gitstats

   1 #!/usr/bin/env python
   2 # Copyright (c) 2007-2010 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 GNUPLOT_COMMON = 'set terminal png transparent\nset size 1.0,0.5\n'
  18 ON_LINUX = (platform.system() == 'Linux')
  19
  20 exectime_internal = 0.0
  21 exectime_external = 0.0
  22 time_start = time.time()
  23
  24 # By default, gnuplot is searched from path, but can be overridden with the
  25 # environment variable "GNUPLOT"
  26 gnuplot_cmd = 'gnuplot'
  27 if 'GNUPLOT' in os.environ:
  28         gnuplot_cmd = os.environ['GNUPLOT']
  29
  30 conf = {
  31         'max_domains': 10,
  32         'max_ext_length': 10,
  33         'style': 'gitstats.css',
  34         'max_authors': 20,
  35 }
  36
  37 def getpipeoutput(cmds, quiet = False):
  38         global exectime_external
  39         start = time.time()
  40         if not quiet and ON_LINUX and os.isatty(1):
  41                 print '>> ' + ' | '.join(cmds),
  42                 sys.stdout.flush()
  43         p0 = subprocess.Popen(cmds[0], stdout = subprocess.PIPE, shell = True)
  44         p = p0
  45         for x in cmds[1:]:
  46                 p = subprocess.Popen(x, stdin = p0.stdout, stdout = subprocess.PIPE, shell = True)
  47                 p0 = p
  48         output = p.communicate()[0]
  49         end = time.time()
  50         if not quiet:
  51                 if ON_LINUX and os.isatty(1):
  52                         print '\r',
  53                 print '[%.5f] >> %s' % (end - start, ' | '.join(cmds))
  54         exectime_external += (end - start)
  55         return output.rstrip('\n')
  56
  57 def getkeyssortedbyvalues(dict):
  58         return map(lambda el : el[1], sorted(map(lambda el : (el[1], el[0]), dict.items())))
  59
  60 # dict['author'] = { 'commits': 512 } - ...key(dict, 'commits')
  61 def getkeyssortedbyvaluekey(d, key):
  62         return map(lambda el : el[1], sorted(map(lambda el : (d[el][key], el), d.keys())))
  63
  64 VERSION = 0
  65 def getversion():
  66         global VERSION
  67         if VERSION == 0:
  68                 VERSION = getpipeoutput(["git rev-parse --short HEAD"]).split('\n')[0]
  69         return VERSION
  70
  71 class DataCollector:
  72         """Manages data collection from a revision control repository."""
  73         def __init__(self):
  74                 self.stamp_created = time.time()
  75                 self.cache = {}
  76
  77         ##
  78         # This should be the main function to extract data from the repository.
  79         def collect(self, dir):
  80                 self.dir = dir
  81                 self.projectname = os.path.basename(os.path.abspath(dir))
  82
  83         ##
  84         # Load cacheable data
  85         def loadCache(self, cachefile):
  86                 if not os.path.exists(cachefile):
  87                         return
  88                 print 'Loading cache...'
  89                 f = open(cachefile)
  90                 try:
  91                         self.cache = pickle.loads(zlib.decompress(f.read()))
  92                 except:
  93                         # temporary hack to upgrade non-compressed caches
  94                         f.seek(0)
  95                         self.cache = pickle.load(f)
  96                 f.close()
  97
  98         ##
  99         # Produce any additional statistics from the extracted data.
 100         def refine(self):
 101                 pass
 102
 103         ##
 104         # : get a dictionary of author
 105         def getAuthorInfo(self, author):
 106                 return None
 107
 108         def getActivityByDayOfWeek(self):
 109                 return {}
 110
 111         def getActivityByHourOfDay(self):
 112                 return {}
 113
 114         # : get a dictionary of domains
 115         def getDomainInfo(self, domain):
 116                 return None
 117
 118         ##
 119         # Get a list of authors
 120         def getAuthors(self):
 121                 return []
 122
 123         def getFirstCommitDate(self):
 124                 return datetime.datetime.now()
 125
 126         def getLastCommitDate(self):
 127                 return datetime.datetime.now()
 128
 129         def getStampCreated(self):
 130                 return self.stamp_created
 131
 132         def getTags(self):
 133                 return []
 134
 135         def getTotalAuthors(self):
 136                 return -1
 137
 138         def getTotalCommits(self):
 139                 return -1
 140
 141         def getTotalFiles(self):
 142                 return -1
 143
 144         def getTotalLOC(self):
 145                 return -1
 146
 147         ##
 148         # Save cacheable data
 149         def saveCache(self, cachefile):
 150                 print 'Saving cache...'
 151                 f = open(cachefile, 'w')
 152                 #pickle.dump(self.cache, f)
 153                 data = zlib.compress(pickle.dumps(self.cache))
 154                 f.write(data)
 155                 f.close()
 156
 157 class GitDataCollector(DataCollector):
 158         def collect(self, dir):
 159                 DataCollector.collect(self, dir)
 160
 161                 try:
 162                         self.total_authors = int(getpipeoutput(['git log', 'git shortlog -s', 'wc -l']))
 163                 except:
 164                         self.total_authors = 0
 165                 #self.total_lines = int(getoutput('git-ls-files -z |xargs -0 cat |wc -l'))
 166
 167                 self.activity_by_hour_of_day = {} # hour -> commits
 168                 self.activity_by_day_of_week = {} # day -> commits
 169                 self.activity_by_month_of_year = {} # month [1-12] -> commits
 170                 self.activity_by_hour_of_week = {} # weekday -> hour -> commits
 171                 self.activity_by_hour_of_day_busiest = 0
 172                 self.activity_by_hour_of_week_busiest = 0
 173                 self.activity_by_year_week = {} # yy_wNN -> commits
 174                 self.activity_by_year_week_peak = 0
 175
 176                 self.authors = {} # name -> {commits, first_commit_stamp, last_commit_stamp, last_active_day, active_days, lines_added, lines_removed}
 177
 178                 # domains
 179                 self.domains = {} # domain -> commits
 180
 181                 # author of the month
 182                 self.author_of_month = {} # month -> author -> commits
 183                 self.author_of_year = {} # year -> author -> commits
 184                 self.commits_by_month = {} # month -> commits
 185                 self.commits_by_year = {} # year -> commits
 186                 self.first_commit_stamp = 0
 187                 self.last_commit_stamp = 0
 188                 self.last_active_day = None
 189                 self.active_days = set()
 190
 191                 # lines
 192                 self.total_lines = 0
 193                 self.total_lines_added = 0
 194                 self.total_lines_removed = 0
 195
 196                 # timezone
 197                 self.commits_by_timezone = {} # timezone -> commits
 198
 199                 # tags
 200                 self.tags = {}
 201                 lines = getpipeoutput(['git show-ref --tags']).split('\n')
 202                 for line in lines:
 203                         if len(line) == 0:
 204                                 continue
 205                         (hash, tag) = line.split(' ')
 206
 207                         tag = tag.replace('refs/tags/', '')
 208                         output = getpipeoutput(['git log "%s" --pretty=format:"%%at %%an" -n 1' % hash])
 209                         if len(output) > 0:
 210                                 parts = output.split(' ')
 211                                 stamp = 0
 212                                 try:
 213                                         stamp = int(parts[0])
 214                                 except ValueError:
 215                                         stamp = 0
 216                                 self.tags[tag] = { 'stamp': stamp, 'hash' : hash, 'date' : datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), 'commits': 0, 'authors': {} }
 217
 218                 # collect info on tags, starting from latest
 219                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), self.tags.items()))))
 220                 prev = None
 221                 for tag in reversed(tags_sorted_by_date_desc):
 222                         cmd = 'git shortlog -s "%s"' % tag
 223                         if prev != None:
 224                                 cmd += ' "^%s"' % prev
 225                         output = getpipeoutput([cmd])
 226                         if len(output) == 0:
 227                                 continue
 228                         prev = tag
 229                         for line in output.split('\n'):
 230                                 parts = re.split('\s+', line, 2)
 231                                 commits = int(parts[1])
 232                                 author = parts[2]
 233                                 self.tags[tag]['commits'] += commits
 234                                 self.tags[tag]['authors'][author] = commits
 235
 236                 # Collect revision statistics
 237                 # Outputs "<stamp> <date> <time> <timezone> <author> '<' <mail> '>'"
 238                 lines = getpipeoutput(['git rev-list --pretty=format:"%at %ai %an <%aE>" HEAD', 'grep -v ^commit']).split('\n')
 239                 for line in lines:
 240                         parts = line.split(' ', 4)
 241                         author = ''
 242                         try:
 243                                 stamp = int(parts[0])
 244                         except ValueError:
 245                                 stamp = 0
 246                         timezone = parts[3]
 247                         author, mail = parts[4].split('<', 1)
 248                         author = author.rstrip()
 249                         mail = mail.rstrip('>')
 250                         domain = '?'
 251                         if mail.find('@') != -1:
 252                                 domain = mail.rsplit('@', 1)[1]
 253                         date = datetime.datetime.fromtimestamp(float(stamp))
 254
 255                         # First and last commit stamp
 256                         if self.last_commit_stamp == 0:
 257                                 self.last_commit_stamp = stamp
 258                         self.first_commit_stamp = stamp
 259
 260                         # activity
 261                         # hour
 262                         hour = date.hour
 263                         self.activity_by_hour_of_day[hour] = self.activity_by_hour_of_day.get(hour, 0) + 1
 264                         # most active hour?
 265                         if self.activity_by_hour_of_day[hour] > self.activity_by_hour_of_day_busiest:
 266                                 self.activity_by_hour_of_day_busiest = self.activity_by_hour_of_day[hour]
 267
 268                         # day of week
 269                         day = date.weekday()
 270                         self.activity_by_day_of_week[day] = self.activity_by_day_of_week.get(day, 0) + 1
 271
 272                         # domain stats
 273                         if domain not in self.domains:
 274                                 self.domains[domain] = {}
 275                         # commits
 276                         self.domains[domain]['commits'] = self.domains[domain].get('commits', 0) + 1
 277
 278                         # hour of week
 279                         if day not in self.activity_by_hour_of_week:
 280                                 self.activity_by_hour_of_week[day] = {}
 281                         self.activity_by_hour_of_week[day][hour] = self.activity_by_hour_of_week[day].get(hour, 0) + 1
 282                         # most active hour?
 283                         if self.activity_by_hour_of_week[day][hour] > self.activity_by_hour_of_week_busiest:
 284                                 self.activity_by_hour_of_week_busiest = self.activity_by_hour_of_week[day][hour]
 285
 286                         # month of year
 287                         month = date.month
 288                         self.activity_by_month_of_year[month] = self.activity_by_month_of_year.get(month, 0) + 1
 289
 290                         # yearly/weekly activity
 291                         yyw = date.strftime('%Y-%W')
 292                         self.activity_by_year_week[yyw] = self.activity_by_year_week.get(yyw, 0) + 1
 293                         if self.activity_by_year_week_peak < self.activity_by_year_week[yyw]:
 294                                 self.activity_by_year_week_peak = self.activity_by_year_week[yyw]
 295
 296                         # author stats
 297                         if author not in self.authors:
 298                                 self.authors[author] = {}
 299                         # commits
 300                         if 'last_commit_stamp' not in self.authors[author]:
 301                                 self.authors[author]['last_commit_stamp'] = stamp
 302                         self.authors[author]['first_commit_stamp'] = stamp
 303                         self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 304
 305                         # author of the month/year
 306                         yymm = date.strftime('%Y-%m')
 307                         if yymm in self.author_of_month:
 308                                 self.author_of_month[yymm][author] = self.author_of_month[yymm].get(author, 0) + 1
 309                         else:
 310                                 self.author_of_month[yymm] = {}
 311                                 self.author_of_month[yymm][author] = 1
 312                         self.commits_by_month[yymm] = self.commits_by_month.get(yymm, 0) + 1
 313
 314                         yy = date.year
 315                         if yy in self.author_of_year:
 316                                 self.author_of_year[yy][author] = self.author_of_year[yy].get(author, 0) + 1
 317                         else:
 318                                 self.author_of_year[yy] = {}
 319                                 self.author_of_year[yy][author] = 1
 320                         self.commits_by_year[yy] = self.commits_by_year.get(yy, 0) + 1
 321
 322                         # authors: active days
 323                         yymmdd = date.strftime('%Y-%m-%d')
 324                         if 'last_active_day' not in self.authors[author]:
 325                                 self.authors[author]['last_active_day'] = yymmdd
 326                                 self.authors[author]['active_days'] = 1
 327                         elif yymmdd != self.authors[author]['last_active_day']:
 328                                 self.authors[author]['last_active_day'] = yymmdd
 329                                 self.authors[author]['active_days'] += 1
 330
 331                         # project: active days
 332                         if yymmdd != self.last_active_day:
 333                                 self.last_active_day = yymmdd
 334                                 self.active_days.add(yymmdd)
 335
 336                         # timezone
 337                         self.commits_by_timezone[timezone] = self.commits_by_timezone.get(timezone, 0) + 1
 338
 339                 # TODO Optimize this, it's the worst bottleneck
 340                 # outputs "<stamp> <files>" for each revision
 341                 self.files_by_stamp = {} # stamp -> files
 342                 revlines = getpipeoutput(['git rev-list --pretty=format:"%at %T" HEAD', 'grep -v ^commit']).strip().split('\n')
 343                 lines = []
 344                 for revline in revlines:
 345                         time, rev = revline.split(' ')
 346                         linecount = self.getFilesInCommit(rev)
 347                         lines.append('%d %d' % (int(time), linecount))
 348
 349                 self.total_commits = len(lines)
 350                 for line in lines:
 351                         parts = line.split(' ')
 352                         if len(parts) != 2:
 353                                 continue
 354                         (stamp, files) = parts[0:2]
 355                         try:
 356                                 self.files_by_stamp[int(stamp)] = int(files)
 357                         except ValueError:
 358                                 print 'Warning: failed to parse line "%s"' % line
 359
 360                 # extensions
 361                 self.extensions = {} # extension -> files, lines
 362                 lines = getpipeoutput(['git ls-tree -r -z HEAD']).split('\000')
 363                 self.total_files = len(lines)
 364                 for line in lines:
 365                         if len(line) == 0:
 366                                 continue
 367                         parts = re.split('\s+', line, 4)
 368                         sha1 = parts[2]
 369                         filename = parts[3]
 370
 371                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 372                                 ext = ''
 373                         else:
 374                                 ext = filename[(filename.rfind('.') + 1):]
 375                         if len(ext) > conf['max_ext_length']:
 376                                 ext = ''
 377
 378                         if ext not in self.extensions:
 379                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 380
 381                         self.extensions[ext]['files'] += 1
 382                         try:
 383                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 384                         except:
 385                                 print 'Warning: Could not count lines for file "%s"' % line
 386
 387                 # line statistics
 388                 # outputs:
 389                 #  N files changed, N insertions (+), N deletions(-)
 390                 # <stamp> <author>
 391                 self.changes_by_date = {} # stamp -> { files, ins, del }
 392                 lines = getpipeoutput(['git log --shortstat --pretty=format:"%at %an"']).split('\n')
 393                 lines.reverse()
 394                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 395                 author = None
 396                 for line in lines:
 397                         if len(line) == 0:
 398                                 continue
 399
 400                         # <stamp> <author>
 401                         if line.find('files changed,') == -1:
 402                                 pos = line.find(' ')
 403                                 if pos != -1:
 404                                         try:
 405                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 406                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 407                                                 if author not in self.authors:
 408                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0 }
 409                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 410                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 411                                         except ValueError:
 412                                                 print 'Warning: unexpected line "%s"' % line
 413                                 else:
 414                                         print 'Warning: unexpected line "%s"' % line
 415                         else:
 416                                 numbers = re.findall('\d+', line)
 417                                 if len(numbers) == 3:
 418                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 419                                         total_lines += inserted
 420                                         total_lines -= deleted
 421                                         self.total_lines_added += inserted
 422                                         self.total_lines_removed += deleted
 423                                 else:
 424                                         print 'Warning: failed to handle line "%s"' % line
 425                                         (files, inserted, deleted) = (0, 0, 0)
 426                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 427                 self.total_lines = total_lines
 428
 429         def refine(self):
 430                 # authors
 431                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 432                 authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 433                 authors_by_commits.reverse() # most first
 434                 for i, name in enumerate(authors_by_commits):
 435                         self.authors[name]['place_by_commits'] = i + 1
 436
 437                 for name in self.authors.keys():
 438                         a = self.authors[name]
 439                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 440                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 441                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 442                         delta = date_last - date_first
 443                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 444                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 445                         a['timedelta'] = delta
 446
 447         def getActiveDays(self):
 448                 return self.active_days
 449
 450         def getActivityByDayOfWeek(self):
 451                 return self.activity_by_day_of_week
 452
 453         def getActivityByHourOfDay(self):
 454                 return self.activity_by_hour_of_day
 455
 456         def getAuthorInfo(self, author):
 457                 return self.authors[author]
 458
 459         def getAuthors(self, limit = None):
 460                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 461                 res.reverse()
 462                 return res[:limit]
 463
 464         def getCommitDeltaDays(self):
 465                 return (self.last_commit_stamp - self.first_commit_stamp) / 86400 + 1
 466
 467         def getDomainInfo(self, domain):
 468                 return self.domains[domain]
 469
 470         def getDomains(self):
 471                 return self.domains.keys()
 472
 473         def getFilesInCommit(self, rev):
 474                 try:
 475                         res = self.cache['files_in_tree'][rev]
 476                 except:
 477                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 478                         if 'files_in_tree' not in self.cache:
 479                                 self.cache['files_in_tree'] = {}
 480                         self.cache['files_in_tree'][rev] = res
 481
 482                 return res
 483
 484         def getFirstCommitDate(self):
 485                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 486
 487         def getLastCommitDate(self):
 488                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 489
 490         def getLinesInBlob(self, sha1):
 491                 try:
 492                         res = self.cache['lines_in_blob'][sha1]
 493                 except:
 494                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 495                         if 'lines_in_blob' not in self.cache:
 496                                 self.cache['lines_in_blob'] = {}
 497                         self.cache['lines_in_blob'][sha1] = res
 498                 return res
 499
 500         def getTags(self):
 501                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 502                 return lines.split('\n')
 503
 504         def getTagDate(self, tag):
 505                 return self.revToDate('tags/' + tag)
 506
 507         def getTotalAuthors(self):
 508                 return self.total_authors
 509
 510         def getTotalCommits(self):
 511                 return self.total_commits
 512
 513         def getTotalFiles(self):
 514                 return self.total_files
 515
 516         def getTotalLOC(self):
 517                 return self.total_lines
 518
 519         def revToDate(self, rev):
 520                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 521                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 522
 523 class ReportCreator:
 524         """Creates the actual report based on given data."""
 525         def __init__(self):
 526                 pass
 527
 528         def create(self, data, path):
 529                 self.data = data
 530                 self.path = path
 531
 532 def html_linkify(text):
 533         return text.lower().replace(' ', '_')
 534
 535 def html_header(level, text):
 536         name = html_linkify(text)
 537         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 538
 539 class HTMLReportCreator(ReportCreator):
 540         def create(self, data, path):
 541                 ReportCreator.create(self, data, path)
 542                 self.title = data.projectname
 543
 544                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 545                 binarypath = os.path.dirname(os.path.abspath(__file__))
 546                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 547                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 548                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 549                         for base in basedirs:
 550                                 src = base + '/' + file
 551                                 if os.path.exists(src):
 552                                         shutil.copyfile(src, path + '/' + file)
 553                                         break
 554                         else:
 555                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 556
 557                 f = open(path + "/index.html", 'w')
 558                 format = '%Y-%m-%d %H:%M:%S'
 559                 self.printHeader(f)
 560
 561                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 562
 563                 self.printNav(f)
 564
 565                 f.write('<dl>')
 566                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 567                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 568                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s)</dd>' % getversion())
 569                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 570                 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())))
 571                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 572                 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))
 573                 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()))
 574                 f.write('<dt>Authors</dt><dd>%s</dd>' % data.getTotalAuthors())
 575                 f.write('</dl>')
 576
 577                 f.write('</body>\n</html>')
 578                 f.close()
 579
 580                 ###
 581                 # Activity
 582                 f = open(path + '/activity.html', 'w')
 583                 self.printHeader(f)
 584                 f.write('<h1>Activity</h1>')
 585                 self.printNav(f)
 586
 587                 #f.write('<h2>Last 30 days</h2>')
 588
 589                 #f.write('<h2>Last 12 months</h2>')
 590
 591                 # Weekly activity
 592                 WEEKS = 32
 593                 f.write(html_header(2, 'Weekly activity'))
 594                 f.write('<p>Last %d weeks</p>' % WEEKS)
 595
 596                 # generate weeks to show (previous N weeks from now)
 597                 now = datetime.datetime.now()
 598                 deltaweek = datetime.timedelta(7)
 599                 weeks = []
 600                 stampcur = now
 601                 for i in range(0, WEEKS):
 602                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 603                         stampcur -= deltaweek
 604
 605                 # top row: commits & bar
 606                 f.write('<table class="noborders"><tr>')
 607                 for i in range(0, WEEKS):
 608                         commits = 0
 609                         if weeks[i] in data.activity_by_year_week:
 610                                 commits = data.activity_by_year_week[weeks[i]]
 611
 612                         percentage = 0
 613                         if weeks[i] in data.activity_by_year_week:
 614                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 615                         height = max(1, int(200 * percentage))
 616                         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))
 617
 618                 # bottom row: year/week
 619                 f.write('</tr><tr>')
 620                 for i in range(0, WEEKS):
 621                         f.write('<td>%s</td>' % (WEEKS - i))
 622                 f.write('</tr></table>')
 623
 624                 # Hour of Day
 625                 f.write(html_header(2, 'Hour of Day'))
 626                 hour_of_day = data.getActivityByHourOfDay()
 627                 f.write('<table><tr><th>Hour</th>')
 628                 for i in range(0, 24):
 629                         f.write('<th>%d</th>' % i)
 630                 f.write('</tr>\n<tr><th>Commits</th>')
 631                 fp = open(path + '/hour_of_day.dat', 'w')
 632                 for i in range(0, 24):
 633                         if i in hour_of_day:
 634                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 635                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 636                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 637                         else:
 638                                 f.write('<td>0</td>')
 639                                 fp.write('%d 0\n' % i)
 640                 fp.close()
 641                 f.write('</tr>\n<tr><th>%</th>')
 642                 totalcommits = data.getTotalCommits()
 643                 for i in range(0, 24):
 644                         if i in hour_of_day:
 645                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 646                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 647                         else:
 648                                 f.write('<td>0.00</td>')
 649                 f.write('</tr></table>')
 650                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 651                 fg = open(path + '/hour_of_day.dat', 'w')
 652                 for i in range(0, 24):
 653                         if i in hour_of_day:
 654                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 655                         else:
 656                                 fg.write('%d 0\n' % (i + 1))
 657                 fg.close()
 658
 659                 # Day of Week
 660                 f.write(html_header(2, 'Day of Week'))
 661                 day_of_week = data.getActivityByDayOfWeek()
 662                 f.write('<div class="vtable"><table>')
 663                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 664                 fp = open(path + '/day_of_week.dat', 'w')
 665                 for d in range(0, 7):
 666                         commits = 0
 667                         if d in day_of_week:
 668                                 commits = day_of_week[d]
 669                         fp.write('%d %d\n' % (d + 1, commits))
 670                         f.write('<tr>')
 671                         f.write('<th>%d</th>' % (d + 1))
 672                         if d in day_of_week:
 673                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 674                         else:
 675                                 f.write('<td>0</td>')
 676                         f.write('</tr>')
 677                 f.write('</table></div>')
 678                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 679                 fp.close()
 680
 681                 # Hour of Week
 682                 f.write(html_header(2, 'Hour of Week'))
 683                 f.write('<table>')
 684
 685                 f.write('<tr><th>Weekday</th>')
 686                 for hour in range(0, 24):
 687                         f.write('<th>%d</th>' % (hour))
 688                 f.write('</tr>')
 689
 690                 for weekday in range(0, 7):
 691                         f.write('<tr><th>%d</th>' % (weekday + 1))
 692                         for hour in range(0, 24):
 693                                 try:
 694                                         commits = data.activity_by_hour_of_week[weekday][hour]
 695                                 except KeyError:
 696                                         commits = 0
 697                                 if commits != 0:
 698                                         f.write('<td')
 699                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 700                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 701                                         f.write('>%d</td>' % commits)
 702                                 else:
 703                                         f.write('<td></td>')
 704                         f.write('</tr>')
 705
 706                 f.write('</table>')
 707
 708                 # Month of Year
 709                 f.write(html_header(2, 'Month of Year'))
 710                 f.write('<div class="vtable"><table>')
 711                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 712                 fp = open (path + '/month_of_year.dat', 'w')
 713                 for mm in range(1, 13):
 714                         commits = 0
 715                         if mm in data.activity_by_month_of_year:
 716                                 commits = data.activity_by_month_of_year[mm]
 717                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 718                         fp.write('%d %d\n' % (mm, commits))
 719                 fp.close()
 720                 f.write('</table></div>')
 721                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 722
 723                 # Commits by year/month
 724                 f.write(html_header(2, 'Commits by year/month'))
 725                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th></tr>')
 726                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 727                         f.write('<tr><td>%s</td><td>%d</td></tr>' % (yymm, data.commits_by_month[yymm]))
 728                 f.write('</table></div>')
 729                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 730                 fg = open(path + '/commits_by_year_month.dat', 'w')
 731                 for yymm in sorted(data.commits_by_month.keys()):
 732                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 733                 fg.close()
 734
 735                 # Commits by year
 736                 f.write(html_header(2, 'Commits by Year'))
 737                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th></tr>')
 738                 for yy in reversed(sorted(data.commits_by_year.keys())):
 739                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td></tr>' % (yy, data.commits_by_year[yy], (100.0 * data.commits_by_year[yy]) / data.getTotalCommits()))
 740                 f.write('</table></div>')
 741                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 742                 fg = open(path + '/commits_by_year.dat', 'w')
 743                 for yy in sorted(data.commits_by_year.keys()):
 744                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 745                 fg.close()
 746
 747                 # Commits by timezone
 748                 f.write(html_header(2, 'Commits by Timezone'))
 749                 f.write('<table><tr>')
 750                 f.write('<th>Timezone</th><th>Commits</th>')
 751                 max_commits_on_tz = max(data.commits_by_timezone.values())
 752                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 753                         commits = data.commits_by_timezone[i]
 754                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 755                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 756                 f.write('</tr></table>')
 757
 758                 f.write('</body></html>')
 759                 f.close()
 760
 761                 ###
 762                 # Authors
 763                 f = open(path + '/authors.html', 'w')
 764                 self.printHeader(f)
 765
 766                 f.write('<h1>Authors</h1>')
 767                 self.printNav(f)
 768
 769                 # Authors :: List of authors
 770                 f.write(html_header(2, 'List of Authors'))
 771
 772                 f.write('<table class="authors sortable" id="authors">')
 773                 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>')
 774                 for author in sorted(data.getAuthors(conf['max_authors'])):
 775                         info = data.getAuthorInfo(author)
 776                         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'], info['active_days'], info['place_by_commits']))
 777                 f.write('</table>')
 778
 779                 allauthors = data.getAuthors()
 780                 if len(allauthors) > conf['max_authors']:
 781                         rest = allauthors[conf['max_authors']:]
 782                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 783
 784                 # Authors :: Author of Month
 785                 f.write(html_header(2, 'Author of Month'))
 786                 f.write('<table class="sortable" id="aom">')
 787                 f.write('<tr><th>Month</th><th>Author</th><th>Commits (%)</th><th class="unsortable">Next top 5</th></tr>')
 788                 for yymm in reversed(sorted(data.author_of_month.keys())):
 789                         authordict = data.author_of_month[yymm]
 790                         authors = getkeyssortedbyvalues(authordict)
 791                         authors.reverse()
 792                         commits = data.author_of_month[yymm][authors[0]]
 793                         next = ', '.join(authors[1:5])
 794                         f.write('<tr><td>%s</td><td>%s</td><td>%d (%.2f%% of %d)</td><td>%s</td></tr>' % (yymm, authors[0], commits, (100.0 * commits) / data.commits_by_month[yymm], data.commits_by_month[yymm], next))
 795
 796                 f.write('</table>')
 797
 798                 f.write(html_header(2, 'Author of Year'))
 799                 f.write('<table class="sortable" id="aoy"><tr><th>Year</th><th>Author</th><th>Commits (%)</th><th class="unsortable">Next top 5</th></tr>')
 800                 for yy in reversed(sorted(data.author_of_year.keys())):
 801                         authordict = data.author_of_year[yy]
 802                         authors = getkeyssortedbyvalues(authordict)
 803                         authors.reverse()
 804                         commits = data.author_of_year[yy][authors[0]]
 805                         next = ', '.join(authors[1:5])
 806                         f.write('<tr><td>%s</td><td>%s</td><td>%d (%.2f%% of %d)</td><td>%s</td></tr>' % (yy, authors[0], commits, (100.0 * commits) / data.commits_by_year[yy], data.commits_by_year[yy], next))
 807                 f.write('</table>')
 808
 809                 # Domains
 810                 f.write(html_header(2, 'Commits by Domains'))
 811                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
 812                 domains_by_commits.reverse() # most first
 813                 f.write('<div class="vtable"><table>')
 814                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
 815                 fp = open(path + '/domains.dat', 'w')
 816                 n = 0
 817                 for domain in domains_by_commits:
 818                         if n == conf['max_domains']:
 819                                 break
 820                         commits = 0
 821                         n += 1
 822                         info = data.getDomainInfo(domain)
 823                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
 824                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
 825                 f.write('</table></div>')
 826                 f.write('<img src="domains.png" alt="Commits by Domains" />')
 827                 fp.close()
 828
 829                 f.write('</body></html>')
 830                 f.close()
 831
 832                 ###
 833                 # Files
 834                 f = open(path + '/files.html', 'w')
 835                 self.printHeader(f)
 836                 f.write('<h1>Files</h1>')
 837                 self.printNav(f)
 838
 839                 f.write('<dl>\n')
 840                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
 841                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
 842                 f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % ((100.0 * data.getTotalLOC()) / data.getTotalFiles()))
 843                 f.write('</dl>\n')
 844
 845                 # Files :: File count by date
 846                 f.write(html_header(2, 'File count by date'))
 847
 848                 # use set to get rid of duplicate/unnecessary entries
 849                 files_by_date = set()
 850                 for stamp in sorted(data.files_by_stamp.keys()):
 851                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
 852
 853                 fg = open(path + '/files_by_date.dat', 'w')
 854                 for line in sorted(list(files_by_date)):
 855                         fg.write('%s\n' % line)
 856                 #for stamp in sorted(data.files_by_stamp.keys()):
 857                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
 858                 fg.close()
 859
 860                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
 861
 862                 #f.write('<h2>Average file size by date</h2>')
 863
 864                 # Files :: Extensions
 865                 f.write(html_header(2, 'Extensions'))
 866                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
 867                 for ext in sorted(data.extensions.keys()):
 868                         files = data.extensions[ext]['files']
 869                         lines = data.extensions[ext]['lines']
 870                         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, (100.0 * lines) / data.getTotalLOC(), lines / files))
 871                 f.write('</table>')
 872
 873                 f.write('</body></html>')
 874                 f.close()
 875
 876                 ###
 877                 # Lines
 878                 f = open(path + '/lines.html', 'w')
 879                 self.printHeader(f)
 880                 f.write('<h1>Lines</h1>')
 881                 self.printNav(f)
 882
 883                 f.write('<dl>\n')
 884                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
 885                 f.write('</dl>\n')
 886
 887                 f.write(html_header(2, 'Lines of Code'))
 888                 f.write('<img src="lines_of_code.png" />')
 889
 890                 fg = open(path + '/lines_of_code.dat', 'w')
 891                 for stamp in sorted(data.changes_by_date.keys()):
 892                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
 893                 fg.close()
 894
 895                 f.write('</body></html>')
 896                 f.close()
 897
 898                 ###
 899                 # tags.html
 900                 f = open(path + '/tags.html', 'w')
 901                 self.printHeader(f)
 902                 f.write('<h1>Tags</h1>')
 903                 self.printNav(f)
 904
 905                 f.write('<dl>')
 906                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
 907                 if len(data.tags) > 0:
 908                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
 909                 f.write('</dl>')
 910
 911                 f.write('<table class="tags">')
 912                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
 913                 # sort the tags by date desc
 914                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
 915                 for tag in tags_sorted_by_date_desc:
 916                         authorinfo = []
 917                         authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
 918                         for i in reversed(authors_by_commits):
 919                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
 920                         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)))
 921                 f.write('</table>')
 922
 923                 f.write('</body></html>')
 924                 f.close()
 925
 926                 self.createGraphs(path)
 927
 928         def createGraphs(self, path):
 929                 print 'Generating graphs...'
 930
 931                 # hour of day
 932                 f = open(path + '/hour_of_day.plot', 'w')
 933                 f.write(GNUPLOT_COMMON)
 934                 f.write(
 935 """
 936 set output 'hour_of_day.png'
 937 unset key
 938 set xrange [0.5:24.5]
 939 set xtics 4
 940 set grid y
 941 set ylabel "Commits"
 942 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
 943 """)
 944                 f.close()
 945
 946                 # day of week
 947                 f = open(path + '/day_of_week.plot', 'w')
 948                 f.write(GNUPLOT_COMMON)
 949                 f.write(
 950 """
 951 set output 'day_of_week.png'
 952 unset key
 953 set xrange [0.5:7.5]
 954 set xtics 1
 955 set grid y
 956 set ylabel "Commits"
 957 plot 'day_of_week.dat' using 1:2:(0.5) w boxes fs solid
 958 """)
 959                 f.close()
 960
 961                 # Domains
 962                 f = open(path + '/domains.plot', 'w')
 963                 f.write(GNUPLOT_COMMON)
 964                 f.write(
 965 """
 966 set output 'domains.png'
 967 unset key
 968 unset xtics
 969 set grid y
 970 set ylabel "Commits"
 971 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
 972 """)
 973                 f.close()
 974
 975                 # Month of Year
 976                 f = open(path + '/month_of_year.plot', 'w')
 977                 f.write(GNUPLOT_COMMON)
 978                 f.write(
 979 """
 980 set output 'month_of_year.png'
 981 unset key
 982 set xrange [0.5:12.5]
 983 set xtics 1
 984 set grid y
 985 set ylabel "Commits"
 986 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
 987 """)
 988                 f.close()
 989
 990                 # commits_by_year_month
 991                 f = open(path + '/commits_by_year_month.plot', 'w')
 992                 f.write(GNUPLOT_COMMON)
 993                 f.write(
 994 """
 995 set output 'commits_by_year_month.png'
 996 unset key
 997 set xdata time
 998 set timefmt "%Y-%m"
 999 set format x "%Y-%m"
1000 set xtics rotate by 90 15768000
1001 set bmargin 5
1002 set grid y
1003 set ylabel "Commits"
1004 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1005 """)
1006                 f.close()
1007
1008                 # commits_by_year
1009                 f = open(path + '/commits_by_year.plot', 'w')
1010                 f.write(GNUPLOT_COMMON)
1011                 f.write(
1012 """
1013 set output 'commits_by_year.png'
1014 unset key
1015 set xtics 1 rotate by 90
1016 set grid y
1017 set ylabel "Commits"
1018 set yrange [0:]
1019 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1020 """)
1021                 f.close()
1022
1023                 # Files by date
1024                 f = open(path + '/files_by_date.plot', 'w')
1025                 f.write(GNUPLOT_COMMON)
1026                 f.write(
1027 """
1028 set output 'files_by_date.png'
1029 unset key
1030 set xdata time
1031 set timefmt "%Y-%m-%d"
1032 set format x "%Y-%m-%d"
1033 set grid y
1034 set ylabel "Files"
1035 set xtics rotate by 90
1036 set ytics autofreq
1037 set bmargin 6
1038 plot 'files_by_date.dat' using 1:2 w steps
1039 """)
1040                 f.close()
1041
1042                 # Lines of Code
1043                 f = open(path + '/lines_of_code.plot', 'w')
1044                 f.write(GNUPLOT_COMMON)
1045                 f.write(
1046 """
1047 set output 'lines_of_code.png'
1048 unset key
1049 set xdata time
1050 set timefmt "%s"
1051 set format x "%Y-%m-%d"
1052 set grid y
1053 set ylabel "Lines"
1054 set xtics rotate by 90
1055 set bmargin 6
1056 plot 'lines_of_code.dat' using 1:2 w lines
1057 """)
1058                 f.close()
1059
1060                 os.chdir(path)
1061                 files = glob.glob(path + '/*.plot')
1062                 for f in files:
1063                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1064                         if len(out) > 0:
1065                                 print out
1066
1067         def printHeader(self, f, title = ''):
1068                 f.write(
1069 """<?xml version="1.0" encoding="UTF-8"?>
1070 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1071 <html xmlns="http://www.w3.org/1999/xhtml">
1072 <head>
1073         <title>GitStats - %s</title>
1074         <link rel="stylesheet" href="%s" type="text/css" />
1075         <meta name="generator" content="GitStats %s" />
1076         <script type="text/javascript" src="sortable.js"></script>
1077 </head>
1078 <body>
1079 """ % (self.title, conf['style'], getversion()))
1080
1081         def printNav(self, f):
1082                 f.write("""
1083 <div class="nav">
1084 <ul>
1085 <li><a href="index.html">General</a></li>
1086 <li><a href="activity.html">Activity</a></li>
1087 <li><a href="authors.html">Authors</a></li>
1088 <li><a href="files.html">Files</a></li>
1089 <li><a href="lines.html">Lines</a></li>
1090 <li><a href="tags.html">Tags</a></li>
1091 </ul>
1092 </div>
1093 """)
1094
1095
1096 class GitStats:
1097         def run(self, args_orig):
1098                 optlist, args = getopt.getopt(args_orig, 'c:')
1099                 for o,v in optlist:
1100                         if o == '-c':
1101                                 key, value = v.split('=', 1)
1102                                 if key not in conf:
1103                                         raise 'Error: no such key "%s" in config' % key
1104                                 if isinstance(conf[key], int):
1105                                         conf[key] = int(value)
1106                                 else:
1107                                         conf[key] = value
1108
1109                 if len(args) < 2:
1110                         print """
1111 Usage: gitstats [options] <gitpath> <outputpath>
1112
1113 Options:
1114 -c key=value     Override configuration value
1115
1116 Default config values:
1117 %s
1118 """ % conf
1119                         sys.exit(0)
1120
1121                 gitpath = args[0]
1122                 outputpath = os.path.abspath(args[1])
1123                 rundir = os.getcwd()
1124
1125                 try:
1126                         os.makedirs(outputpath)
1127                 except OSError:
1128                         pass
1129                 if not os.path.isdir(outputpath):
1130                         print 'FATAL: Output path is not a directory or does not exist'
1131                         sys.exit(1)
1132
1133                 print 'Git path: %s' % gitpath
1134                 print 'Output path: %s' % outputpath
1135
1136                 os.chdir(gitpath)
1137
1138                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1139
1140                 print 'Collecting data...'
1141                 data = GitDataCollector()
1142                 data.loadCache(cachefile)
1143                 data.collect(gitpath)
1144                 print 'Refining data...'
1145                 data.saveCache(cachefile)
1146                 data.refine()
1147
1148                 os.chdir(rundir)
1149
1150                 print 'Generating report...'
1151                 report = HTMLReportCreator()
1152                 report.create(data, outputpath)
1153
1154                 time_end = time.time()
1155                 exectime_internal = time_end - time_start
1156                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1157
1158 g = GitStats()
1159 g.run(sys.argv[1:])
1160