prev
 279                         output = getpipeoutput([cmd])
 280                         if len(output) == 0:
 281                                 continue
 282                         prev = tag
 283                         for line in output.split('\n'):
 284                                 parts = re.split('\s+', line, 2)
 285                                 commits = int(parts[1])
 286                                 author = parts[2]
 287                                 self.tags[tag]['commits'] += commits
 288                                 self.tags[tag]['authors'][author] = commits
 289
 290                 # Collect revision statistics
 291                 # Outputs "<stamp> <date> <time> <timezone> <author> '<' <mail> '>'"
 292                 lines = getpipeoutput(['git rev-list --pretty=format:"%%at %%ai %%aN <%%aE>" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).split('\n')
 293                 for line in lines:
 294                         parts = line.split(' ', 4)
 295                         author = ''
 296                         try:
 297                                 stamp = int(parts[0])
 298                         except ValueError:
 299                                 stamp = 0
 300                         timezone = parts[3]
 301                         author, mail = parts[4].split('<', 1)
 302                         author = author.rstrip()
 303                         mail = mail.rstrip('>')
 304                         domain = '?'
 305                         if mail.find('@') != -1:
 306                                 domain = mail.rsplit('@', 1)[1]
 307                         date = datetime.datetime.fromtimestamp(float(stamp))
 308
 309                         # First and last commit stamp (may be in any order because of cherry-picking and patches)
 310                         if stamp > self.last_commit_stamp:
 311                                 self.last_commit_stamp = stamp
 312                         if self.first_commit_stamp == 0 or stamp < self.first_commit_stamp:
 313                                 self.first_commit_stamp = stamp
 314
 315                         # activity
 316                         # hour
 317                         hour = date.hour
 318                         self.activity_by_hour_of_day[hour] = self.activity_by_hour_of_day.get(hour, 0) + 1
 319                         # most active hour?
 320                         if self.activity_by_hour_of_day[hour] > self.activity_by_hour_of_day_busiest:
 321                                 self.activity_by_hour_of_day_busiest = self.activity_by_hour_of_day[hour]
 322
 323                         # day of week
 324                         day = date.weekday()
 325                         self.activity_by_day_of_week[day] = self.activity_by_day_of_week.get(day, 0) + 1
 326
 327                         # domain stats
 328                         if domain not in self.domains:
 329                                 self.domains[domain] = {}
 330                         # commits
 331                         self.domains[domain]['commits'] = self.domains[domain].get('commits', 0) + 1
 332
 333                         # hour of week
 334                         if day not in self.activity_by_hour_of_week:
 335                                 self.activity_by_hour_of_week[day] = {}
 336                         self.activity_by_hour_of_week[day][hour] = self.activity_by_hour_of_week[day].get(hour, 0) + 1
 337                         # most active hour?
 338                         if self.activity_by_hour_of_week[day][hour] > self.activity_by_hour_of_week_busiest:
 339                                 self.activity_by_hour_of_week_busiest = self.activity_by_hour_of_week[day][hour]
 340
 341                         # month of year
 342                         month = date.month
 343                         self.activity_by_month_of_year[month] = self.activity_by_month_of_year.get(month, 0) + 1
 344
 345                         # yearly/weekly activity
 346                         yyw = date.strftime('%Y-%W')
 347                         self.activity_by_year_week[yyw] = self.activity_by_year_week.get(yyw, 0) + 1
 348                         if self.activity_by_year_week_peak < self.activity_by_year_week[yyw]:
 349                                 self.activity_by_year_week_peak = self.activity_by_year_week[yyw]
 350
 351                         # author stats
 352                         if author not in self.authors:
 353                                 self.authors[author] = {}
 354                         # commits, note again that commits may be in any date order because of cherry-picking and patches
 355                         if 'last_commit_stamp' not in self.authors[author]:
 356                                 self.authors[author]['last_commit_stamp'] = stamp
 357                         if stamp > self.authors[author]['last_commit_stamp']:
 358                                 self.authors[author]['last_commit_stamp'] = stamp
 359                         if 'first_commit_stamp' not in self.authors[author]:
 360                                 self.authors[author]['first_commit_stamp'] = stamp
 361                         if stamp < self.authors[author]['first_commit_stamp']:
 362                                 self.authors[author]['first_commit_stamp'] = stamp
 363
 364                         # author of the month/year
 365                         yymm = date.strftime('%Y-%m')
 366                         if yymm in self.author_of_month:
 367                                 self.author_of_month[yymm][author] = self.author_of_month[yymm].get(author, 0) + 1
 368                         else:
 369                                 self.author_of_month[yymm] = {}
 370                                 self.author_of_month[yymm][author] = 1
 371                         self.commits_by_month[yymm] = self.commits_by_month.get(yymm, 0) + 1
 372
 373                         yy = date.year
 374                         if yy in self.author_of_year:
 375                                 self.author_of_year[yy][author] = self.author_of_year[yy].get(author, 0) + 1
 376                         else:
 377                                 self.author_of_year[yy] = {}
 378                                 self.author_of_year[yy][author] = 1
 379                         self.commits_by_year[yy] = self.commits_by_year.get(yy, 0) + 1
 380
 381                         # authors: active days
 382                         yymmdd = date.strftime('%Y-%m-%d')
 383                         if 'last_active_day' not in self.authors[author]:
 384                                 self.authors[author]['last_active_day'] = yymmdd
 385                                 self.authors[author]['active_days'] = set([yymmdd])
 386                         elif yymmdd != self.authors[author]['last_active_day']:
 387                                 self.authors[author]['last_active_day'] = yymmdd
 388                                 self.authors[author]['active_days'].add(yymmdd)
 389
 390                         # project: active days
 391                         if yymmdd != self.last_active_day:
 392                                 self.last_active_day = yymmdd
 393                                 self.active_days.add(yymmdd)
 394
 395                         # timezone
 396                         self.commits_by_timezone[timezone] = self.commits_by_timezone.get(timezone, 0) + 1
 397
 398                 # TODO Optimize this, it's the worst bottleneck
 399                 # outputs "<stamp> <files>" for each revision
 400                 revlines = getpipeoutput(['git rev-list --pretty=format:"%%at %%T" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).strip().split('\n')
 401                 lines = []
 402                 for revline in revlines:
 403                         time, rev = revline.split(' ')
 404                         linecount = self.getFilesInCommit(rev)
 405                         lines.append('%d %d' % (int(time), linecount))
 406
 407                 self.total_commits += len(lines)
 408                 for line in lines:
 409                         parts = line.split(' ')
 410                         if len(parts) != 2:
 411                                 continue
 412                         (stamp, files) = parts[0:2]
 413                         try:
 414                                 self.files_by_stamp[int(stamp)] = int(files)
 415                         except ValueError:
 416                                 print 'Warning: failed to parse line "%s"' % line
 417
 418                 # extensions and size of files
 419                 lines = getpipeoutput(['git ls-tree -r -l -z %s' % getcommitrange('HEAD', end_only = True)]).split('\000')
 420                 for line in lines:
 421                         if len(line) == 0:
 422                                 continue
 423                         parts = re.split('\s+', line, 5)
 424                         if parts[0] == '160000' and parts[3] == '-':
 425                                 # skip submodules
 426                                 continue
 427                         sha1 = parts[2]
 428                         size = int(parts[3])
 429                         fullpath = parts[4]
 430
 431                         self.total_size += size
 432                         self.total_files += 1
 433
 434                         filename = fullpath.split('/')[-1] # strip directories
 435                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 436                                 ext = ''
 437                         else:
 438                                 ext = filename[(filename.rfind('.') + 1):]
 439                         if len(ext) > conf['max_ext_length']:
 440                                 ext = ''
 441
 442                         if ext not in self.extensions:
 443                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 444
 445                         self.extensions[ext]['files'] += 1
 446                         try:
 447                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 448                         except:
 449                                 print 'Warning: Could not count lines for file "%s"' % line
 450
 451                 # line statistics
 452                 # outputs:
 453                 #  N files changed, N insertions (+), N deletions(-)
 454                 # <stamp> <author>
 455                 self.changes_by_date = {} # stamp -> { files, ins, del }
 456                 # computation of lines of code by date is better done
 457                 # on a linear history.
 458                 extra = ''
 459                 if conf['linear_linestats']:
 460                         extra = '--first-parent -m'
 461                 lines = getpipeoutput(['git log --shortstat %s --pretty=format:"%%at %%aN" %s' % (extra, getcommitrange('HEAD'))]).split('\n')
 462                 lines.reverse()
 463                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 464                 author = None
 465                 for line in lines:
 466                         if len(line) == 0:
 467                                 continue
 468
 469                         # <stamp> <author>
 470                         if re.search('files? changed', line) == None:
 471                                 pos = line.find(' ')
 472                                 if pos != -1:
 473                                         try:
 474                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 475                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 476
 477                                                 date = datetime.datetime.fromtimestamp(stamp)
 478                                                 yymm = date.strftime('%Y-%m')
 479                                                 self.lines_added_by_month[yymm] = self.lines_added_by_month.get(yymm, 0) + inserted
 480                                                 self.lines_removed_by_month[yymm] = self.lines_removed_by_month.get(yymm, 0) + deleted
 481
 482                                                 yy = date.year
 483                                                 self.lines_added_by_year[yy] = self.lines_added_by_year.get(yy,0) + inserted
 484                                                 self.lines_removed_by_year[yy] = self.lines_removed_by_year.get(yy, 0) + deleted
 485
 486                                                 files, inserted, deleted = 0, 0, 0
 487                                         except ValueError:
 488                                                 print 'Warning: unexpected line "%s"' % line
 489                                 else:
 490                                         print 'Warning: unexpected line "%s"' % line
 491                         else:
 492                                 numbers = getstatsummarycounts(line)
 493
 494                                 if len(numbers) == 3:
 495                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 496                                         total_lines += inserted
 497                                         total_lines -= deleted
 498                                         self.total_lines_added += inserted
 499                                         self.total_lines_removed += deleted
 500
 501                                 else:
 502                                         print 'Warning: failed to handle line "%s"' % line
 503                                         (files, inserted, deleted) = (0, 0, 0)
 504                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 505                 self.total_lines = total_lines
 506
 507                 # Per-author statistics
 508
 509                 # defined for stamp, author only if author commited at this timestamp.
 510                 self.changes_by_date_by_author = {} # stamp -> author -> lines_added
 511
 512                 # Similar to the above, but never use --first-parent
 513                 # (we need to walk through every commit to know who
 514                 # committed what, not just through mainline)
 515                 lines = getpipeoutput(['git log --shortstat --date-order --pretty=format:"%%at %%aN" %s' % (getcommitrange('HEAD'))]).split('\n')
 516                 lines.reverse()
 517                 files = 0; inserted = 0; deleted = 0
 518                 author = None
 519                 stamp = 0
 520                 for line in lines:
 521                         if len(line) == 0:
 522                                 continue
 523
 524                         # <stamp> <author>
 525                         if re.search('files? changed', line) == None:
 526                                 pos = line.find(' ')
 527                                 if pos != -1:
 528                                         try:
 529                                                 oldstamp = stamp
 530                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 531                                                 if oldstamp > stamp:
 532                                                         # clock skew, keep old timestamp to avoid having ugly graph
 533                                                         stamp = oldstamp
 534                                                 if author not in self.authors:
 535                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0, 'commits' : 0}
 536                                                 self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 537                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 538                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 539                                                 if stamp not in self.changes_by_date_by_author:
 540                                                         self.changes_by_date_by_author[stamp] = {}
 541                                                 if author not in self.changes_by_date_by_author[stamp]:
 542                                                         self.changes_by_date_by_author[stamp][author] = {}
 543                                                 self.changes_by_date_by_author[stamp][author]['lines_added'] = self.authors[author]['lines_added']
 544                                                 self.changes_by_date_by_author[stamp][author]['commits'] = self.authors[author]['commits']
 545                                                 files, inserted, deleted = 0, 0, 0
 546                                         except ValueError:
 547                                                 print 'Warning: unexpected line "%s"' % line
 548                                 else:
 549                                         print 'Warning: unexpected line "%s"' % line
 550                         else:
 551                                 numbers = getstatsummarycounts(line);
 552
 553                                 if len(numbers) == 3:
 554                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 555                                 else:
 556                                         print 'Warning: failed to handle line "%s"' % line
 557                                         (files, inserted, deleted) = (0, 0, 0)
 558
 559         def refine(self):
 560                 # authors
 561                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 562                 self.authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 563                 self.authors_by_commits.reverse() # most first
 564                 for i, name in enumerate(self.authors_by_commits):
 565                         self.authors[name]['place_by_commits'] = i + 1
 566
 567                 for name in self.authors.keys():
 568                         a = self.authors[name]
 569                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 570                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 571                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 572                         delta = date_last - date_first
 573                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 574                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 575                         a['timedelta'] = delta
 576                         if 'lines_added' not in a: a['lines_added'] = 0
 577                         if 'lines_removed' not in a: a['lines_removed'] = 0
 578
 579         def getActiveDays(self):
 580                 return self.active_days
 581
 582         def getActivityByDayOfWeek(self):
 583                 return self.activity_by_day_of_week
 584
 585         def getActivityByHourOfDay(self):
 586                 return self.activity_by_hour_of_day
 587
 588         def getAuthorInfo(self, author):
 589                 return self.authors[author]
 590
 591         def getAuthors(self, limit = None):
 592                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 593                 res.reverse()
 594                 return res[:limit]
 595
 596         def getCommitDeltaDays(self):
 597                 return (self.last_commit_stamp / 86400 - self.first_commit_stamp / 86400) + 1
 598
 599         def getDomainInfo(self, domain):
 600                 return self.domains[domain]
 601
 602         def getDomains(self):
 603                 return self.domains.keys()
 604
 605         def getFilesInCommit(self, rev):
 606                 try:
 607                         res = self.cache['files_in_tree'][rev]
 608                 except:
 609                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 610                         if 'files_in_tree' not in self.cache:
 611                                 self.cache['files_in_tree'] = {}
 612                         self.cache['files_in_tree'][rev] = res
 613
 614                 return res
 615
 616         def getFirstCommitDate(self):
 617                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 618
 619         def getLastCommitDate(self):
 620                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 621
 622         def getLinesInBlob(self, sha1):
 623                 try:
 624                         res = self.cache['lines_in_blob'][sha1]
 625                 except:
 626                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 627                         if 'lines_in_blob' not in self.cache:
 628                                 self.cache['lines_in_blob'] = {}
 629                         self.cache['lines_in_blob'][sha1] = res
 630                 return res
 631
 632         def getTags(self):
 633                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 634                 return lines.split('\n')
 635
 636         def getTagDate(self, tag):
 637                 return self.revToDate('tags/' + tag)
 638
 639         def getTotalAuthors(self):
 640                 return self.total_authors
 641
 642         def getTotalCommits(self):
 643                 return self.total_commits
 644
 645         def getTotalFiles(self):
 646                 return self.total_files
 647
 648         def getTotalLOC(self):
 649                 return self.total_lines
 650
 651         def getTotalSize(self):
 652                 return self.total_size
 653
 654         def revToDate(self, rev):
 655                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 656                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 657
 658 class ReportCreator:
 659         """Creates the actual report based on given data."""
 660         def __init__(self):
 661                 pass
 662
 663         def create(self, data, path):
 664                 self.data = data
 665                 self.path = path
 666
 667 def html_linkify(text):
 668         return text.lower().replace(' ', '_')
 669
 670 def html_header(level, text):
 671         name = html_linkify(text)
 672         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 673
 674 class HTMLReportCreator(ReportCreator):
 675         def create(self, data, path):
 676                 ReportCreator.create(self, data, path)
 677                 self.title = data.projectname
 678
 679                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 680                 binarypath = os.path.dirname(os.path.abspath(__file__))
 681                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 682                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 683                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 684                         for base in basedirs:
 685                                 src = base + '/' + file
 686                                 if os.path.exists(src):
 687                                         shutil.copyfile(src, path + '/' + file)
 688                                         break
 689                         else:
 690                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 691
 692                 f = open(path + "/index.html", 'w')
 693                 format = '%Y-%m-%d %H:%M:%S'
 694                 self.printHeader(f)
 695
 696                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 697
 698                 self.printNav(f)
 699
 700                 f.write('<dl>')
 701                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 702                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 703                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s), %s, %s</dd>' % (getversion(), getgitversion(), getgnuplotversion()))
 704                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 705                 f.write('<dt>Age</dt><dd>%d days, %d active days (%3.2f%%)</dd>' % (data.getCommitDeltaDays(), len(data.getActiveDays()), (100.0 * len(data.getActiveDays()) / data.getCommitDeltaDays())))
 706                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 707                 f.write('<dt>Total Lines of Code</dt><dd>%s (%d added, %d removed)</dd>' % (data.getTotalLOC(), data.total_lines_added, data.total_lines_removed))
 708                 f.write('<dt>Total Commits</dt><dd>%s (average %.1f commits per active day, %.1f per all days)</dd>' % (data.getTotalCommits(), float(data.getTotalCommits()) / len(data.getActiveDays()), float(data.getTotalCommits()) / data.getCommitDeltaDays()))
 709                 f.write('<dt>Authors</dt><dd>%s (average %.1f commits per author)</dd>' % (data.getTotalAuthors(), (1.0 * data.getTotalCommits()) / data.getTotalAuthors()))
 710                 f.write('</dl>')
 711
 712                 f.write('</body>\n</html>')
 713                 f.close()
 714
 715                 ###
 716                 # Activity
 717                 f = open(path + '/activity.html', 'w')
 718                 self.printHeader(f)
 719                 f.write('<h1>Activity</h1>')
 720                 self.printNav(f)
 721
 722                 #f.write('<h2>Last 30 days</h2>')
 723
 724                 #f.write('<h2>Last 12 months</h2>')
 725
 726                 # Weekly activity
 727                 WEEKS = 32
 728                 f.write(html_header(2, 'Weekly activity'))
 729                 f.write('<p>Last %d weeks</p>' % WEEKS)
 730
 731                 # generate weeks to show (previous N weeks from now)
 732                 now = datetime.datetime.now()
 733                 deltaweek = datetime.timedelta(7)
 734                 weeks = []
 735                 stampcur = now
 736                 for i in range(0, WEEKS):
 737                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 738                         stampcur -= deltaweek
 739
 740                 # top row: commits & bar
 741                 f.write('<table class="noborders"><tr>')
 742                 for i in range(0, WEEKS):
 743                         commits = 0
 744                         if weeks[i] in data.activity_by_year_week:
 745                                 commits = data.activity_by_year_week[weeks[i]]
 746
 747                         percentage = 0
 748                         if weeks[i] in data.activity_by_year_week:
 749                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 750                         height = max(1, int(200 * percentage))
 751                         f.write('<td style="text-align: center; vertical-align: bottom">%d<div style="display: block; background-color: red; width: 20px; height: %dpx"></div></td>' % (commits, height))
 752
 753                 # bottom row: year/week
 754                 f.write('</tr><tr>')
 755                 for i in range(0, WEEKS):
 756                         f.write('<td>%s</td>' % (WEEKS - i))
 757                 f.write('</tr></table>')
 758
 759                 # Hour of Day
 760                 f.write(html_header(2, 'Hour of Day'))
 761                 hour_of_day = data.getActivityByHourOfDay()
 762                 f.write('<table><tr><th>Hour</th>')
 763                 for i in range(0, 24):
 764                         f.write('<th>%d</th>' % i)
 765                 f.write('</tr>\n<tr><th>Commits</th>')
 766                 fp = open(path + '/hour_of_day.dat', 'w')
 767                 for i in range(0, 24):
 768                         if i in hour_of_day:
 769                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 770                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 771                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 772                         else:
 773                                 f.write('<td>0</td>')
 774                                 fp.write('%d 0\n' % i)
 775                 fp.close()
 776                 f.write('</tr>\n<tr><th>%</th>')
 777                 totalcommits = data.getTotalCommits()
 778                 for i in range(0, 24):
 779                         if i in hour_of_day:
 780                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 781                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 782                         else:
 783                                 f.write('<td>0.00</td>')
 784                 f.write('</tr></table>')
 785                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 786                 fg = open(path + '/hour_of_day.dat', 'w')
 787                 for i in range(0, 24):
 788                         if i in hour_of_day:
 789                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 790                         else:
 791                                 fg.write('%d 0\n' % (i + 1))
 792                 fg.close()
 793
 794                 # Day of Week
 795                 f.write(html_header(2, 'Day of Week'))
 796                 day_of_week = data.getActivityByDayOfWeek()
 797                 f.write('<div class="vtable"><table>')
 798                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 799                 fp = open(path + '/day_of_week.dat', 'w')
 800                 for d in range(0, 7):
 801                         commits = 0
 802                         if d in day_of_week:
 803                                 commits = day_of_week[d]
 804                         fp.write('%d %s %d\n' % (d + 1, WEEKDAYS[d], commits))
 805                         f.write('<tr>')
 806                         f.write('<th>%s</th>' % (WEEKDAYS[d]))
 807                         if d in day_of_week:
 808                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 809                         else:
 810                                 f.write('<td>0</td>')
 811                         f.write('</tr>')
 812                 f.write('</table></div>')
 813                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 814                 fp.close()
 815
 816                 # Hour of Week
 817                 f.write(html_header(2, 'Hour of Week'))
 818                 f.write('<table>')
 819
 820                 f.write('<tr><th>Weekday</th>')
 821                 for hour in range(0, 24):
 822                         f.write('<th>%d</th>' % (hour))
 823                 f.write('</tr>')
 824
 825                 for weekday in range(0, 7):
 826                         f.write('<tr><th>%s</th>' % (WEEKDAYS[weekday]))
 827                         for hour in range(0, 24):
 828                                 try:
 829                                         commits = data.activity_by_hour_of_week[weekday][hour]
 830                                 except KeyError:
 831                                         commits = 0
 832                                 if commits != 0:
 833                                         f.write('<td')
 834                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 835                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 836                                         f.write('>%d</td>' % commits)
 837                                 else:
 838                                         f.write('<td></td>')
 839                         f.write('</tr>')
 840
 841                 f.write('</table>')
 842
 843                 # Month of Year
 844                 f.write(html_header(2, 'Month of Year'))
 845                 f.write('<div class="vtable"><table>')
 846                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 847                 fp = open (path + '/month_of_year.dat', 'w')
 848                 for mm in range(1, 13):
 849                         commits = 0
 850                         if mm in data.activity_by_month_of_year:
 851                                 commits = data.activity_by_month_of_year[mm]
 852                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 853                         fp.write('%d %d\n' % (mm, commits))
 854                 fp.close()
 855                 f.write('</table></div>')
 856                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 857
 858                 # Commits by year/month
 859                 f.write(html_header(2, 'Commits by year/month'))
 860                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th><th>Lines added</th><th>Lines removed</th></tr>')
 861                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 862                         f.write('<tr><td>%s</td><td>%d</td><td>%d</td><td>%d</td></tr>' % (yymm, data.commits_by_month.get(yymm,0), data.lines_added_by_month.get(yymm,0), data.lines_removed_by_month.get(yymm,0)))
 863                 f.write('</table></div>')
 864                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 865                 fg = open(path + '/commits_by_year_month.dat', 'w')
 866                 for yymm in sorted(data.commits_by_month.keys()):
 867                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 868                 fg.close()
 869
 870                 # Commits by year
 871                 f.write(html_header(2, 'Commits by Year'))
 872                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th><th>Lines added</th><th>Lines removed</th></tr>')
 873                 for yy in reversed(sorted(data.commits_by_year.keys())):
 874                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d</td><td>%d</td></tr>' % (yy, data.commits_by_year.get(yy,0), (100.0 * data.commits_by_year.get(yy,0)) / data.getTotalCommits(), data.lines_added_by_year.get(yy,0), data.lines_removed_by_year.get(yy,0)))
 875                 f.write('</table></div>')
 876                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 877                 fg = open(path + '/commits_by_year.dat', 'w')
 878                 for yy in sorted(data.commits_by_year.keys()):
 879                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 880                 fg.close()
 881
 882                 # Commits by timezone
 883                 f.write(html_header(2, 'Commits by Timezone'))
 884                 f.write('<table><tr>')
 885                 f.write('<th>Timezone</th><th>Commits</th>')
 886                 max_commits_on_tz = max(data.commits_by_timezone.values())
 887                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 888                         commits = data.commits_by_timezone[i]
 889                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 890                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 891                 f.write('</tr></table>')
 892
 893                 f.write('</body></html>')
 894                 f.close()
 895
 896                 ###
 897                 # Authors
 898                 f = open(path + '/authors.html', 'w')
 899                 self.printHeader(f)
 900
 901                 f.write('<h1>Authors</h1>')
 902                 self.printNav(f)
 903
 904                 # Authors :: List of authors
 905                 f.write(html_header(2, 'List of Authors'))
 906
 907                 f.write('<table class="authors sortable" id="authors">')
 908                 f.write('<tr><th>Author</th><th>Commits (%)</th><th>+ lines</th><th>- lines</th><th>First commit</th><th>Last commit</th><th class="unsortable">Age</th><th>Active days</th><th># by commits</th></tr>')
 909                 for author in data.getAuthors(conf['max_authors']):
 910                         info = data.getAuthorInfo(author)
 911                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d</td><td>%d</td><td>%s</td><td>%s</td><td>%s</td><td>%d</td><td>%d</td></tr>' % (author, info['commits'], info['commits_frac'], info['lines_added'], info['lines_removed'], info['date_first'], info['date_last'], info['timedelta'], len(info['active_days']), info['place_by_commits']))
 912                 f.write('</table>')
 913
 914                 allauthors = data.getAuthors()
 915                 if len(allauthors) > conf['max_authors']:
 916                         rest = allauthors[conf['max_authors']:]
 917                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 918
 919                 f.write(html_header(2, 'Cumulated Added Lines of Code per Author'))
 920                 f.write('<img src="lines_of_code_by_author.png" alt="Lines of code per Author" />')
 921                 if len(allauthors) > conf['max_authors']:
 922                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 923
 924                 f.write(html_header(2, 'Commits per Author'))
 925                 f.write('<img src="commits_by_author.png" alt="Commits per Author" />')
 926                 if len(allauthors) > conf['max_authors']:
 927                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 928
 929                 fgl = open(path + '/lines_of_code_by_author.dat', 'w')
 930                 fgc = open(path + '/commits_by_author.dat', 'w')
 931
 932                 lines_by_authors = {} # cumulated added lines by
 933                 # author. to save memory,
 934                 # changes_by_date_by_author[stamp][author] is defined
 935                 # only at points where author commits.
 936                 # lines_by_authors allows us to generate all the
 937                 # points in the .dat file.
 938
 939                 # Don't rely on getAuthors to give the same order each
 940                 # time. Be robust and keep the list in a variable.
 941                 commits_by_authors = {} # cumulated added lines by
 942
 943                 self.authors_to_plot = data.getAuthors(conf['max_authors'])
 944                 for author in self.authors_to_plot:
 945                         lines_by_authors[author] = 0
 946                         commits_by_authors[author] = 0
 947                 for stamp in sorted(data.changes_by_date_by_author.keys()):
 948                         fgl.write('%d' % stamp)
 949                         fgc.write('%d' % stamp)
 950                         for author in self.authors_to_plot:
 951                                 if author in data.changes_by_date_by_author[stamp].keys():
 952                                         lines_by_authors[author] = data.changes_by_date_by_author[stamp][author]['lines_added']
 953                                         commits_by_authors[author] = data.changes_by_date_by_author[stamp][author]['commits']
 954                                 fgl.write(' %d' % lines_by_authors[author])
 955                                 fgc.write(' %d' % commits_by_authors[author])
 956                         fgl.write('\n')
 957                         fgc.write('\n')
 958                 fgl.close()
 959                 fgc.close()
 960
 961                 # Authors :: Author of Month
 962                 f.write(html_header(2, 'Author of Month'))
 963                 f.write('<table class="sortable" id="aom">')
 964                 f.write('<tr><th>Month</th><th>Author</th><th>Commits (%%)</th><th class="unsortable">Next top %d</th><th>Number of authors</th></tr>' % conf['authors_top'])
 965                 for yymm in reversed(sorted(data.author_of_month.keys())):
 966                         authordict = data.author_of_month[yymm]
 967                         authors = getkeyssortedbyvalues(authordict)
 968                         authors.reverse()
 969                         commits = data.author_of_month[yymm][authors[0]]
 970                         next = ', '.join(authors[1:conf['authors_top']+1])
 971                         f.write('<tr><td>%s</td><td>%s</td><td>%d (%.2f%% of %d)</td><td>%s</td><td>%d</td></tr>' % (yymm, authors[0], commits, (100.0 * commits) / data.commits_by_month[yymm], data.commits_by_month[yymm], next, len(authors)))
 972
 973                 f.write('</table>')
 974
 975                 f.write(html_header(2, 'Author of Year'))
 976                 f.write('<table class="sortable" id="aoy"><tr><th>Year</th><th>Author</th><th>Commits (%%)</th><th class="unsortable">Next top %d</th><th>Number of authors</th></tr>' % conf['authors_top'])
 977                 for yy in reversed(sorted(data.author_of_year.keys())):
 978                         authordict = data.author_of_year[yy]
 979                         authors = getkeyssortedbyvalues(authordict)
 980                         authors.reverse()
 981                         commits = data.author_of_year[yy][authors[0]]
 982                         next = ', '.join(authors[1:conf['authors_top']+1])
 983                         f.write('<tr><td>%s</td><td>%s</td><td>%d (%.2f%% of %d)</td><td>%s</td><td>%d</td></tr>' % (yy, authors[0], commits, (100.0 * commits) / data.commits_by_year[yy], data.commits_by_year[yy], next, len(authors)))
 984                 f.write('</table>')
 985
 986                 # Domains
 987                 f.write(html_header(2, 'Commits by Domains'))
 988                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
 989                 domains_by_commits.reverse() # most first
 990                 f.write('<div class="vtable"><table>')
 991                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
 992                 fp = open(path + '/domains.dat', 'w')
 993                 n = 0
 994                 for domain in domains_by_commits:
 995                         if n == conf['max_domains']:
 996                                 break
 997                         commits = 0
 998                         n += 1
 999                         info = data.getDomainInfo(domain)
1000                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
1001                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
1002                 f.write('</table></div>')
1003                 f.write('<img src="domains.png" alt="Commits by Domains" />')
1004                 fp.close()
1005
1006                 f.write('</body></html>')
1007                 f.close()
1008
1009                 ###
1010                 # Files
1011                 f = open(path + '/files.html', 'w')
1012                 self.printHeader(f)
1013                 f.write('<h1>Files</h1>')
1014                 self.printNav(f)
1015
1016                 f.write('<dl>\n')
1017                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
1018                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1019                 try:
1020                         f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % (float(data.getTotalSize()) / data.getTotalFiles()))
1021                 except ZeroDivisionError:
1022                         pass
1023                 f.write('</dl>\n')
1024
1025                 # Files :: File count by date
1026                 f.write(html_header(2, 'File count by date'))
1027
1028                 # use set to get rid of duplicate/unnecessary entries
1029                 files_by_date = set()
1030                 for stamp in sorted(data.files_by_stamp.keys()):
1031                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1032
1033                 fg = open(path + '/files_by_date.dat', 'w')
1034                 for line in sorted(list(files_by_date)):
1035                         fg.write('%s\n' % line)
1036                 #for stamp in sorted(data.files_by_stamp.keys()):
1037                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1038                 fg.close()
1039
1040                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
1041
1042                 #f.write('<h2>Average file size by date</h2>')
1043
1044                 # Files :: Extensions
1045                 f.write(html_header(2, 'Extensions'))
1046                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
1047                 for ext in sorted(data.extensions.keys()):
1048                         files = data.extensions[ext]['files']
1049                         lines = data.extensions[ext]['lines']
1050                         try:
1051                                 loc_percentage = (100.0 * lines) / data.getTotalLOC()
1052                         except ZeroDivisionError:
1053                                 loc_percentage = 0
1054                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d (%.2f%%)</td><td>%d</td></tr>' % (ext, files, (100.0 * files) / data.getTotalFiles(), lines, loc_percentage, lines / files))
1055                 f.write('</table>')
1056
1057                 f.write('</body></html>')
1058                 f.close()
1059
1060                 ###
1061                 # Lines
1062                 f = open(path + '/lines.html', 'w')
1063                 self.printHeader(f)
1064                 f.write('<h1>Lines</h1>')
1065                 self.printNav(f)
1066
1067                 f.write('<dl>\n')
1068                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1069                 f.write('</dl>\n')
1070
1071                 f.write(html_header(2, 'Lines of Code'))
1072                 f.write('<img src="lines_of_code.png" />')
1073
1074                 fg = open(path + '/lines_of_code.dat', 'w')
1075                 for stamp in sorted(data.changes_by_date.keys()):
1076                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
1077                 fg.close()
1078
1079                 f.write('</body></html>')
1080                 f.close()
1081
1082                 ###
1083                 # tags.html
1084                 f = open(path + '/tags.html', 'w')
1085                 self.printHeader(f)
1086                 f.write('<h1>Tags</h1>')
1087                 self.printNav(f)
1088
1089                 f.write('<dl>')
1090                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
1091                 if len(data.tags) > 0:
1092                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
1093                 f.write('</dl>')
1094
1095                 f.write('<table class="tags">')
1096                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
1097                 # sort the tags by date desc
1098                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
1099                 for tag in tags_sorted_by_date_desc:
1100                         authorinfo = []
1101                         self.authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
1102                         for i in reversed(self.authors_by_commits):
1103                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
1104                         f.write('<tr><td>%s</td><td>%s</td><td>%d</td><td>%s</td></tr>' % (tag, data.tags[tag]['date'], data.tags[tag]['commits'], ', '.join(authorinfo)))
1105                 f.write('</table>')
1106
1107                 f.write('</body></html>')
1108                 f.close()
1109
1110                 self.createGraphs(path)
1111
1112         def createGraphs(self, path):
1113                 print 'Generating graphs...'
1114
1115                 # hour of day
1116                 f = open(path + '/hour_of_day.plot', 'w')
1117                 f.write(GNUPLOT_COMMON)
1118                 f.write(
1119 """
1120 set output 'hour_of_day.png'
1121 unset key
1122 set xrange [0.5:24.5]
1123 set xtics 4
1124 set grid y
1125 set ylabel "Commits"
1126 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
1127 """)
1128                 f.close()
1129
1130                 # day of week
1131                 f = open(path + '/day_of_week.plot', 'w')
1132                 f.write(GNUPLOT_COMMON)
1133                 f.write(
1134 """
1135 set output 'day_of_week.png'
1136 unset key
1137 set xrange [0.5:7.5]
1138 set xtics 1
1139 set grid y
1140 set ylabel "Commits"
1141 plot 'day_of_week.dat' using 1:3:(0.5):xtic(2) w boxes fs solid
1142 """)
1143                 f.close()
1144
1145                 # Domains
1146                 f = open(path + '/domains.plot', 'w')
1147                 f.write(GNUPLOT_COMMON)
1148                 f.write(
1149 """
1150 set output 'domains.png'
1151 unset key
1152 unset xtics
1153 set yrange [0:]
1154 set grid y
1155 set ylabel "Commits"
1156 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
1157 """)
1158                 f.close()
1159
1160                 # Month of Year
1161                 f = open(path + '/month_of_year.plot', 'w')
1162                 f.write(GNUPLOT_COMMON)
1163                 f.write(
1164 """
1165 set output 'month_of_year.png'
1166 unset key
1167 set xrange [0.5:12.5]
1168 set xtics 1
1169 set grid y
1170 set ylabel "Commits"
1171 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
1172 """)
1173                 f.close()
1174
1175                 # commits_by_year_month
1176                 f = open(path + '/commits_by_year_month.plot', 'w')
1177                 f.write(GNUPLOT_COMMON)
1178                 f.write(
1179 """
1180 set output 'commits_by_year_month.png'
1181 unset key
1182 set xdata time
1183 set timefmt "%Y-%m"
1184 set format x "%Y-%m"
1185 set xtics rotate
1186 set bmargin 5
1187 set grid y
1188 set ylabel "Commits"
1189 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1190 """)
1191                 f.close()
1192
1193                 # commits_by_year
1194                 f = open(path + '/commits_by_year.plot', 'w')
1195                 f.write(GNUPLOT_COMMON)
1196                 f.write(
1197 """
1198 set output 'commits_by_year.png'
1199 unset key
1200 set xtics 1 rotate
1201 set grid y
1202 set ylabel "Commits"
1203 set yrange [0:]
1204 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1205 """)
1206                 f.close()
1207
1208                 # Files by date
1209                 f = open(path + '/files_by_date.plot', 'w')
1210                 f.write(GNUPLOT_COMMON)
1211                 f.write(
1212 """
1213 set output 'files_by_date.png'
1214 unset key
1215 set xdata time
1216 set timefmt "%Y-%m-%d"
1217 set format x "%Y-%m-%d"
1218 set grid y
1219 set ylabel "Files"
1220 set xtics rotate
1221 set ytics autofreq
1222 set bmargin 6
1223 plot 'files_by_date.dat' using 1:2 w steps
1224 """)
1225                 f.close()
1226
1227                 # Lines of Code
1228                 f = open(path + '/lines_of_code.plot', 'w')
1229                 f.write(GNUPLOT_COMMON)
1230                 f.write(
1231 """
1232 set output 'lines_of_code.png'
1233 unset key
1234 set xdata time
1235 set timefmt "%s"
1236 set format x "%Y-%m-%d"
1237 set grid y
1238 set ylabel "Lines"
1239 set xtics rotate
1240 set bmargin 6
1241 plot 'lines_of_code.dat' using 1:2 w lines
1242 """)
1243                 f.close()
1244
1245                 # Lines of Code Added per author
1246                 f = open(path + '/lines_of_code_by_author.plot', 'w')
1247                 f.write(GNUPLOT_COMMON)
1248                 f.write(
1249 """
1250 set terminal png transparent size 640,480
1251 set output 'lines_of_code_by_author.png'
1252 set key left top
1253 set xdata time
1254 set timefmt "%s"
1255 set format x "%Y-%m-%d"
1256 set grid y
1257 set ylabel "Lines"
1258 set xtics rotate
1259 set bmargin 6
1260 plot """
1261 )
1262                 i = 1
1263                 plots = []
1264                 for a in self.authors_to_plot:
1265                         i = i + 1
1266                         plots.append("""'lines_of_code_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1267                 f.write(", ".join(plots))
1268                 f.write('\n')
1269
1270                 f.close()
1271
1272                 # Commits per author
1273                 f = open(path + '/commits_by_author.plot', 'w')
1274                 f.write(GNUPLOT_COMMON)
1275                 f.write(
1276 """
1277 set terminal png transparent size 640,480
1278 set output 'commits_by_author.png'
1279 set key left top
1280 set xdata time
1281 set timefmt "%s"
1282 set format x "%Y-%m-%d"
1283 set grid y
1284 set ylabel "Commits"
1285 set xtics rotate
1286 set bmargin 6
1287 plot """
1288 )
1289                 i = 1
1290                 plots = []
1291                 for a in self.authors_to_plot:
1292                         i = i + 1
1293                         plots.append("""'commits_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1294                 f.write(", ".join(plots))
1295                 f.write('\n')
1296
1297                 f.close()
1298
1299                 os.chdir(path)
1300                 files = glob.glob(path + '/*.plot')
1301                 for f in files:
1302                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1303                         if len(out) > 0:
1304                                 print out
1305
1306         def printHeader(self, f, title = ''):
1307                 f.write(
1308 """<?xml version="1.0" encoding="UTF-8"?>
1309 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1310 <html xmlns="http://www.w3.org/1999/xhtml">
1311 <head>
1312         <title>GitStats - %s</title>
1313         <link rel="stylesheet" href="%s" type="text/css" />
1314         <meta name="generator" content="GitStats %s" />
1315         <script type="text/javascript" src="sortable.js"></script>
1316 </head>
1317 <body>
1318 """ % (self.title, conf['style'], getversion()))
1319
1320         def printNav(self, f):
1321                 f.write("""
1322 <div class="nav">
1323 <ul>
1324 <li><a href="index.html">General</a></li>
1325 <li><a href="activity.html">Activity</a></li>
1326 <li><a href="authors.html">Authors</a></li>
1327 <li><a href="files.html">Files</a></li>
1328 <li><a href="lines.html">Lines</a></li>
1329 <li><a href="tags.html">Tags</a></li>
1330 </ul>
1331 </div>
1332 """)
1333
1334
1335 class GitStats:
1336         def run(self, args_orig):
1337                 optlist, args = getopt.getopt(args_orig, 'c:')
1338                 for o,v in optlist:
1339                         if o == '-c':
1340                                 key, value = v.split('=', 1)
1341                                 if key not in conf:
1342                                         raise KeyError('no such key "%s" in config' % key)
1343                                 if isinstance(conf[key], int):
1344                                         conf[key] = int(value)
1345                                 else:
1346                                         conf[key] = value
1347
1348                 if len(args) < 2:
1349                         print """
1350 Usage: gitstats [options] <gitpath..> <outputpath>
1351
1352 Options:
1353 -c key=value     Override configuration value
1354
1355 Default config values:
1356 %s
1357 """ % conf
1358                         sys.exit(0)
1359
1360                 outputpath = os.path.abspath(args[-1])
1361                 rundir = os.getcwd()
1362
1363                 try:
1364                         os.makedirs(outputpath)
1365                 except OSError:
1366                         pass
1367                 if not os.path.isdir(outputpath):
1368                         print 'FATAL: Output path is not a directory or does not exist'
1369                         sys.exit(1)
1370
1371                 print 'Output path: %s' % outputpath
1372                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1373
1374                 data = GitDataCollector()
1375                 data.loadCache(cachefile)
1376
1377                 for gitpath in args[0:-1]:
1378                         print 'Git path: %s' % gitpath
1379
1380                         os.chdir(gitpath)
1381
1382                         print 'Collecting data...'
1383                         data.collect(gitpath)
1384
1385                 print 'Refining data...'
1386                 data.saveCache(cachefile)
1387                 data.refine()
1388
1389                 os.chdir(rundir)
1390
1391                 print 'Generating report...'
1392                 report = HTMLReportCreator()
1393                 report.create(data, outputpath)
1394
1395                 time_end = time.time()
1396                 exectime_internal = time_end - time_start
1397                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1398                 if sys.stdin.isatty():
1399                         print 'You may now run:'
1400                         print
1401                         print '   sensible-browser \'%s\'' % os.path.join(outputpath, 'index.html').replace("'", "'\\''")
1402                         print
1403
1404 if __name__=='__main__':
1405         g = GitStats()
1406         g.run(sys.argv[1:])
1407