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                 self.total_size = 0
 410                 for line in lines:
 411                         if len(line) == 0:
 412                                 continue
 413                         parts = re.split('\s+', line, 5)
 414                         sha1 = parts[2]
 415                         size = int(parts[3])
 416                         fullpath = parts[4]
 417
 418                         self.total_size += size
 419                         self.total_files += 1
 420
 421                         filename = fullpath.split('/')[-1] # strip directories
 422                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 423                                 ext = ''
 424                         else:
 425                                 ext = filename[(filename.rfind('.') + 1):]
 426                         if len(ext) > conf['max_ext_length']:
 427                                 ext = ''
 428
 429                         if ext not in self.extensions:
 430                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 431
 432                         self.extensions[ext]['files'] += 1
 433                         try:
 434                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 435                         except:
 436                                 print 'Warning: Could not count lines for file "%s"' % line
 437
 438                 # line statistics
 439                 # outputs:
 440                 #  N files changed, N insertions (+), N deletions(-)
 441                 # <stamp> <author>
 442                 self.changes_by_date = {} # stamp -> { files, ins, del }
 443                 # computation of lines of code by date is better done
 444                 # on a linear history.
 445                 extra = ''
 446                 if conf['linear_linestats']:
 447                         extra = '--first-parent -m'
 448                 lines = getpipeoutput(['git log --shortstat %s --pretty=format:"%%at %%aN" %s' % (extra, getcommitrange('HEAD'))]).split('\n')
 449                 lines.reverse()
 450                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 451                 author = None
 452                 for line in lines:
 453                         if len(line) == 0:
 454                                 continue
 455
 456                         # <stamp> <author>
 457                         if line.find('files changed,') == -1:
 458                                 pos = line.find(' ')
 459                                 if pos != -1:
 460                                         try:
 461                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 462                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 463
 464                                                 date = datetime.datetime.fromtimestamp(stamp)
 465                                                 yymm = date.strftime('%Y-%m')
 466                                                 self.lines_added_by_month[yymm] = self.lines_added_by_month.get(yymm, 0) + inserted
 467                                                 self.lines_removed_by_month[yymm] = self.lines_removed_by_month.get(yymm, 0) + deleted
 468
 469                                                 yy = date.year
 470                                                 self.lines_added_by_year[yy] = self.lines_added_by_year.get(yy,0) + inserted
 471                                                 self.lines_removed_by_year[yy] = self.lines_removed_by_year.get(yy, 0) + deleted
 472
 473                                                 files, inserted, deleted = 0, 0, 0
 474                                         except ValueError:
 475                                                 print 'Warning: unexpected line "%s"' % line
 476                                 else:
 477                                         print 'Warning: unexpected line "%s"' % line
 478                         else:
 479                                 numbers = re.findall('\d+', line)
 480                                 if len(numbers) == 3:
 481                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 482                                         total_lines += inserted
 483                                         total_lines -= deleted
 484                                         self.total_lines_added += inserted
 485                                         self.total_lines_removed += deleted
 486
 487                                 else:
 488                                         print 'Warning: failed to handle line "%s"' % line
 489                                         (files, inserted, deleted) = (0, 0, 0)
 490                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 491                 self.total_lines = total_lines
 492
 493                 # Per-author statistics
 494
 495                 # defined for stamp, author only if author commited at this timestamp.
 496                 self.changes_by_date_by_author = {} # stamp -> author -> lines_added
 497
 498                 # Similar to the above, but never use --first-parent
 499                 # (we need to walk through every commit to know who
 500                 # committed what, not just through mainline)
 501                 lines = getpipeoutput(['git log --shortstat --date-order --pretty=format:"%%at %%aN" %s' % (getcommitrange('HEAD'))]).split('\n')
 502                 lines.reverse()
 503                 files = 0; inserted = 0; deleted = 0
 504                 author = None
 505                 stamp = 0
 506                 for line in lines:
 507                         if len(line) == 0:
 508                                 continue
 509
 510                         # <stamp> <author>
 511                         if line.find('files changed,') == -1:
 512                                 pos = line.find(' ')
 513                                 if pos != -1:
 514                                         try:
 515                                                 oldstamp = stamp
 516                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 517                                                 if oldstamp > stamp:
 518                                                         # clock skew, keep old timestamp to avoid having ugly graph
 519                                                         stamp = oldstamp
 520                                                 if author not in self.authors:
 521                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0, 'commits' : 0}
 522                                                 self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 523                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 524                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 525                                                 if stamp not in self.changes_by_date_by_author:
 526                                                         self.changes_by_date_by_author[stamp] = {}
 527                                                 if author not in self.changes_by_date_by_author[stamp]:
 528                                                         self.changes_by_date_by_author[stamp][author] = {}
 529                                                 self.changes_by_date_by_author[stamp][author]['lines_added'] = self.authors[author]['lines_added']
 530                                                 self.changes_by_date_by_author[stamp][author]['commits'] = self.authors[author]['commits']
 531                                                 files, inserted, deleted = 0, 0, 0
 532                                         except ValueError:
 533                                                 print 'Warning: unexpected line "%s"' % line
 534                                 else:
 535                                         print 'Warning: unexpected line "%s"' % line
 536                         else:
 537                                 numbers = re.findall('\d+', line)
 538                                 if len(numbers) == 3:
 539                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 540                                 else:
 541                                         print 'Warning: failed to handle line "%s"' % line
 542                                         (files, inserted, deleted) = (0, 0, 0)
 543
 544         def refine(self):
 545                 # authors
 546                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 547                 self.authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 548                 self.authors_by_commits.reverse() # most first
 549                 for i, name in enumerate(self.authors_by_commits):
 550                         self.authors[name]['place_by_commits'] = i + 1
 551
 552                 for name in self.authors.keys():
 553                         a = self.authors[name]
 554                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 555                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 556                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 557                         delta = date_last - date_first
 558                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 559                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 560                         a['timedelta'] = delta
 561                         if 'lines_added' not in a: a['lines_added'] = 0
 562                         if 'lines_removed' not in a: a['lines_removed'] = 0
 563
 564         def getActiveDays(self):
 565                 return self.active_days
 566
 567         def getActivityByDayOfWeek(self):
 568                 return self.activity_by_day_of_week
 569
 570         def getActivityByHourOfDay(self):
 571                 return self.activity_by_hour_of_day
 572
 573         def getAuthorInfo(self, author):
 574                 return self.authors[author]
 575
 576         def getAuthors(self, limit = None):
 577                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 578                 res.reverse()
 579                 return res[:limit]
 580
 581         def getCommitDeltaDays(self):
 582                 return (self.last_commit_stamp - self.first_commit_stamp) / 86400 + 1
 583
 584         def getDomainInfo(self, domain):
 585                 return self.domains[domain]
 586
 587         def getDomains(self):
 588                 return self.domains.keys()
 589
 590         def getFilesInCommit(self, rev):
 591                 try:
 592                         res = self.cache['files_in_tree'][rev]
 593                 except:
 594                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 595                         if 'files_in_tree' not in self.cache:
 596                                 self.cache['files_in_tree'] = {}
 597                         self.cache['files_in_tree'][rev] = res
 598
 599                 return res
 600
 601         def getFirstCommitDate(self):
 602                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 603
 604         def getLastCommitDate(self):
 605                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 606
 607         def getLinesInBlob(self, sha1):
 608                 try:
 609                         res = self.cache['lines_in_blob'][sha1]
 610                 except:
 611                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 612                         if 'lines_in_blob' not in self.cache:
 613                                 self.cache['lines_in_blob'] = {}
 614                         self.cache['lines_in_blob'][sha1] = res
 615                 return res
 616
 617         def getTags(self):
 618                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 619                 return lines.split('\n')
 620
 621         def getTagDate(self, tag):
 622                 return self.revToDate('tags/' + tag)
 623
 624         def getTotalAuthors(self):
 625                 return self.total_authors
 626
 627         def getTotalCommits(self):
 628                 return self.total_commits
 629
 630         def getTotalFiles(self):
 631                 return self.total_files
 632
 633         def getTotalLOC(self):
 634                 return self.total_lines
 635
 636         def getTotalSize(self):
 637                 return self.total_size
 638
 639         def revToDate(self, rev):
 640                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 641                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 642
 643 class ReportCreator:
 644         """Creates the actual report based on given data."""
 645         def __init__(self):
 646                 pass
 647
 648         def create(self, data, path):
 649                 self.data = data
 650                 self.path = path
 651
 652 def html_linkify(text):
 653         return text.lower().replace(' ', '_')
 654
 655 def html_header(level, text):
 656         name = html_linkify(text)
 657         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 658
 659 class HTMLReportCreator(ReportCreator):
 660         def create(self, data, path):
 661                 ReportCreator.create(self, data, path)
 662                 self.title = data.projectname
 663
 664                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 665                 binarypath = os.path.dirname(os.path.abspath(__file__))
 666                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 667                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 668                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 669                         for base in basedirs:
 670                                 src = base + '/' + file
 671                                 if os.path.exists(src):
 672                                         shutil.copyfile(src, path + '/' + file)
 673                                         break
 674                         else:
 675                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 676
 677                 f = open(path + "/index.html", 'w')
 678                 format = '%Y-%m-%d %H:%M:%S'
 679                 self.printHeader(f)
 680
 681                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 682
 683                 self.printNav(f)
 684
 685                 f.write('<dl>')
 686                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 687                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 688                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s), %s, %s</dd>' % (getversion(), getgitversion(), getgnuplotversion()))
 689                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 690                 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())))
 691                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 692                 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))
 693                 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()))
 694                 f.write('<dt>Authors</dt><dd>%s (average %.1f commits per author)</dd>' % (data.getTotalAuthors(), (1.0 * data.getTotalCommits()) / data.getTotalAuthors()))
 695                 f.write('</dl>')
 696
 697                 f.write('</body>\n</html>')
 698                 f.close()
 699
 700                 ###
 701                 # Activity
 702                 f = open(path + '/activity.html', 'w')
 703                 self.printHeader(f)
 704                 f.write('<h1>Activity</h1>')
 705                 self.printNav(f)
 706
 707                 #f.write('<h2>Last 30 days</h2>')
 708
 709                 #f.write('<h2>Last 12 months</h2>')
 710
 711                 # Weekly activity
 712                 WEEKS = 32
 713                 f.write(html_header(2, 'Weekly activity'))
 714                 f.write('<p>Last %d weeks</p>' % WEEKS)
 715
 716                 # generate weeks to show (previous N weeks from now)
 717                 now = datetime.datetime.now()
 718                 deltaweek = datetime.timedelta(7)
 719                 weeks = []
 720                 stampcur = now
 721                 for i in range(0, WEEKS):
 722                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 723                         stampcur -= deltaweek
 724
 725                 # top row: commits & bar
 726                 f.write('<table class="noborders"><tr>')
 727                 for i in range(0, WEEKS):
 728                         commits = 0
 729                         if weeks[i] in data.activity_by_year_week:
 730                                 commits = data.activity_by_year_week[weeks[i]]
 731
 732                         percentage = 0
 733                         if weeks[i] in data.activity_by_year_week:
 734                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 735                         height = max(1, int(200 * percentage))
 736                         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))
 737
 738                 # bottom row: year/week
 739                 f.write('</tr><tr>')
 740                 for i in range(0, WEEKS):
 741                         f.write('<td>%s</td>' % (WEEKS - i))
 742                 f.write('</tr></table>')
 743
 744                 # Hour of Day
 745                 f.write(html_header(2, 'Hour of Day'))
 746                 hour_of_day = data.getActivityByHourOfDay()
 747                 f.write('<table><tr><th>Hour</th>')
 748                 for i in range(0, 24):
 749                         f.write('<th>%d</th>' % i)
 750                 f.write('</tr>\n<tr><th>Commits</th>')
 751                 fp = open(path + '/hour_of_day.dat', 'w')
 752                 for i in range(0, 24):
 753                         if i in hour_of_day:
 754                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 755                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 756                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 757                         else:
 758                                 f.write('<td>0</td>')
 759                                 fp.write('%d 0\n' % i)
 760                 fp.close()
 761                 f.write('</tr>\n<tr><th>%</th>')
 762                 totalcommits = data.getTotalCommits()
 763                 for i in range(0, 24):
 764                         if i in hour_of_day:
 765                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 766                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 767                         else:
 768                                 f.write('<td>0.00</td>')
 769                 f.write('</tr></table>')
 770                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 771                 fg = open(path + '/hour_of_day.dat', 'w')
 772                 for i in range(0, 24):
 773                         if i in hour_of_day:
 774                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 775                         else:
 776                                 fg.write('%d 0\n' % (i + 1))
 777                 fg.close()
 778
 779                 # Day of Week
 780                 f.write(html_header(2, 'Day of Week'))
 781                 day_of_week = data.getActivityByDayOfWeek()
 782                 f.write('<div class="vtable"><table>')
 783                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 784                 fp = open(path + '/day_of_week.dat', 'w')
 785                 for d in range(0, 7):
 786                         commits = 0
 787                         if d in day_of_week:
 788                                 commits = day_of_week[d]
 789                         fp.write('%d %s %d\n' % (d + 1, WEEKDAYS[d], commits))
 790                         f.write('<tr>')
 791                         f.write('<th>%s</th>' % (WEEKDAYS[d]))
 792                         if d in day_of_week:
 793                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 794                         else:
 795                                 f.write('<td>0</td>')
 796                         f.write('</tr>')
 797                 f.write('</table></div>')
 798                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 799                 fp.close()
 800
 801                 # Hour of Week
 802                 f.write(html_header(2, 'Hour of Week'))
 803                 f.write('<table>')
 804
 805                 f.write('<tr><th>Weekday</th>')
 806                 for hour in range(0, 24):
 807                         f.write('<th>%d</th>' % (hour))
 808                 f.write('</tr>')
 809
 810                 for weekday in range(0, 7):
 811                         f.write('<tr><th>%s</th>' % (WEEKDAYS[weekday]))
 812                         for hour in range(0, 24):
 813                                 try:
 814                                         commits = data.activity_by_hour_of_week[weekday][hour]
 815                                 except KeyError:
 816                                         commits = 0
 817                                 if commits != 0:
 818                                         f.write('<td')
 819                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 820                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 821                                         f.write('>%d</td>' % commits)
 822                                 else:
 823                                         f.write('<td></td>')
 824                         f.write('</tr>')
 825
 826                 f.write('</table>')
 827
 828                 # Month of Year
 829                 f.write(html_header(2, 'Month of Year'))
 830                 f.write('<div class="vtable"><table>')
 831                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 832                 fp = open (path + '/month_of_year.dat', 'w')
 833                 for mm in range(1, 13):
 834                         commits = 0
 835                         if mm in data.activity_by_month_of_year:
 836                                 commits = data.activity_by_month_of_year[mm]
 837                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 838                         fp.write('%d %d\n' % (mm, commits))
 839                 fp.close()
 840                 f.write('</table></div>')
 841                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 842
 843                 # Commits by year/month
 844                 f.write(html_header(2, 'Commits by year/month'))
 845                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th><th>Lines added</th><th>Lines removed</th></tr>')
 846                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 847                         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)))
 848                 f.write('</table></div>')
 849                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 850                 fg = open(path + '/commits_by_year_month.dat', 'w')
 851                 for yymm in sorted(data.commits_by_month.keys()):
 852                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 853                 fg.close()
 854
 855                 # Commits by year
 856                 f.write(html_header(2, 'Commits by Year'))
 857                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th><th>Lines added</th><th>Lines removed</th></tr>')
 858                 for yy in reversed(sorted(data.commits_by_year.keys())):
 859                         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)))
 860                 f.write('</table></div>')
 861                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 862                 fg = open(path + '/commits_by_year.dat', 'w')
 863                 for yy in sorted(data.commits_by_year.keys()):
 864                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 865                 fg.close()
 866
 867                 # Commits by timezone
 868                 f.write(html_header(2, 'Commits by Timezone'))
 869                 f.write('<table><tr>')
 870                 f.write('<th>Timezone</th><th>Commits</th>')
 871                 max_commits_on_tz = max(data.commits_by_timezone.values())
 872                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 873                         commits = data.commits_by_timezone[i]
 874                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 875                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 876                 f.write('</tr></table>')
 877
 878                 f.write('</body></html>')
 879                 f.close()
 880
 881                 ###
 882                 # Authors
 883                 f = open(path + '/authors.html', 'w')
 884                 self.printHeader(f)
 885
 886                 f.write('<h1>Authors</h1>')
 887                 self.printNav(f)
 888
 889                 # Authors :: List of authors
 890                 f.write(html_header(2, 'List of Authors'))
 891
 892                 f.write('<table class="authors sortable" id="authors">')
 893                 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>')
 894                 for author in data.getAuthors(conf['max_authors']):
 895                         info = data.getAuthorInfo(author)
 896                         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']))
 897                 f.write('</table>')
 898
 899                 allauthors = data.getAuthors()
 900                 if len(allauthors) > conf['max_authors']:
 901                         rest = allauthors[conf['max_authors']:]
 902                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 903
 904                 f.write(html_header(2, 'Cumulated Added Lines of Code per Author'))
 905                 f.write('<img src="lines_of_code_by_author.png" alt="Lines of code per Author" />')
 906                 if len(allauthors) > conf['max_authors']:
 907                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 908
 909                 f.write(html_header(2, 'Commits per Author'))
 910                 f.write('<img src="commits_by_author.png" alt="Commits per Author" />')
 911                 if len(allauthors) > conf['max_authors']:
 912                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 913
 914                 fgl = open(path + '/lines_of_code_by_author.dat', 'w')
 915                 fgc = open(path + '/commits_by_author.dat', 'w')
 916
 917                 lines_by_authors = {} # cumulated added lines by
 918                 # author. to save memory,
 919                 # changes_by_date_by_author[stamp][author] is defined
 920                 # only at points where author commits.
 921                 # lines_by_authors allows us to generate all the
 922                 # points in the .dat file.
 923
 924                 # Don't rely on getAuthors to give the same order each
 925                 # time. Be robust and keep the list in a variable.
 926                 commits_by_authors = {} # cumulated added lines by
 927
 928                 self.authors_to_plot = data.getAuthors(conf['max_authors'])
 929                 for author in self.authors_to_plot:
 930                         lines_by_authors[author] = 0
 931                         commits_by_authors[author] = 0
 932                 for stamp in sorted(data.changes_by_date_by_author.keys()):
 933                         fgl.write('%d' % stamp)
 934                         fgc.write('%d' % stamp)
 935                         for author in self.authors_to_plot:
 936                                 if author in data.changes_by_date_by_author[stamp].keys():
 937                                         lines_by_authors[author] = data.changes_by_date_by_author[stamp][author]['lines_added']
 938                                         commits_by_authors[author] = data.changes_by_date_by_author[stamp][author]['commits']
 939                                 fgl.write(' %d' % lines_by_authors[author])
 940                                 fgc.write(' %d' % commits_by_authors[author])
 941                         fgl.write('\n')
 942                         fgc.write('\n')
 943                 fgl.close()
 944                 fgc.close()
 945
 946                 # Authors :: Author of Month
 947                 f.write(html_header(2, 'Author of Month'))
 948                 f.write('<table class="sortable" id="aom">')
 949                 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'])
 950                 for yymm in reversed(sorted(data.author_of_month.keys())):
 951                         authordict = data.author_of_month[yymm]
 952                         authors = getkeyssortedbyvalues(authordict)
 953                         authors.reverse()
 954                         commits = data.author_of_month[yymm][authors[0]]
 955                         next = ', '.join(authors[1:conf['authors_top']+1])
 956                         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)))
 957
 958                 f.write('</table>')
 959
 960                 f.write(html_header(2, 'Author of Year'))
 961                 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'])
 962                 for yy in reversed(sorted(data.author_of_year.keys())):
 963                         authordict = data.author_of_year[yy]
 964                         authors = getkeyssortedbyvalues(authordict)
 965                         authors.reverse()
 966                         commits = data.author_of_year[yy][authors[0]]
 967                         next = ', '.join(authors[1:conf['authors_top']+1])
 968                         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)))
 969                 f.write('</table>')
 970
 971                 # Domains
 972                 f.write(html_header(2, 'Commits by Domains'))
 973                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
 974                 domains_by_commits.reverse() # most first
 975                 f.write('<div class="vtable"><table>')
 976                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
 977                 fp = open(path + '/domains.dat', 'w')
 978                 n = 0
 979                 for domain in domains_by_commits:
 980                         if n == conf['max_domains']:
 981                                 break
 982                         commits = 0
 983                         n += 1
 984                         info = data.getDomainInfo(domain)
 985                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
 986                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
 987                 f.write('</table></div>')
 988                 f.write('<img src="domains.png" alt="Commits by Domains" />')
 989                 fp.close()
 990
 991                 f.write('</body></html>')
 992                 f.close()
 993
 994                 ###
 995                 # Files
 996                 f = open(path + '/files.html', 'w')
 997                 self.printHeader(f)
 998                 f.write('<h1>Files</h1>')
 999                 self.printNav(f)
1000
1001                 f.write('<dl>\n')
1002                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
1003                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1004                 f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % (float(data.getTotalSize()) / data.getTotalFiles()))
1005                 f.write('</dl>\n')
1006
1007                 # Files :: File count by date
1008                 f.write(html_header(2, 'File count by date'))
1009
1010                 # use set to get rid of duplicate/unnecessary entries
1011                 files_by_date = set()
1012                 for stamp in sorted(data.files_by_stamp.keys()):
1013                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1014
1015                 fg = open(path + '/files_by_date.dat', 'w')
1016                 for line in sorted(list(files_by_date)):
1017                         fg.write('%s\n' % line)
1018                 #for stamp in sorted(data.files_by_stamp.keys()):
1019                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1020                 fg.close()
1021
1022                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
1023
1024                 #f.write('<h2>Average file size by date</h2>')
1025
1026                 # Files :: Extensions
1027                 f.write(html_header(2, 'Extensions'))
1028                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
1029                 for ext in sorted(data.extensions.keys()):
1030                         files = data.extensions[ext]['files']
1031                         lines = data.extensions[ext]['lines']
1032                         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))
1033                 f.write('</table>')
1034
1035                 f.write('</body></html>')
1036                 f.close()
1037
1038                 ###
1039                 # Lines
1040                 f = open(path + '/lines.html', 'w')
1041                 self.printHeader(f)
1042                 f.write('<h1>Lines</h1>')
1043                 self.printNav(f)
1044
1045                 f.write('<dl>\n')
1046                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1047                 f.write('</dl>\n')
1048
1049                 f.write(html_header(2, 'Lines of Code'))
1050                 f.write('<img src="lines_of_code.png" />')
1051
1052                 fg = open(path + '/lines_of_code.dat', 'w')
1053                 for stamp in sorted(data.changes_by_date.keys()):
1054                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
1055                 fg.close()
1056
1057                 f.write('</body></html>')
1058                 f.close()
1059
1060                 ###
1061                 # tags.html
1062                 f = open(path + '/tags.html', 'w')
1063                 self.printHeader(f)
1064                 f.write('<h1>Tags</h1>')
1065                 self.printNav(f)
1066
1067                 f.write('<dl>')
1068                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
1069                 if len(data.tags) > 0:
1070                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
1071                 f.write('</dl>')
1072
1073                 f.write('<table class="tags">')
1074                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
1075                 # sort the tags by date desc
1076                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
1077                 for tag in tags_sorted_by_date_desc:
1078                         authorinfo = []
1079                         self.authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
1080                         for i in reversed(self.authors_by_commits):
1081                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
1082                         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)))
1083                 f.write('</table>')
1084
1085                 f.write('</body></html>')
1086                 f.close()
1087
1088                 self.createGraphs(path)
1089
1090         def createGraphs(self, path):
1091                 print 'Generating graphs...'
1092
1093                 # hour of day
1094                 f = open(path + '/hour_of_day.plot', 'w')
1095                 f.write(GNUPLOT_COMMON)
1096                 f.write(
1097 """
1098 set output 'hour_of_day.png'
1099 unset key
1100 set xrange [0.5:24.5]
1101 set xtics 4
1102 set grid y
1103 set ylabel "Commits"
1104 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
1105 """)
1106                 f.close()
1107
1108                 # day of week
1109                 f = open(path + '/day_of_week.plot', 'w')
1110                 f.write(GNUPLOT_COMMON)
1111                 f.write(
1112 """
1113 set output 'day_of_week.png'
1114 unset key
1115 set xrange [0.5:7.5]
1116 set xtics 1
1117 set grid y
1118 set ylabel "Commits"
1119 plot 'day_of_week.dat' using 1:3:(0.5):xtic(2) w boxes fs solid
1120 """)
1121                 f.close()
1122
1123                 # Domains
1124                 f = open(path + '/domains.plot', 'w')
1125                 f.write(GNUPLOT_COMMON)
1126                 f.write(
1127 """
1128 set output 'domains.png'
1129 unset key
1130 unset xtics
1131 set yrange [0:]
1132 set grid y
1133 set ylabel "Commits"
1134 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
1135 """)
1136                 f.close()
1137
1138                 # Month of Year
1139                 f = open(path + '/month_of_year.plot', 'w')
1140                 f.write(GNUPLOT_COMMON)
1141                 f.write(
1142 """
1143 set output 'month_of_year.png'
1144 unset key
1145 set xrange [0.5:12.5]
1146 set xtics 1
1147 set grid y
1148 set ylabel "Commits"
1149 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
1150 """)
1151                 f.close()
1152
1153                 # commits_by_year_month
1154                 f = open(path + '/commits_by_year_month.plot', 'w')
1155                 f.write(GNUPLOT_COMMON)
1156                 f.write(
1157 """
1158 set output 'commits_by_year_month.png'
1159 unset key
1160 set xdata time
1161 set timefmt "%Y-%m"
1162 set format x "%Y-%m"
1163 set xtics rotate
1164 set bmargin 5
1165 set grid y
1166 set ylabel "Commits"
1167 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1168 """)
1169                 f.close()
1170
1171                 # commits_by_year
1172                 f = open(path + '/commits_by_year.plot', 'w')
1173                 f.write(GNUPLOT_COMMON)
1174                 f.write(
1175 """
1176 set output 'commits_by_year.png'
1177 unset key
1178 set xtics 1 rotate
1179 set grid y
1180 set ylabel "Commits"
1181 set yrange [0:]
1182 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1183 """)
1184                 f.close()
1185
1186                 # Files by date
1187                 f = open(path + '/files_by_date.plot', 'w')
1188                 f.write(GNUPLOT_COMMON)
1189                 f.write(
1190 """
1191 set output 'files_by_date.png'
1192 unset key
1193 set xdata time
1194 set timefmt "%Y-%m-%d"
1195 set format x "%Y-%m-%d"
1196 set grid y
1197 set ylabel "Files"
1198 set xtics rotate
1199 set ytics autofreq
1200 set bmargin 6
1201 plot 'files_by_date.dat' using 1:2 w steps
1202 """)
1203                 f.close()
1204
1205                 # Lines of Code
1206                 f = open(path + '/lines_of_code.plot', 'w')
1207                 f.write(GNUPLOT_COMMON)
1208                 f.write(
1209 """
1210 set output 'lines_of_code.png'
1211 unset key
1212 set xdata time
1213 set timefmt "%s"
1214 set format x "%Y-%m-%d"
1215 set grid y
1216 set ylabel "Lines"
1217 set xtics rotate
1218 set bmargin 6
1219 plot 'lines_of_code.dat' using 1:2 w lines
1220 """)
1221                 f.close()
1222
1223                 # Lines of Code Added per author
1224                 f = open(path + '/lines_of_code_by_author.plot', 'w')
1225                 f.write(GNUPLOT_COMMON)
1226                 f.write(
1227 """
1228 set terminal png transparent size 640,480
1229 set output 'lines_of_code_by_author.png'
1230 set key left top
1231 set xdata time
1232 set timefmt "%s"
1233 set format x "%Y-%m-%d"
1234 set grid y
1235 set ylabel "Lines"
1236 set xtics rotate
1237 set bmargin 6
1238 plot """
1239 )
1240                 i = 1
1241                 plots = []
1242                 for a in self.authors_to_plot:
1243                         i = i + 1
1244                         plots.append("""'lines_of_code_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1245                 f.write(", ".join(plots))
1246                 f.write('\n')
1247
1248                 f.close()
1249
1250                 # Commits per author
1251                 f = open(path + '/commits_by_author.plot', 'w')
1252                 f.write(GNUPLOT_COMMON)
1253                 f.write(
1254 """
1255 set terminal png transparent size 640,480
1256 set output 'commits_by_author.png'
1257 set key left top
1258 set xdata time
1259 set timefmt "%s"
1260 set format x "%Y-%m-%d"
1261 set grid y
1262 set ylabel "Commits"
1263 set xtics rotate
1264 set bmargin 6
1265 plot """
1266 )
1267                 i = 1
1268                 plots = []
1269                 for a in self.authors_to_plot:
1270                         i = i + 1
1271                         plots.append("""'commits_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1272                 f.write(", ".join(plots))
1273                 f.write('\n')
1274
1275                 f.close()
1276
1277                 os.chdir(path)
1278                 files = glob.glob(path + '/*.plot')
1279                 for f in files:
1280                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1281                         if len(out) > 0:
1282                                 print out
1283
1284         def printHeader(self, f, title = ''):
1285                 f.write(
1286 """<?xml version="1.0" encoding="UTF-8"?>
1287 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1288 <html xmlns="http://www.w3.org/1999/xhtml">
1289 <head>
1290         <title>GitStats - %s</title>
1291         <link rel="stylesheet" href="%s" type="text/css" />
1292         <meta name="generator" content="GitStats %s" />
1293         <script type="text/javascript" src="sortable.js"></script>
1294 </head>
1295 <body>
1296 """ % (self.title, conf['style'], getversion()))
1297
1298         def printNav(self, f):
1299                 f.write("""
1300 <div class="nav">
1301 <ul>
1302 <li><a href="index.html">General</a></li>
1303 <li><a href="activity.html">Activity</a></li>
1304 <li><a href="authors.html">Authors</a></li>
1305 <li><a href="files.html">Files</a></li>
1306 <li><a href="lines.html">Lines</a></li>
1307 <li><a href="tags.html">Tags</a></li>
1308 </ul>
1309 </div>
1310 """)
1311
1312
1313 class GitStats:
1314         def run(self, args_orig):
1315                 optlist, args = getopt.getopt(args_orig, 'c:')
1316                 for o,v in optlist:
1317                         if o == '-c':
1318                                 key, value = v.split('=', 1)
1319                                 if key not in conf:
1320                                         raise KeyError('no such key "%s" in config' % key)
1321                                 if isinstance(conf[key], int):
1322                                         conf[key] = int(value)
1323                                 else:
1324                                         conf[key] = value
1325
1326                 if len(args) < 2:
1327                         print """
1328 Usage: gitstats [options] <gitpath..> <outputpath>
1329
1330 Options:
1331 -c key=value     Override configuration value
1332
1333 Default config values:
1334 %s
1335 """ % conf
1336                         sys.exit(0)
1337
1338                 outputpath = os.path.abspath(args[-1])
1339                 rundir = os.getcwd()
1340
1341                 try:
1342                         os.makedirs(outputpath)
1343                 except OSError:
1344                         pass
1345                 if not os.path.isdir(outputpath):
1346                         print 'FATAL: Output path is not a directory or does not exist'
1347                         sys.exit(1)
1348
1349                 print 'Output path: %s' % outputpath
1350                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1351
1352                 data = GitDataCollector()
1353                 data.loadCache(cachefile)
1354
1355                 for gitpath in args[0:-1]:
1356                         print 'Git path: %s' % gitpath
1357
1358                         os.chdir(gitpath)
1359
1360                         print 'Collecting data...'
1361                         data.collect(gitpath)
1362
1363                 print 'Refining data...'
1364                 data.saveCache(cachefile)
1365                 data.refine()
1366
1367                 os.chdir(rundir)
1368
1369                 print 'Generating report...'
1370                 report = HTMLReportCreator()
1371                 report.create(data, outputpath)
1372
1373                 time_end = time.time()
1374                 exectime_internal = time_end - time_start
1375                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1376
1377 if __name__=='__main__':
1378         g = GitStats()
1379         g.run(sys.argv[1:])
1380