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