gitstats

   1 #!/usr/bin/env python
   2 # Copyright (c) 2007-2011 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 size 640,240\nset size 1.0,1.0\n'
  18 ON_LINUX = (platform.system() == 'Linux')
  19 WEEKDAYS = ('Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun')
  20
  21 exectime_internal = 0.0
  22 exectime_external = 0.0
  23 time_start = time.time()
  24
  25 # By default, gnuplot is searched from path, but can be overridden with the
  26 # environment variable "GNUPLOT"
  27 gnuplot_cmd = 'gnuplot'
  28 if 'GNUPLOT' in os.environ:
  29         gnuplot_cmd = os.environ['GNUPLOT']
  30
  31 conf = {
  32         'max_domains': 10,
  33         'max_ext_length': 10,
  34         'style': 'gitstats.css',
  35         'max_authors': 20,
  36         'authors_top': 5,
  37         'commit_begin': '',
  38         'commit_end': 'HEAD',
  39         'linear_linestats': 1,
  40         'project_name': '',
  41 }
  42
  43 def getpipeoutput(cmds, quiet = False):
  44         global exectime_external
  45         start = time.time()
  46         if not quiet and ON_LINUX and os.isatty(1):
  47                 print '>> ' + ' | '.join(cmds),
  48                 sys.stdout.flush()
  49         p0 = subprocess.Popen(cmds[0], stdout = subprocess.PIPE, shell = True)
  50         p = p0
  51         for x in cmds[1:]:
  52                 p = subprocess.Popen(x, stdin = p0.stdout, stdout = subprocess.PIPE, shell = True)
  53                 p0 = p
  54         output = p.communicate()[0]
  55         end = time.time()
  56         if not quiet:
  57                 if ON_LINUX and os.isatty(1):
  58                         print '\r',
  59                 print '[%.5f] >> %s' % (end - start, ' | '.join(cmds))
  60         exectime_external += (end - start)
  61         return output.rstrip('\n')
  62
  63 def getcommitrange(defaultrange = 'HEAD', end_only = False):
  64         if len(conf['commit_end']) > 0:
  65                 if end_only or len(conf['commit_begin']) == 0:
  66                         return conf['commit_end']
  67                 return '%s..%s' % (conf['commit_begin'], conf['commit_end'])
  68         return defaultrange
  69
  70 def getkeyssortedbyvalues(dict):
  71         return map(lambda el : el[1], sorted(map(lambda el : (el[1], el[0]), dict.items())))
  72
  73 # dict['author'] = { 'commits': 512 } - ...key(dict, 'commits')
  74 def getkeyssortedbyvaluekey(d, key):
  75         return map(lambda el : el[1], sorted(map(lambda el : (d[el][key], el), d.keys())))
  76
  77 VERSION = 0
  78 def getversion():
  79         global VERSION
  80         if VERSION == 0:
  81                 VERSION = getpipeoutput(["git rev-parse --short %s" % getcommitrange('HEAD')]).split('\n')[0]
  82         return VERSION
  83
  84 def getgitversion():
  85         return getpipeoutput(['git --version']).split('\n')[0]
  86
  87 def getgnuplotversion():
  88         return getpipeoutput(['gnuplot --version']).split('\n')[0]
  89
  90 class DataCollector:
  91         """Manages data collection from a revision control repository."""
  92         def __init__(self):
  93                 self.stamp_created = time.time()
  94                 self.cache = {}
  95                 self.total_authors = 0
  96                 self.activity_by_hour_of_day = {} # hour -> commits
  97                 self.activity_by_day_of_week = {} # day -> commits
  98                 self.activity_by_month_of_year = {} # month [1-12] -> commits
  99                 self.activity_by_hour_of_week = {} # weekday -> hour -> commits
 100                 self.activity_by_hour_of_day_busiest = 0
 101                 self.activity_by_hour_of_week_busiest = 0
 102                 self.activity_by_year_week = {} # yy_wNN -> commits
 103                 self.activity_by_year_week_peak = 0
 104
 105                 self.authors = {} # name -> {commits, first_commit_stamp, last_commit_stamp, last_active_day, active_days, lines_added, lines_removed}
 106
 107                 self.total_commits = 0
 108                 self.total_files = 0
 109                 self.authors_by_commits = 0
 110
 111                 # domains
 112                 self.domains = {} # domain -> commits
 113
 114                 # author of the month
 115                 self.author_of_month = {} # month -> author -> commits
 116                 self.author_of_year = {} # year -> author -> commits
 117                 self.commits_by_month = {} # month -> commits
 118                 self.commits_by_year = {} # year -> commits
 119                 self.lines_added_by_month = {} # month -> lines added
 120                 self.lines_added_by_year = {} # year -> lines added
 121                 self.lines_removed_by_month = {} # month -> lines removed
 122                 self.lines_removed_by_year = {} # year -> lines removed
 123                 self.first_commit_stamp = 0
 124                 self.last_commit_stamp = 0
 125                 self.last_active_day = None
 126                 self.active_days = set()
 127
 128                 # lines
 129                 self.total_lines = 0
 130                 self.total_lines_added = 0
 131                 self.total_lines_removed = 0
 132
 133                 # size
 134                 self.total_size = 0
 135
 136                 # timezone
 137                 self.commits_by_timezone = {} # timezone -> commits
 138
 139                 # tags
 140                 self.tags = {}
 141
 142                 self.files_by_stamp = {} # stamp -> files
 143
 144                 # extensions
 145                 self.extensions = {} # extension -> files, lines
 146
 147                 # line statistics
 148                 self.changes_by_date = {} # stamp -> { files, ins, del }
 149
 150         ##
 151         # This should be the main function to extract data from the repository.
 152         def collect(self, dir):
 153                 self.dir = dir
 154                 if len(conf['project_name']) == 0:
 155                         self.projectname = os.path.basename(os.path.abspath(dir))
 156                 else:
 157                         self.projectname = conf['project_name']
 158
 159         ##
 160         # Load cacheable data
 161         def loadCache(self, cachefile):
 162                 if not os.path.exists(cachefile):
 163                         return
 164                 print 'Loading cache...'
 165                 f = open(cachefile, 'rb')
 166                 try:
 167                         self.cache = pickle.loads(zlib.decompress(f.read()))
 168                 except:
 169                         # temporary hack to upgrade non-compressed caches
 170                         f.seek(0)
 171                         self.cache = pickle.load(f)
 172                 f.close()
 173
 174         ##
 175         # Produce any additional statistics from the extracted data.
 176         def refine(self):
 177                 pass
 178
 179         ##
 180         # : get a dictionary of author
 181         def getAuthorInfo(self, author):
 182                 return None
 183
 184         def getActivityByDayOfWeek(self):
 185                 return {}
 186
 187         def getActivityByHourOfDay(self):
 188                 return {}
 189
 190         # : get a dictionary of domains
 191         def getDomainInfo(self, domain):
 192                 return None
 193
 194         ##
 195         # Get a list of authors
 196         def getAuthors(self):
 197                 return []
 198
 199         def getFirstCommitDate(self):
 200                 return datetime.datetime.now()
 201
 202         def getLastCommitDate(self):
 203                 return datetime.datetime.now()
 204
 205         def getStampCreated(self):
 206                 return self.stamp_created
 207
 208         def getTags(self):
 209                 return []
 210
 211         def getTotalAuthors(self):
 212                 return -1
 213
 214         def getTotalCommits(self):
 215                 return -1
 216
 217         def getTotalFiles(self):
 218                 return -1
 219
 220         def getTotalLOC(self):
 221                 return -1
 222
 223         ##
 224         # Save cacheable data
 225         def saveCache(self, cachefile):
 226                 print 'Saving cache...'
 227                 f = open(cachefile, 'wb')
 228                 #pickle.dump(self.cache, f)
 229                 data = zlib.compress(pickle.dumps(self.cache))
 230                 f.write(data)
 231                 f.close()
 232
 233 class GitDataCollector(DataCollector):
 234         def collect(self, dir):
 235                 DataCollector.collect(self, dir)
 236
 237                 try:
 238                         self.total_authors += int(getpipeoutput(['git shortlog -s %s' % getcommitrange(), 'wc -l']))
 239                 except:
 240                         self.total_authors = 0
 241                 #self.total_lines = int(getoutput('git-ls-files -z |xargs -0 cat |wc -l'))
 242
 243                 # tags
 244                 lines = getpipeoutput(['git show-ref --tags']).split('\n')
 245                 for line in lines:
 246                         if len(line) == 0:
 247                                 continue
 248                         (hash, tag) = line.split(' ')
 249
 250                         tag = tag.replace('refs/tags/', '')
 251                         output = getpipeoutput(['git log "%s" --pretty=format:"%%at %%aN" -n 1' % hash])
 252                         if len(output) > 0:
 253                                 parts = output.split(' ')
 254                                 stamp = 0
 255                                 try:
 256                                         stamp = int(parts[0])
 257                                 except ValueError:
 258                                         stamp = 0
 259                                 self.tags[tag] = { 'stamp': stamp, 'hash' : hash, 'date' : datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), 'commits': 0, 'authors': {} }
 260
 261                 # collect info on tags, starting from latest
 262                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), self.tags.items()))))
 263                 prev = None
 264                 for tag in reversed(tags_sorted_by_date_desc):
 265                         cmd = 'git shortlog -s "%s"' % tag
 266                         if prev != None:
 267                                 cmd += ' "^%s"' % prev
 268                         output = getpipeoutput([cmd])
 269                         if len(output) == 0:
 270                                 continue
 271                         prev = tag
 272                         for line in output.split('\n'):
 273                                 parts = re.split('\s+', line, 2)
 274                                 commits = int(parts[1])
 275                                 author = parts[2]
 276                                 self.tags[tag]['commits'] = commits
 277                                 self.tags[tag]['authors'][author] = commits
 278
 279                 # Collect revision statistics
 280                 # Outputs "<stamp> <date> <time> <timezone> <author> '<' <mail> '>'"
 281                 lines = getpipeoutput(['git rev-list --pretty=format:"%%at %%ai %%aN <%%aE>" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).split('\n')
 282                 for line in lines:
 283                         parts = line.split(' ', 4)
 284                         author = ''
 285                         try:
 286                                 stamp = int(parts[0])
 287                         except ValueError:
 288                                 stamp = 0
 289                         timezone = parts[3]
 290                         author, mail = parts[4].split('<', 1)
 291                         author = author.rstrip()
 292                         mail = mail.rstrip('>')
 293                         domain = '?'
 294                         if mail.find('@') != -1:
 295                                 domain = mail.rsplit('@', 1)[1]
 296                         date = datetime.datetime.fromtimestamp(float(stamp))
 297
 298                         # First and last commit stamp (may be in any order because of cherry-picking and patches)
 299                         if stamp > self.last_commit_stamp:
 300                                 self.last_commit_stamp = stamp
 301                         if self.first_commit_stamp == 0 or stamp < self.first_commit_stamp:
 302                                 self.first_commit_stamp = stamp
 303
 304                         # activity
 305                         # hour
 306                         hour = date.hour
 307                         self.activity_by_hour_of_day[hour] = self.activity_by_hour_of_day.get(hour, 0) + 1
 308                         # most active hour?
 309                         if self.activity_by_hour_of_day[hour] > self.activity_by_hour_of_day_busiest:
 310                                 self.activity_by_hour_of_day_busiest = self.activity_by_hour_of_day[hour]
 311
 312                         # day of week
 313                         day = date.weekday()
 314                         self.activity_by_day_of_week[day] = self.activity_by_day_of_week.get(day, 0) + 1
 315
 316                         # domain stats
 317                         if domain not in self.domains:
 318                                 self.domains[domain] = {}
 319                         # commits
 320                         self.domains[domain]['commits'] = self.domains[domain].get('commits', 0) + 1
 321
 322                         # hour of week
 323                         if day not in self.activity_by_hour_of_week:
 324                                 self.activity_by_hour_of_week[day] = {}
 325                         self.activity_by_hour_of_week[day][hour] = self.activity_by_hour_of_week[day].get(hour, 0) + 1
 326                         # most active hour?
 327                         if self.activity_by_hour_of_week[day][hour] > self.activity_by_hour_of_week_busiest:
 328                                 self.activity_by_hour_of_week_busiest = self.activity_by_hour_of_week[day][hour]
 329
 330                         # month of year
 331                         month = date.month
 332                         self.activity_by_month_of_year[month] = self.activity_by_month_of_year.get(month, 0) + 1
 333
 334                         # yearly/weekly activity
 335                         yyw = date.strftime('%Y-%W')
 336                         self.activity_by_year_week[yyw] = self.activity_by_year_week.get(yyw, 0) + 1
 337                         if self.activity_by_year_week_peak < self.activity_by_year_week[yyw]:
 338                                 self.activity_by_year_week_peak = self.activity_by_year_week[yyw]
 339
 340                         # author stats
 341                         if author not in self.authors:
 342                                 self.authors[author] = {}
 343                         # commits, note again that commits may be in any date order because of cherry-picking and patches
 344                         if 'last_commit_stamp' not in self.authors[author]:
 345                                 self.authors[author]['last_commit_stamp'] = stamp
 346                         if stamp > self.authors[author]['last_commit_stamp']:
 347                                 self.authors[author]['last_commit_stamp'] = stamp
 348                         if 'first_commit_stamp' not in self.authors[author]:
 349                                 self.authors[author]['first_commit_stamp'] = stamp
 350                         if stamp < self.authors[author]['first_commit_stamp']:
 351                                 self.authors[author]['first_commit_stamp'] = stamp
 352
 353                         # author of the month/year
 354                         yymm = date.strftime('%Y-%m')
 355                         if yymm in self.author_of_month:
 356                                 self.author_of_month[yymm][author] = self.author_of_month[yymm].get(author, 0) + 1
 357                         else:
 358                                 self.author_of_month[yymm] = {}
 359                                 self.author_of_month[yymm][author] = 1
 360                         self.commits_by_month[yymm] = self.commits_by_month.get(yymm, 0) + 1
 361
 362                         yy = date.year
 363                         if yy in self.author_of_year:
 364                                 self.author_of_year[yy][author] = self.author_of_year[yy].get(author, 0) + 1
 365                         else:
 366                                 self.author_of_year[yy] = {}
 367                                 self.author_of_year[yy][author] = 1
 368                         self.commits_by_year[yy] = self.commits_by_year.get(yy, 0) + 1
 369
 370                         # authors: active days
 371                         yymmdd = date.strftime('%Y-%m-%d')
 372                         if 'last_active_day' not in self.authors[author]:
 373                                 self.authors[author]['last_active_day'] = yymmdd
 374                                 self.authors[author]['active_days'] = set([yymmdd])
 375                         elif yymmdd != self.authors[author]['last_active_day']:
 376                                 self.authors[author]['last_active_day'] = yymmdd
 377                                 self.authors[author]['active_days'].add(yymmdd)
 378
 379                         # project: active days
 380                         if yymmdd != self.last_active_day:
 381                                 self.last_active_day = yymmdd
 382                                 self.active_days.add(yymmdd)
 383
 384                         # timezone
 385                         self.commits_by_timezone[timezone] = self.commits_by_timezone.get(timezone, 0) + 1
 386
 387                 # TODO Optimize this, it's the worst bottleneck
 388                 # outputs "<stamp> <files>" for each revision
 389                 revlines = getpipeoutput(['git rev-list --pretty=format:"%%at %%T" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).strip().split('\n')
 390                 lines = []
 391                 for revline in revlines:
 392                         time, rev = revline.split(' ')
 393                         linecount = self.getFilesInCommit(rev)
 394                         lines.append('%d %d' % (int(time), linecount))
 395
 396                 self.total_commits += len(lines)
 397                 for line in lines:
 398                         parts = line.split(' ')
 399                         if len(parts) != 2:
 400                                 continue
 401                         (stamp, files) = parts[0:2]
 402                         try:
 403                                 self.files_by_stamp[int(stamp)] = int(files)
 404                         except ValueError:
 405                                 print 'Warning: failed to parse line "%s"' % line
 406
 407                 # extensions and size of files
 408                 lines = getpipeoutput(['git ls-tree -r -l -z %s' % getcommitrange('HEAD', end_only = True)]).split('\000')
 409                 for line in lines:
 410                         if len(line) == 0:
 411                                 continue
 412                         parts = re.split('\s+', line, 5)
 413                         sha1 = parts[2]
 414                         size = int(parts[3])
 415                         fullpath = parts[4]
 416
 417                         self.total_size += size
 418                         self.total_files += 1
 419
 420                         filename = fullpath.split('/')[-1] # strip directories
 421                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 422                                 ext = ''
 423                         else:
 424                                 ext = filename[(filename.rfind('.') + 1):]
 425                         if len(ext) > conf['max_ext_length']:
 426                                 ext = ''
 427
 428                         if ext not in self.extensions:
 429                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 430
 431                         self.extensions[ext]['files'] += 1
 432                         try:
 433                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 434                         except:
 435                                 print 'Warning: Could not count lines for file "%s"' % line
 436
 437                 # line statistics
 438                 # outputs:
 439                 #  N files changed, N insertions (+), N deletions(-)
 440                 # <stamp> <author>
 441                 self.changes_by_date = {} # stamp -> { files, ins, del }
 442                 # computation of lines of code by date is better done
 443                 # on a linear history.
 444                 extra = ''
 445                 if conf['linear_linestats']:
 446                         extra = '--first-parent -m'
 447                 lines = getpipeoutput(['git log --shortstat %s --pretty=format:"%%at %%aN" %s' % (extra, getcommitrange('HEAD'))]).split('\n')
 448                 lines.reverse()
 449                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 450                 author = None
 451                 for line in lines:
 452                         if len(line) == 0:
 453                                 continue
 454
 455                         # <stamp> <author>
 456                         if line.find('files changed,') == -1:
 457                                 pos = line.find(' ')
 458                                 if pos != -1:
 459                                         try:
 460                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 461                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 462
 463                                                 date = datetime.datetime.fromtimestamp(stamp)
 464                                                 yymm = date.strftime('%Y-%m')
 465                                                 self.lines_added_by_month[yymm] = self.lines_added_by_month.get(yymm, 0) + inserted
 466                                                 self.lines_removed_by_month[yymm] = self.lines_removed_by_month.get(yymm, 0) + deleted
 467
 468                                                 yy = date.year
 469                                                 self.lines_added_by_year[yy] = self.lines_added_by_year.get(yy,0) + inserted
 470                                                 self.lines_removed_by_year[yy] = self.lines_removed_by_year.get(yy, 0) + deleted
 471
 472                                                 files, inserted, deleted = 0, 0, 0
 473                                         except ValueError:
 474                                                 print 'Warning: unexpected line "%s"' % line
 475                                 else:
 476                                         print 'Warning: unexpected line "%s"' % line
 477                         else:
 478                                 numbers = re.findall('\d+', line)
 479                                 if len(numbers) == 3:
 480                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 481                                         total_lines += inserted
 482                                         total_lines -= deleted
 483                                         self.total_lines_added += inserted
 484                                         self.total_lines_removed += deleted
 485
 486                                 else:
 487                                         print 'Warning: failed to handle line "%s"' % line
 488                                         (files, inserted, deleted) = (0, 0, 0)
 489                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 490                 self.total_lines = total_lines
 491
 492                 # Per-author statistics
 493
 494                 # defined for stamp, author only if author commited at this timestamp.
 495                 self.changes_by_date_by_author = {} # stamp -> author -> lines_added
 496
 497                 # Similar to the above, but never use --first-parent
 498                 # (we need to walk through every commit to know who
 499                 # committed what, not just through mainline)
 500                 lines = getpipeoutput(['git log --shortstat --date-order --pretty=format:"%%at %%aN" %s' % (getcommitrange('HEAD'))]).split('\n')
 501                 lines.reverse()
 502                 files = 0; inserted = 0; deleted = 0
 503                 author = None
 504                 stamp = 0
 505                 for line in lines:
 506                         if len(line) == 0:
 507                                 continue
 508
 509                         # <stamp> <author>
 510                         if line.find('files changed,') == -1:
 511                                 pos = line.find(' ')
 512                                 if pos != -1:
 513                                         try:
 514                                                 oldstamp = stamp
 515                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 516                                                 if oldstamp > stamp:
 517                                                         # clock skew, keep old timestamp to avoid having ugly graph
 518                                                         stamp = oldstamp
 519                                                 if author not in self.authors:
 520                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0, 'commits' : 0}
 521                                                 self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 522                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 523                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 524                                                 if stamp not in self.changes_by_date_by_author:
 525                                                         self.changes_by_date_by_author[stamp] = {}
 526                                                 if author not in self.changes_by_date_by_author[stamp]:
 527                                                         self.changes_by_date_by_author[stamp][author] = {}
 528                                                 self.changes_by_date_by_author[stamp][author]['lines_added'] = self.authors[author]['lines_added']
 529                                                 self.changes_by_date_by_author[stamp][author]['commits'] = self.authors[author]['commits']
 530                                                 files, inserted, deleted = 0, 0, 0
 531                                         except ValueError:
 532                                                 print 'Warning: unexpected line "%s"' % line
 533                                 else:
 534                                         print 'Warning: unexpected line "%s"' % line
 535                         else:
 536                                 numbers = re.findall('\d+', line)
 537                                 if len(numbers) == 3:
 538                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 539                                 else:
 540                                         print 'Warning: failed to handle line "%s"' % line
 541                                         (files, inserted, deleted) = (0, 0, 0)
 542
 543         def refine(self):
 544                 # authors
 545                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 546                 self.authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 547                 self.authors_by_commits.reverse() # most first
 548                 for i, name in enumerate(self.authors_by_commits):
 549                         self.authors[name]['place_by_commits'] = i + 1
 550
 551                 for name in self.authors.keys():
 552                         a = self.authors[name]
 553                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 554                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 555                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 556                         delta = date_last - date_first
 557                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 558                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 559                         a['timedelta'] = delta
 560                         if 'lines_added' not in a: a['lines_added'] = 0
 561                         if 'lines_removed' not in a: a['lines_removed'] = 0
 562
 563         def getActiveDays(self):
 564                 return self.active_days
 565
 566         def getActivityByDayOfWeek(self):
 567                 return self.activity_by_day_of_week
 568
 569         def getActivityByHourOfDay(self):
 570                 return self.activity_by_hour_of_day
 571
 572         def getAuthorInfo(self, author):
 573                 return self.authors[author]
 574
 575         def getAuthors(self, limit = None):
 576                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 577                 res.reverse()
 578                 return res[:limit]
 579
 580         def getCommitDeltaDays(self):
 581                 return (self.last_commit_stamp - self.first_commit_stamp) / 86400 + 1
 582
 583         def getDomainInfo(self, domain):
 584                 return self.domains[domain]
 585
 586         def getDomains(self):
 587                 return self.domains.keys()
 588
 589         def getFilesInCommit(self, rev):
 590                 try:
 591                         res = self.cache['files_in_tree'][rev]
 592                 except:
 593                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 594                         if 'files_in_tree' not in self.cache:
 595                                 self.cache['files_in_tree'] = {}
 596                         self.cache['files_in_tree'][rev] = res
 597
 598                 return res
 599
 600         def getFirstCommitDate(self):
 601                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 602
 603         def getLastCommitDate(self):
 604                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 605
 606         def getLinesInBlob(self, sha1):
 607                 try:
 608                         res = self.cache['lines_in_blob'][sha1]
 609                 except:
 610                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 611                         if 'lines_in_blob' not in self.cache:
 612                                 self.cache['lines_in_blob'] = {}
 613                         self.cache['lines_in_blob'][sha1] = res
 614                 return res
 615
 616         def getTags(self):
 617                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 618                 return lines.split('\n')
 619
 620         def getTagDate(self, tag):
 621                 return self.revToDate('tags/' + tag)
 622
 623         def getTotalAuthors(self):
 624                 return self.total_authors
 625
 626         def getTotalCommits(self):
 627                 return self.total_commits
 628
 629         def getTotalFiles(self):
 630                 return self.total_files
 631
 632         def getTotalLOC(self):
 633                 return self.total_lines
 634
 635         def getTotalSize(self):
 636                 return self.total_size
 637
 638         def revToDate(self, rev):
 639                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 640                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 641
 642 class ReportCreator:
 643         """Creates the actual report based on given data."""
 644         def __init__(self):
 645                 pass
 646
 647         def create(self, data, path):
 648                 self.data = data
 649                 self.path = path
 650
 651 def html_linkify(text):
 652         return text.lower().replace(' ', '_')
 653
 654 def html_header(level, text):
 655         name = html_linkify(text)
 656         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 657
 658 class HTMLReportCreator(ReportCreator):
 659         def create(self, data, path):
 660                 ReportCreator.create(self, data, path)
 661                 self.title = data.projectname
 662
 663                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 664                 binarypath = os.path.dirname(os.path.abspath(__file__))
 665                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 666                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 667                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 668                         for base in basedirs:
 669                                 src = base + '/' + file
 670                                 if os.path.exists(src):
 671                                         shutil.copyfile(src, path + '/' + file)
 672                                         break
 673                         else:
 674                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 675
 676                 f = open(path + "/index.html", 'w')
 677                 format = '%Y-%m-%d %H:%M:%S'
 678                 self.printHeader(f)
 679
 680                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 681
 682                 self.printNav(f)
 683
 684                 f.write('<dl>')
 685                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 686                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 687                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s), %s, %s</dd>' % (getversion(), getgitversion(), getgnuplotversion()))
 688                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 689                 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())))
 690                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 691                 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))
 692                 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()))
 693                 f.write('<dt>Authors</dt><dd>%s (average %.1f commits per author)</dd>' % (data.getTotalAuthors(), (1.0 * data.getTotalCommits()) / data.getTotalAuthors()))
 694                 f.write('</dl>')
 695
 696                 f.write('</body>\n</html>')
 697                 f.close()
 698
 699                 ###
 700                 # Activity
 701                 f = open(path + '/activity.html', 'w')
 702                 self.printHeader(f)
 703                 f.write('<h1>Activity</h1>')
 704                 self.printNav(f)
 705
 706                 #f.write('<h2>Last 30 days</h2>')
 707
 708                 #f.write('<h2>Last 12 months</h2>')
 709
 710                 # Weekly activity
 711                 WEEKS = 32
 712                 f.write(html_header(2, 'Weekly activity'))
 713                 f.write('<p>Last %d weeks</p>' % WEEKS)
 714
 715                 # generate weeks to show (previous N weeks from now)
 716                 now = datetime.datetime.now()
 717                 deltaweek = datetime.timedelta(7)
 718                 weeks = []
 719                 stampcur = now
 720                 for i in range(0, WEEKS):
 721                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 722                         stampcur -= deltaweek
 723
 724                 # top row: commits & bar
 725                 f.write('<table class="noborders"><tr>')
 726                 for i in range(0, WEEKS):
 727                         commits = 0
 728                         if weeks[i] in data.activity_by_year_week:
 729                                 commits = data.activity_by_year_week[weeks[i]]
 730
 731                         percentage = 0
 732                         if weeks[i] in data.activity_by_year_week:
 733                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 734                         height = max(1, int(200 * percentage))
 735                         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))
 736
 737                 # bottom row: year/week
 738                 f.write('</tr><tr>')
 739                 for i in range(0, WEEKS):
 740                         f.write('<td>%s</td>' % (WEEKS - i))
 741                 f.write('</tr></table>')
 742
 743                 # Hour of Day
 744                 f.write(html_header(2, 'Hour of Day'))
 745                 hour_of_day = data.getActivityByHourOfDay()
 746                 f.write('<table><tr><th>Hour</th>')
 747                 for i in range(0, 24):
 748                         f.write('<th>%d</th>' % i)
 749                 f.write('</tr>\n<tr><th>Commits</th>')
 750                 fp = open(path + '/hour_of_day.dat', 'w')
 751                 for i in range(0, 24):
 752                         if i in hour_of_day:
 753                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 754                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 755                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 756                         else:
 757                                 f.write('<td>0</td>')
 758                                 fp.write('%d 0\n' % i)
 759                 fp.close()
 760                 f.write('</tr>\n<tr><th>%</th>')
 761                 totalcommits = data.getTotalCommits()
 762                 for i in range(0, 24):
 763                         if i in hour_of_day:
 764                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 765                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 766                         else:
 767                                 f.write('<td>0.00</td>')
 768                 f.write('</tr></table>')
 769                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 770                 fg = open(path + '/hour_of_day.dat', 'w')
 771                 for i in range(0, 24):
 772                         if i in hour_of_day:
 773                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 774                         else:
 775                                 fg.write('%d 0\n' % (i + 1))
 776                 fg.close()
 777
 778                 # Day of Week
 779                 f.write(html_header(2, 'Day of Week'))
 780                 day_of_week = data.getActivityByDayOfWeek()
 781                 f.write('<div class="vtable"><table>')
 782                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 783                 fp = open(path + '/day_of_week.dat', 'w')
 784                 for d in range(0, 7):
 785                         commits = 0
 786                         if d in day_of_week:
 787                                 commits = day_of_week[d]
 788                         fp.write('%d %s %d\n' % (d + 1, WEEKDAYS[d], commits))
 789                         f.write('<tr>')
 790                         f.write('<th>%s</th>' % (WEEKDAYS[d]))
 791                         if d in day_of_week:
 792                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 793                         else:
 794                                 f.write('<td>0</td>')
 795                         f.write('</tr>')
 796                 f.write('</table></div>')
 797                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 798                 fp.close()
 799
 800                 # Hour of Week
 801                 f.write(html_header(2, 'Hour of Week'))
 802                 f.write('<table>')
 803
 804                 f.write('<tr><th>Weekday</th>')
 805                 for hour in range(0, 24):
 806                         f.write('<th>%d</th>' % (hour))
 807                 f.write('</tr>')
 808
 809                 for weekday in range(0, 7):
 810                         f.write('<tr><th>%s</th>' % (WEEKDAYS[weekday]))
 811                         for hour in range(0, 24):
 812                                 try:
 813                                         commits = data.activity_by_hour_of_week[weekday][hour]
 814                                 except KeyError:
 815                                         commits = 0
 816                                 if commits != 0:
 817                                         f.write('<td')
 818                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 819                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 820                                         f.write('>%d</td>' % commits)
 821                                 else:
 822                                         f.write('<td></td>')
 823                         f.write('</tr>')
 824
 825                 f.write('</table>')
 826
 827                 # Month of Year
 828                 f.write(html_header(2, 'Month of Year'))
 829                 f.write('<div class="vtable"><table>')
 830                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 831                 fp = open (path + '/month_of_year.dat', 'w')
 832                 for mm in range(1, 13):
 833                         commits = 0
 834                         if mm in data.activity_by_month_of_year:
 835                                 commits = data.activity_by_month_of_year[mm]
 836                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 837                         fp.write('%d %d\n' % (mm, commits))
 838                 fp.close()
 839                 f.write('</table></div>')
 840                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 841
 842                 # Commits by year/month
 843                 f.write(html_header(2, 'Commits by year/month'))
 844                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th><th>Lines added</th><th>Lines removed</th></tr>')
 845                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 846                         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)))
 847                 f.write('</table></div>')
 848                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 849                 fg = open(path + '/commits_by_year_month.dat', 'w')
 850                 for yymm in sorted(data.commits_by_month.keys()):
 851                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 852                 fg.close()
 853
 854                 # Commits by year
 855                 f.write(html_header(2, 'Commits by Year'))
 856                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th><th>Lines added</th><th>Lines removed</th></tr>')
 857                 for yy in reversed(sorted(data.commits_by_year.keys())):
 858                         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)))
 859                 f.write('</table></div>')
 860                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 861                 fg = open(path + '/commits_by_year.dat', 'w')
 862                 for yy in sorted(data.commits_by_year.keys()):
 863                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 864                 fg.close()
 865
 866                 # Commits by timezone
 867                 f.write(html_header(2, 'Commits by Timezone'))
 868                 f.write('<table><tr>')
 869                 f.write('<th>Timezone</th><th>Commits</th>')
 870                 max_commits_on_tz = max(data.commits_by_timezone.values())
 871                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 872                         commits = data.commits_by_timezone[i]
 873                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 874                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 875                 f.write('</tr></table>')
 876
 877                 f.write('</body></html>')
 878                 f.close()
 879
 880                 ###
 881                 # Authors
 882                 f = open(path + '/authors.html', 'w')
 883                 self.printHeader(f)
 884
 885                 f.write('<h1>Authors</h1>')
 886                 self.printNav(f)
 887
 888                 # Authors :: List of authors
 889                 f.write(html_header(2, 'List of Authors'))
 890
 891                 f.write('<table class="authors sortable" id="authors">')
 892                 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>')
 893                 for author in data.getAuthors(conf['max_authors']):
 894                         info = data.getAuthorInfo(author)
 895                         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']))
 896                 f.write('</table>')
 897
 898                 allauthors = data.getAuthors()
 899                 if len(allauthors) > conf['max_authors']:
 900                         rest = allauthors[conf['max_authors']:]
 901                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 902
 903                 f.write(html_header(2, 'Cumulated Added Lines of Code per Author'))
 904                 f.write('<img src="lines_of_code_by_author.png" alt="Lines of code per Author" />')
 905                 if len(allauthors) > conf['max_authors']:
 906                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 907
 908                 f.write(html_header(2, 'Commits per Author'))
 909                 f.write('<img src="commits_by_author.png" alt="Commits per Author" />')
 910                 if len(allauthors) > conf['max_authors']:
 911                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 912
 913                 fgl = open(path + '/lines_of_code_by_author.dat', 'w')
 914                 fgc = open(path + '/commits_by_author.dat', 'w')
 915
 916                 lines_by_authors = {} # cumulated added lines by
 917                 # author. to save memory,
 918                 # changes_by_date_by_author[stamp][author] is defined
 919                 # only at points where author commits.
 920                 # lines_by_authors allows us to generate all the
 921                 # points in the .dat file.
 922
 923                 # Don't rely on getAuthors to give the same order each
 924                 # time. Be robust and keep the list in a variable.
 925                 commits_by_authors = {} # cumulated added lines by
 926
 927                 self.authors_to_plot = data.getAuthors(conf['max_authors'])
 928                 for author in self.authors_to_plot:
 929                         lines_by_authors[author] = 0
 930                         commits_by_authors[author] = 0
 931                 for stamp in sorted(data.changes_by_date_by_author.keys()):
 932                         fgl.write('%d' % stamp)
 933                         fgc.write('%d' % stamp)
 934                         for author in self.authors_to_plot:
 935                                 if author in data.changes_by_date_by_author[stamp].keys():
 936                                         lines_by_authors[author] = data.changes_by_date_by_author[stamp][author]['lines_added']
 937                                         commits_by_authors[author] = data.changes_by_date_by_author[stamp][author]['commits']
 938                                 fgl.write(' %d' % lines_by_authors[author])
 939                                 fgc.write(' %d' % commits_by_authors[author])
 940                         fgl.write('\n')
 941                         fgc.write('\n')
 942                 fgl.close()
 943                 fgc.close()
 944
 945                 # Authors :: Author of Month
 946                 f.write(html_header(2, 'Author of Month'))
 947                 f.write('<table class="sortable" id="aom">')
 948                 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'])
 949                 for yymm in reversed(sorted(data.author_of_month.keys())):
 950                         authordict = data.author_of_month[yymm]
 951                         authors = getkeyssortedbyvalues(authordict)
 952                         authors.reverse()
 953                         commits = data.author_of_month[yymm][authors[0]]
 954                         next = ', '.join(authors[1:conf['authors_top']+1])
 955                         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)))
 956
 957                 f.write('</table>')
 958
 959                 f.write(html_header(2, 'Author of Year'))
 960                 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'])
 961                 for yy in reversed(sorted(data.author_of_year.keys())):
 962                         authordict = data.author_of_year[yy]
 963                         authors = getkeyssortedbyvalues(authordict)
 964                         authors.reverse()
 965                         commits = data.author_of_year[yy][authors[0]]
 966                         next = ', '.join(authors[1:conf['authors_top']+1])
 967                         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)))
 968                 f.write('</table>')
 969
 970                 # Domains
 971                 f.write(html_header(2, 'Commits by Domains'))
 972                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
 973                 domains_by_commits.reverse() # most first
 974                 f.write('<div class="vtable"><table>')
 975                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
 976                 fp = open(path + '/domains.dat', 'w')
 977                 n = 0
 978                 for domain in domains_by_commits:
 979                         if n == conf['max_domains']:
 980                                 break
 981                         commits = 0
 982                         n += 1
 983                         info = data.getDomainInfo(domain)
 984                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
 985                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
 986                 f.write('</table></div>')
 987                 f.write('<img src="domains.png" alt="Commits by Domains" />')
 988                 fp.close()
 989
 990                 f.write('</body></html>')
 991                 f.close()
 992
 993                 ###
 994                 # Files
 995                 f = open(path + '/files.html', 'w')
 996                 self.printHeader(f)
 997                 f.write('<h1>Files</h1>')
 998                 self.printNav(f)
 999
1000                 f.write('<dl>\n')
1001                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
1002                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1003                 f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % (float(data.getTotalSize()) / data.getTotalFiles()))
1004                 f.write('</dl>\n')
1005
1006                 # Files :: File count by date
1007                 f.write(html_header(2, 'File count by date'))
1008
1009                 # use set to get rid of duplicate/unnecessary entries
1010                 files_by_date = set()
1011                 for stamp in sorted(data.files_by_stamp.keys()):
1012                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1013
1014                 fg = open(path + '/files_by_date.dat', 'w')
1015                 for line in sorted(list(files_by_date)):
1016                         fg.write('%s\n' % line)
1017                 #for stamp in sorted(data.files_by_stamp.keys()):
1018                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1019                 fg.close()
1020
1021                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
1022
1023                 #f.write('<h2>Average file size by date</h2>')
1024
1025                 # Files :: Extensions
1026                 f.write(html_header(2, 'Extensions'))
1027                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
1028                 for ext in sorted(data.extensions.keys()):
1029                         files = data.extensions[ext]['files']
1030                         lines = data.extensions[ext]['lines']
1031                         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))
1032                 f.write('</table>')
1033
1034                 f.write('</body></html>')
1035                 f.close()
1036
1037                 ###
1038                 # Lines
1039                 f = open(path + '/lines.html', 'w')
1040                 self.printHeader(f)
1041                 f.write('<h1>Lines</h1>')
1042                 self.printNav(f)
1043
1044                 f.write('<dl>\n')
1045                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1046                 f.write('</dl>\n')
1047
1048                 f.write(html_header(2, 'Lines of Code'))
1049                 f.write('<img src="lines_of_code.png" />')
1050
1051                 fg = open(path + '/lines_of_code.dat', 'w')
1052                 for stamp in sorted(data.changes_by_date.keys()):
1053                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
1054                 fg.close()
1055
1056                 f.write('</body></html>')
1057                 f.close()
1058
1059                 ###
1060                 # tags.html
1061                 f = open(path + '/tags.html', 'w')
1062                 self.printHeader(f)
1063                 f.write('<h1>Tags</h1>')
1064                 self.printNav(f)
1065
1066                 f.write('<dl>')
1067                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
1068                 if len(data.tags) > 0:
1069                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
1070                 f.write('</dl>')
1071
1072                 f.write('<table class="tags">')
1073                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
1074                 # sort the tags by date desc
1075                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
1076                 for tag in tags_sorted_by_date_desc:
1077                         authorinfo = []
1078                         self.authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
1079                         for i in reversed(self.authors_by_commits):
1080                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
1081                         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)))
1082                 f.write('</table>')
1083
1084                 f.write('</body></html>')
1085                 f.close()
1086
1087                 self.createGraphs(path)
1088
1089         def createGraphs(self, path):
1090                 print 'Generating graphs...'
1091
1092                 # hour of day
1093                 f = open(path + '/hour_of_day.plot', 'w')
1094                 f.write(GNUPLOT_COMMON)
1095                 f.write(
1096 """
1097 set output 'hour_of_day.png'
1098 unset key
1099 set xrange [0.5:24.5]
1100 set xtics 4
1101 set grid y
1102 set ylabel "Commits"
1103 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
1104 """)
1105                 f.close()
1106
1107                 # day of week
1108                 f = open(path + '/day_of_week.plot', 'w')
1109                 f.write(GNUPLOT_COMMON)
1110                 f.write(
1111 """
1112 set output 'day_of_week.png'
1113 unset key
1114 set xrange [0.5:7.5]
1115 set xtics 1
1116 set grid y
1117 set ylabel "Commits"
1118 plot 'day_of_week.dat' using 1:3:(0.5):xtic(2) w boxes fs solid
1119 """)
1120                 f.close()
1121
1122                 # Domains
1123                 f = open(path + '/domains.plot', 'w')
1124                 f.write(GNUPLOT_COMMON)
1125                 f.write(
1126 """
1127 set output 'domains.png'
1128 unset key
1129 unset xtics
1130 set yrange [0:]
1131 set grid y
1132 set ylabel "Commits"
1133 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
1134 """)
1135                 f.close()
1136
1137                 # Month of Year
1138                 f = open(path + '/month_of_year.plot', 'w')
1139                 f.write(GNUPLOT_COMMON)
1140                 f.write(
1141 """
1142 set output 'month_of_year.png'
1143 unset key
1144 set xrange [0.5:12.5]
1145 set xtics 1
1146 set grid y
1147 set ylabel "Commits"
1148 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
1149 """)
1150                 f.close()
1151
1152                 # commits_by_year_month
1153                 f = open(path + '/commits_by_year_month.plot', 'w')
1154                 f.write(GNUPLOT_COMMON)
1155                 f.write(
1156 """
1157 set output 'commits_by_year_month.png'
1158 unset key
1159 set xdata time
1160 set timefmt "%Y-%m"
1161 set format x "%Y-%m"
1162 set xtics rotate
1163 set bmargin 5
1164 set grid y
1165 set ylabel "Commits"
1166 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1167 """)
1168                 f.close()
1169
1170                 # commits_by_year
1171                 f = open(path + '/commits_by_year.plot', 'w')
1172                 f.write(GNUPLOT_COMMON)
1173                 f.write(
1174 """
1175 set output 'commits_by_year.png'
1176 unset key
1177 set xtics 1 rotate
1178 set grid y
1179 set ylabel "Commits"
1180 set yrange [0:]
1181 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1182 """)
1183                 f.close()
1184
1185                 # Files by date
1186                 f = open(path + '/files_by_date.plot', 'w')
1187                 f.write(GNUPLOT_COMMON)
1188                 f.write(
1189 """
1190 set output 'files_by_date.png'
1191 unset key
1192 set xdata time
1193 set timefmt "%Y-%m-%d"
1194 set format x "%Y-%m-%d"
1195 set grid y
1196 set ylabel "Files"
1197 set xtics rotate
1198 set ytics autofreq
1199 set bmargin 6
1200 plot 'files_by_date.dat' using 1:2 w steps
1201 """)
1202                 f.close()
1203
1204                 # Lines of Code
1205                 f = open(path + '/lines_of_code.plot', 'w')
1206                 f.write(GNUPLOT_COMMON)
1207                 f.write(
1208 """
1209 set output 'lines_of_code.png'
1210 unset key
1211 set xdata time
1212 set timefmt "%s"
1213 set format x "%Y-%m-%d"
1214 set grid y
1215 set ylabel "Lines"
1216 set xtics rotate
1217 set bmargin 6
1218 plot 'lines_of_code.dat' using 1:2 w lines
1219 """)
1220                 f.close()
1221
1222                 # Lines of Code Added per author
1223                 f = open(path + '/lines_of_code_by_author.plot', 'w')
1224                 f.write(GNUPLOT_COMMON)
1225                 f.write(
1226 """
1227 set terminal png transparent size 640,480
1228 set output 'lines_of_code_by_author.png'
1229 set key left top
1230 set xdata time
1231 set timefmt "%s"
1232 set format x "%Y-%m-%d"
1233 set grid y
1234 set ylabel "Lines"
1235 set xtics rotate
1236 set bmargin 6
1237 plot """
1238 )
1239                 i = 1
1240                 plots = []
1241                 for a in self.authors_to_plot:
1242                         i = i + 1
1243                         plots.append("""'lines_of_code_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1244                 f.write(", ".join(plots))
1245                 f.write('\n')
1246
1247                 f.close()
1248
1249                 # Commits per author
1250                 f = open(path + '/commits_by_author.plot', 'w')
1251                 f.write(GNUPLOT_COMMON)
1252                 f.write(
1253 """
1254 set terminal png transparent size 640,480
1255 set output 'commits_by_author.png'
1256 set key left top
1257 set xdata time
1258 set timefmt "%s"
1259 set format x "%Y-%m-%d"
1260 set grid y
1261 set ylabel "Commits"
1262 set xtics rotate
1263 set bmargin 6
1264 plot """
1265 )
1266                 i = 1
1267                 plots = []
1268                 for a in self.authors_to_plot:
1269                         i = i + 1
1270                         plots.append("""'commits_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1271                 f.write(", ".join(plots))
1272                 f.write('\n')
1273
1274                 f.close()
1275
1276                 os.chdir(path)
1277                 files = glob.glob(path + '/*.plot')
1278                 for f in files:
1279                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1280                         if len(out) > 0:
1281                                 print out
1282
1283         def printHeader(self, f, title = ''):
1284                 f.write(
1285 """<?xml version="1.0" encoding="UTF-8"?>
1286 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1287 <html xmlns="http://www.w3.org/1999/xhtml">
1288 <head>
1289         <title>GitStats - %s</title>
1290         <link rel="stylesheet" href="%s" type="text/css" />
1291         <meta name="generator" content="GitStats %s" />
1292         <script type="text/javascript" src="sortable.js"></script>
1293 </head>
1294 <body>
1295 """ % (self.title, conf['style'], getversion()))
1296
1297         def printNav(self, f):
1298                 f.write("""
1299 <div class="nav">
1300 <ul>
1301 <li><a href="index.html">General</a></li>
1302 <li><a href="activity.html">Activity</a></li>
1303 <li><a href="authors.html">Authors</a></li>
1304 <li><a href="files.html">Files</a></li>
1305 <li><a href="lines.html">Lines</a></li>
1306 <li><a href="tags.html">Tags</a></li>
1307 </ul>
1308 </div>
1309 """)
1310
1311
1312 class GitStats:
1313         def run(self, args_orig):
1314                 optlist, args = getopt.getopt(args_orig, 'c:')
1315                 for o,v in optlist:
1316                         if o == '-c':
1317                                 key, value = v.split('=', 1)
1318                                 if key not in conf:
1319                                         raise KeyError('no such key "%s" in config' % key)
1320                                 if isinstance(conf[key], int):
1321                                         conf[key] = int(value)
1322                                 else:
1323                                         conf[key] = value
1324
1325                 if len(args) < 2:
1326                         print """
1327 Usage: gitstats [options] <gitpath..> <outputpath>
1328
1329 Options:
1330 -c key=value     Override configuration value
1331
1332 Default config values:
1333 %s
1334 """ % conf
1335                         sys.exit(0)
1336
1337                 outputpath = os.path.abspath(args[-1])
1338                 rundir = os.getcwd()
1339
1340                 try:
1341                         os.makedirs(outputpath)
1342                 except OSError:
1343                         pass
1344                 if not os.path.isdir(outputpath):
1345                         print 'FATAL: Output path is not a directory or does not exist'
1346                         sys.exit(1)
1347
1348                 print 'Output path: %s' % outputpath
1349                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1350
1351                 data = GitDataCollector()
1352                 data.loadCache(cachefile)
1353
1354                 for gitpath in args[0:-1]:
1355                         print 'Git path: %s' % gitpath
1356
1357                         os.chdir(gitpath)
1358
1359                         print 'Collecting data...'
1360                         data.collect(gitpath)
1361
1362                 print 'Refining data...'
1363                 data.saveCache(cachefile)
1364                 data.refine()
1365
1366                 os.chdir(rundir)
1367
1368                 print 'Generating report...'
1369                 report = HTMLReportCreator()
1370                 report.create(data, outputpath)
1371
1372                 time_end = time.time()
1373                 exectime_internal = time_end - time_start
1374                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1375                 if sys.stdin.isatty():
1376                         print 'You may now run:'
1377                         print
1378                         print '   sensible-browser \'%s\'' % os.path.join(outputpath, 'index.html').replace("'", "'\\''")
1379                         print
1380
1381 if __name__=='__main__':
1382         g = GitStats()
1383         g.run(sys.argv[1:])
1384