gitstats

   1 #!/usr/bin/env python
   2 # Copyright (c) 2007-2010 Heikki Hokkanen <hoxu@users.sf.net> & others (see doc/author.txt)
   3 # GPLv2 / GPLv3
   4 import datetime
   5 import getopt
   6 import glob
   7 import os
   8 import pickle
   9 import platform
  10 import re
  11 import shutil
  12 import subprocess
  13 import sys
  14 import time
  15 import zlib
  16
  17 GNUPLOT_COMMON = 'set terminal png transparent 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': '',
  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
  96         ##
  97         # This should be the main function to extract data from the repository.
  98         def collect(self, dir):
  99                 self.dir = dir
 100                 if len(conf['project_name']) == 0:
 101                         self.projectname = os.path.basename(os.path.abspath(dir))
 102                 else:
 103                         self.projectname = conf['project_name']
 104
 105         ##
 106         # Load cacheable data
 107         def loadCache(self, cachefile):
 108                 if not os.path.exists(cachefile):
 109                         return
 110                 print 'Loading cache...'
 111                 f = open(cachefile, 'rb')
 112                 try:
 113                         self.cache = pickle.loads(zlib.decompress(f.read()))
 114                 except:
 115                         # temporary hack to upgrade non-compressed caches
 116                         f.seek(0)
 117                         self.cache = pickle.load(f)
 118                 f.close()
 119
 120         ##
 121         # Produce any additional statistics from the extracted data.
 122         def refine(self):
 123                 pass
 124
 125         ##
 126         # : get a dictionary of author
 127         def getAuthorInfo(self, author):
 128                 return None
 129
 130         def getActivityByDayOfWeek(self):
 131                 return {}
 132
 133         def getActivityByHourOfDay(self):
 134                 return {}
 135
 136         # : get a dictionary of domains
 137         def getDomainInfo(self, domain):
 138                 return None
 139
 140         ##
 141         # Get a list of authors
 142         def getAuthors(self):
 143                 return []
 144
 145         def getFirstCommitDate(self):
 146                 return datetime.datetime.now()
 147
 148         def getLastCommitDate(self):
 149                 return datetime.datetime.now()
 150
 151         def getStampCreated(self):
 152                 return self.stamp_created
 153
 154         def getTags(self):
 155                 return []
 156
 157         def getTotalAuthors(self):
 158                 return -1
 159
 160         def getTotalCommits(self):
 161                 return -1
 162
 163         def getTotalFiles(self):
 164                 return -1
 165
 166         def getTotalLOC(self):
 167                 return -1
 168
 169         ##
 170         # Save cacheable data
 171         def saveCache(self, cachefile):
 172                 print 'Saving cache...'
 173                 f = open(cachefile, 'wb')
 174                 #pickle.dump(self.cache, f)
 175                 data = zlib.compress(pickle.dumps(self.cache))
 176                 f.write(data)
 177                 f.close()
 178
 179 class GitDataCollector(DataCollector):
 180         def collect(self, dir):
 181                 DataCollector.collect(self, dir)
 182
 183                 try:
 184                         self.total_authors = int(getpipeoutput(['git shortlog -s %s' % getcommitrange(), 'wc -l']))
 185                 except:
 186                         self.total_authors = 0
 187                 #self.total_lines = int(getoutput('git-ls-files -z |xargs -0 cat |wc -l'))
 188
 189                 self.activity_by_hour_of_day = {} # hour -> commits
 190                 self.activity_by_day_of_week = {} # day -> commits
 191                 self.activity_by_month_of_year = {} # month [1-12] -> commits
 192                 self.activity_by_hour_of_week = {} # weekday -> hour -> commits
 193                 self.activity_by_hour_of_day_busiest = 0
 194                 self.activity_by_hour_of_week_busiest = 0
 195                 self.activity_by_year_week = {} # yy_wNN -> commits
 196                 self.activity_by_year_week_peak = 0
 197
 198                 self.authors = {} # name -> {commits, first_commit_stamp, last_commit_stamp, last_active_day, active_days, lines_added, lines_removed}
 199
 200                 # domains
 201                 self.domains = {} # domain -> commits
 202
 203                 # author of the month
 204                 self.author_of_month = {} # month -> author -> commits
 205                 self.author_of_year = {} # year -> author -> commits
 206                 self.commits_by_month = {} # month -> commits
 207                 self.commits_by_year = {} # year -> commits
 208                 self.first_commit_stamp = 0
 209                 self.last_commit_stamp = 0
 210                 self.last_active_day = None
 211                 self.active_days = set()
 212
 213                 # lines
 214                 self.total_lines = 0
 215                 self.total_lines_added = 0
 216                 self.total_lines_removed = 0
 217
 218                 # timezone
 219                 self.commits_by_timezone = {} # timezone -> commits
 220
 221                 # tags
 222                 self.tags = {}
 223                 lines = getpipeoutput(['git show-ref --tags']).split('\n')
 224                 for line in lines:
 225                         if len(line) == 0:
 226                                 continue
 227                         (hash, tag) = line.split(' ')
 228
 229                         tag = tag.replace('refs/tags/', '')
 230                         output = getpipeoutput(['git log "%s" --pretty=format:"%%at %%aN" -n 1' % hash])
 231                         if len(output) > 0:
 232                                 parts = output.split(' ')
 233                                 stamp = 0
 234                                 try:
 235                                         stamp = int(parts[0])
 236                                 except ValueError:
 237                                         stamp = 0
 238                                 self.tags[tag] = { 'stamp': stamp, 'hash' : hash, 'date' : datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), 'commits': 0, 'authors': {} }
 239
 240                 # collect info on tags, starting from latest
 241                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), self.tags.items()))))
 242                 prev = None
 243                 for tag in reversed(tags_sorted_by_date_desc):
 244                         cmd = 'git shortlog -s "%s"' % tag
 245                         if prev != None:
 246                                 cmd += ' "^%s"' % prev
 247                         output = getpipeoutput([cmd])
 248                         if len(output) == 0:
 249                                 continue
 250                         prev = tag
 251                         for line in output.split('\n'):
 252                                 parts = re.split('\s+', line, 2)
 253                                 commits = int(parts[1])
 254                                 author = parts[2]
 255                                 self.tags[tag]['commits'] += commits
 256                                 self.tags[tag]['authors'][author] = commits
 257
 258                 # Collect revision statistics
 259                 # Outputs "<stamp> <date> <time> <timezone> <author> '<' <mail> '>'"
 260                 lines = getpipeoutput(['git rev-list --pretty=format:"%%at %%ai %%aN <%%aE>" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).split('\n')
 261                 for line in lines:
 262                         parts = line.split(' ', 4)
 263                         author = ''
 264                         try:
 265                                 stamp = int(parts[0])
 266                         except ValueError:
 267                                 stamp = 0
 268                         timezone = parts[3]
 269                         author, mail = parts[4].split('<', 1)
 270                         author = author.rstrip()
 271                         mail = mail.rstrip('>')
 272                         domain = '?'
 273                         if mail.find('@') != -1:
 274                                 domain = mail.rsplit('@', 1)[1]
 275                         date = datetime.datetime.fromtimestamp(float(stamp))
 276
 277                         # First and last commit stamp
 278                         if self.last_commit_stamp == 0:
 279                                 self.last_commit_stamp = stamp
 280                         self.first_commit_stamp = stamp
 281
 282                         # activity
 283                         # hour
 284                         hour = date.hour
 285                         self.activity_by_hour_of_day[hour] = self.activity_by_hour_of_day.get(hour, 0) + 1
 286                         # most active hour?
 287                         if self.activity_by_hour_of_day[hour] > self.activity_by_hour_of_day_busiest:
 288                                 self.activity_by_hour_of_day_busiest = self.activity_by_hour_of_day[hour]
 289
 290                         # day of week
 291                         day = date.weekday()
 292                         self.activity_by_day_of_week[day] = self.activity_by_day_of_week.get(day, 0) + 1
 293
 294                         # domain stats
 295                         if domain not in self.domains:
 296                                 self.domains[domain] = {}
 297                         # commits
 298                         self.domains[domain]['commits'] = self.domains[domain].get('commits', 0) + 1
 299
 300                         # hour of week
 301                         if day not in self.activity_by_hour_of_week:
 302                                 self.activity_by_hour_of_week[day] = {}
 303                         self.activity_by_hour_of_week[day][hour] = self.activity_by_hour_of_week[day].get(hour, 0) + 1
 304                         # most active hour?
 305                         if self.activity_by_hour_of_week[day][hour] > self.activity_by_hour_of_week_busiest:
 306                                 self.activity_by_hour_of_week_busiest = self.activity_by_hour_of_week[day][hour]
 307
 308                         # month of year
 309                         month = date.month
 310                         self.activity_by_month_of_year[month] = self.activity_by_month_of_year.get(month, 0) + 1
 311
 312                         # yearly/weekly activity
 313                         yyw = date.strftime('%Y-%W')
 314                         self.activity_by_year_week[yyw] = self.activity_by_year_week.get(yyw, 0) + 1
 315                         if self.activity_by_year_week_peak < self.activity_by_year_week[yyw]:
 316                                 self.activity_by_year_week_peak = self.activity_by_year_week[yyw]
 317
 318                         # author stats
 319                         if author not in self.authors:
 320                                 self.authors[author] = {}
 321                         # commits
 322                         if 'last_commit_stamp' not in self.authors[author]:
 323                                 self.authors[author]['last_commit_stamp'] = stamp
 324                         self.authors[author]['first_commit_stamp'] = stamp
 325                         self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 326
 327                         # author of the month/year
 328                         yymm = date.strftime('%Y-%m')
 329                         if yymm in self.author_of_month:
 330                                 self.author_of_month[yymm][author] = self.author_of_month[yymm].get(author, 0) + 1
 331                         else:
 332                                 self.author_of_month[yymm] = {}
 333                                 self.author_of_month[yymm][author] = 1
 334                         self.commits_by_month[yymm] = self.commits_by_month.get(yymm, 0) + 1
 335
 336                         yy = date.year
 337                         if yy in self.author_of_year:
 338                                 self.author_of_year[yy][author] = self.author_of_year[yy].get(author, 0) + 1
 339                         else:
 340                                 self.author_of_year[yy] = {}
 341                                 self.author_of_year[yy][author] = 1
 342                         self.commits_by_year[yy] = self.commits_by_year.get(yy, 0) + 1
 343
 344                         # authors: active days
 345                         yymmdd = date.strftime('%Y-%m-%d')
 346                         if 'last_active_day' not in self.authors[author]:
 347                                 self.authors[author]['last_active_day'] = yymmdd
 348                                 self.authors[author]['active_days'] = 1
 349                         elif yymmdd != self.authors[author]['last_active_day']:
 350                                 self.authors[author]['last_active_day'] = yymmdd
 351                                 self.authors[author]['active_days'] += 1
 352
 353                         # project: active days
 354                         if yymmdd != self.last_active_day:
 355                                 self.last_active_day = yymmdd
 356                                 self.active_days.add(yymmdd)
 357
 358                         # timezone
 359                         self.commits_by_timezone[timezone] = self.commits_by_timezone.get(timezone, 0) + 1
 360
 361                 # TODO Optimize this, it's the worst bottleneck
 362                 # outputs "<stamp> <files>" for each revision
 363                 self.files_by_stamp = {} # stamp -> files
 364                 revlines = getpipeoutput(['git rev-list --pretty=format:"%%at %%T" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).strip().split('\n')
 365                 lines = []
 366                 for revline in revlines:
 367                         time, rev = revline.split(' ')
 368                         linecount = self.getFilesInCommit(rev)
 369                         lines.append('%d %d' % (int(time), linecount))
 370
 371                 self.total_commits = len(lines)
 372                 for line in lines:
 373                         parts = line.split(' ')
 374                         if len(parts) != 2:
 375                                 continue
 376                         (stamp, files) = parts[0:2]
 377                         try:
 378                                 self.files_by_stamp[int(stamp)] = int(files)
 379                         except ValueError:
 380                                 print 'Warning: failed to parse line "%s"' % line
 381
 382                 # extensions
 383                 self.extensions = {} # extension -> files, lines
 384                 lines = getpipeoutput(['git ls-tree -r -z %s' % getcommitrange('HEAD', end_only = True)]).split('\000')
 385                 self.total_files = len(lines)
 386                 for line in lines:
 387                         if len(line) == 0:
 388                                 continue
 389                         parts = re.split('\s+', line, 4)
 390                         sha1 = parts[2]
 391                         filename = parts[3]
 392
 393                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 394                                 ext = ''
 395                         else:
 396                                 ext = filename[(filename.rfind('.') + 1):]
 397                         if len(ext) > conf['max_ext_length']:
 398                                 ext = ''
 399
 400                         if ext not in self.extensions:
 401                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 402
 403                         self.extensions[ext]['files'] += 1
 404                         try:
 405                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 406                         except:
 407                                 print 'Warning: Could not count lines for file "%s"' % line
 408
 409                 # line statistics
 410                 # outputs:
 411                 #  N files changed, N insertions (+), N deletions(-)
 412                 # <stamp> <author>
 413                 self.changes_by_date = {} # stamp -> { files, ins, del }
 414                 # computation of lines of code by date is better done
 415                 # on a linear history.
 416                 extra = ''
 417                 if conf['linear_linestats']:
 418                         extra = '--first-parent -m'
 419                 lines = getpipeoutput(['git log --shortstat %s --pretty=format:"%%at %%aN" %s' % (extra, getcommitrange('HEAD'))]).split('\n')
 420                 lines.reverse()
 421                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 422                 author = None
 423                 for line in lines:
 424                         if len(line) == 0:
 425                                 continue
 426
 427                         # <stamp> <author>
 428                         if line.find('files changed,') == -1:
 429                                 pos = line.find(' ')
 430                                 if pos != -1:
 431                                         try:
 432                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 433                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 434                                                 files, inserted, deleted = 0, 0, 0
 435                                         except ValueError:
 436                                                 print 'Warning: unexpected line "%s"' % line
 437                                 else:
 438                                         print 'Warning: unexpected line "%s"' % line
 439                         else:
 440                                 numbers = re.findall('\d+', line)
 441                                 if len(numbers) == 3:
 442                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 443                                         total_lines += inserted
 444                                         total_lines -= deleted
 445                                         self.total_lines_added += inserted
 446                                         self.total_lines_removed += deleted
 447                                 else:
 448                                         print 'Warning: failed to handle line "%s"' % line
 449                                         (files, inserted, deleted) = (0, 0, 0)
 450                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 451                 self.total_lines = total_lines
 452
 453                 # Per-author statistics
 454                 self.changes_by_date_by_author = {} # stamp -> author -> lines_added
 455                                                     # defined for stamp, author only if
 456                                                     # author commited at this timestamp.
 457                 # Similar to the above, but never use --first-parent
 458                 # (we need to walk through every commits to know who
 459                 # commited what, not just through mainline)
 460                 lines = getpipeoutput(['git log --shortstat --pretty=format:"%%at %%aN" %s' % (getcommitrange('HEAD'))]).split('\n')
 461                 lines.reverse()
 462                 files = 0; inserted = 0; deleted = 0;
 463                 author = None
 464                 for line in lines:
 465                         if len(line) == 0:
 466                                 continue
 467
 468                         # <stamp> <author>
 469                         if line.find('files changed,') == -1:
 470                                 pos = line.find(' ')
 471                                 if pos != -1:
 472                                         try:
 473                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 474                                                 if author not in self.authors:
 475                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0 }
 476                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 477                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 478                                                 if stamp not in self.changes_by_date_by_author:
 479                                                         self.changes_by_date_by_author[stamp] = {}
 480                                                 self.changes_by_date_by_author[stamp][author] = self.authors[author]['lines_added']
 481                                                 files, inserted, deleted = 0, 0, 0
 482                                         except ValueError:
 483                                                 print 'Warning: unexpected line "%s"' % line
 484                                 else:
 485                                         print 'Warning: unexpected line "%s"' % line
 486                         else:
 487                                 numbers = re.findall('\d+', line)
 488                                 if len(numbers) == 3:
 489                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 490                                 else:
 491                                         print 'Warning: failed to handle line "%s"' % line
 492                                         (files, inserted, deleted) = (0, 0, 0)
 493
 494         def refine(self):
 495                 # authors
 496                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 497                 authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 498                 authors_by_commits.reverse() # most first
 499                 for i, name in enumerate(authors_by_commits):
 500                         self.authors[name]['place_by_commits'] = i + 1
 501
 502                 for name in self.authors.keys():
 503                         a = self.authors[name]
 504                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 505                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 506                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 507                         delta = date_last - date_first
 508                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 509                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 510                         a['timedelta'] = delta
 511                         if 'lines_added' not in a: a['lines_added'] = 0
 512                         if 'lines_removed' not in a: a['lines_removed'] = 0
 513
 514         def getActiveDays(self):
 515                 return self.active_days
 516
 517         def getActivityByDayOfWeek(self):
 518                 return self.activity_by_day_of_week
 519
 520         def getActivityByHourOfDay(self):
 521                 return self.activity_by_hour_of_day
 522
 523         def getAuthorInfo(self, author):
 524                 return self.authors[author]
 525
 526         def getAuthors(self, limit = None):
 527                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 528                 res.reverse()
 529                 return res[:limit]
 530
 531         def getCommitDeltaDays(self):
 532                 return (self.last_commit_stamp - self.first_commit_stamp) / 86400 + 1
 533
 534         def getDomainInfo(self, domain):
 535                 return self.domains[domain]
 536
 537         def getDomains(self):
 538                 return self.domains.keys()
 539
 540         def getFilesInCommit(self, rev):
 541                 try:
 542                         res = self.cache['files_in_tree'][rev]
 543                 except:
 544                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 545                         if 'files_in_tree' not in self.cache:
 546                                 self.cache['files_in_tree'] = {}
 547                         self.cache['files_in_tree'][rev] = res
 548
 549                 return res
 550
 551         def getFirstCommitDate(self):
 552                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 553
 554         def getLastCommitDate(self):
 555                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 556
 557         def getLinesInBlob(self, sha1):
 558                 try:
 559                         res = self.cache['lines_in_blob'][sha1]
 560                 except:
 561                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 562                         if 'lines_in_blob' not in self.cache:
 563                                 self.cache['lines_in_blob'] = {}
 564                         self.cache['lines_in_blob'][sha1] = res
 565                 return res
 566
 567         def getTags(self):
 568                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 569                 return lines.split('\n')
 570
 571         def getTagDate(self, tag):
 572                 return self.revToDate('tags/' + tag)
 573
 574         def getTotalAuthors(self):
 575                 return self.total_authors
 576
 577         def getTotalCommits(self):
 578                 return self.total_commits
 579
 580         def getTotalFiles(self):
 581                 return self.total_files
 582
 583         def getTotalLOC(self):
 584                 return self.total_lines
 585
 586         def revToDate(self, rev):
 587                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 588                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 589
 590 class ReportCreator:
 591         """Creates the actual report based on given data."""
 592         def __init__(self):
 593                 pass
 594
 595         def create(self, data, path):
 596                 self.data = data
 597                 self.path = path
 598
 599 def html_linkify(text):
 600         return text.lower().replace(' ', '_')
 601
 602 def html_header(level, text):
 603         name = html_linkify(text)
 604         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 605
 606 class HTMLReportCreator(ReportCreator):
 607         def create(self, data, path):
 608                 ReportCreator.create(self, data, path)
 609                 self.title = data.projectname
 610
 611                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 612                 binarypath = os.path.dirname(os.path.abspath(__file__))
 613                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 614                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 615                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 616                         for base in basedirs:
 617                                 src = base + '/' + file
 618                                 if os.path.exists(src):
 619                                         shutil.copyfile(src, path + '/' + file)
 620                                         break
 621                         else:
 622                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 623
 624                 f = open(path + "/index.html", 'w')
 625                 format = '%Y-%m-%d %H:%M:%S'
 626                 self.printHeader(f)
 627
 628                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 629
 630                 self.printNav(f)
 631
 632                 f.write('<dl>')
 633                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 634                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 635                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s), %s, %s</dd>' % (getversion(), getgitversion(), getgnuplotversion()))
 636                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 637                 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())))
 638                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 639                 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))
 640                 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()))
 641                 f.write('<dt>Authors</dt><dd>%s (average %.1f commits per author)</dd>' % (data.getTotalAuthors(), (1.0 * data.getTotalCommits()) / data.getTotalAuthors()))
 642                 f.write('</dl>')
 643
 644                 f.write('</body>\n</html>')
 645                 f.close()
 646
 647                 ###
 648                 # Activity
 649                 f = open(path + '/activity.html', 'w')
 650                 self.printHeader(f)
 651                 f.write('<h1>Activity</h1>')
 652                 self.printNav(f)
 653
 654                 #f.write('<h2>Last 30 days</h2>')
 655
 656                 #f.write('<h2>Last 12 months</h2>')
 657
 658                 # Weekly activity
 659                 WEEKS = 32
 660                 f.write(html_header(2, 'Weekly activity'))
 661                 f.write('<p>Last %d weeks</p>' % WEEKS)
 662
 663                 # generate weeks to show (previous N weeks from now)
 664                 now = datetime.datetime.now()
 665                 deltaweek = datetime.timedelta(7)
 666                 weeks = []
 667                 stampcur = now
 668                 for i in range(0, WEEKS):
 669                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 670                         stampcur -= deltaweek
 671
 672                 # top row: commits & bar
 673                 f.write('<table class="noborders"><tr>')
 674                 for i in range(0, WEEKS):
 675                         commits = 0
 676                         if weeks[i] in data.activity_by_year_week:
 677                                 commits = data.activity_by_year_week[weeks[i]]
 678
 679                         percentage = 0
 680                         if weeks[i] in data.activity_by_year_week:
 681                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 682                         height = max(1, int(200 * percentage))
 683                         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))
 684
 685                 # bottom row: year/week
 686                 f.write('</tr><tr>')
 687                 for i in range(0, WEEKS):
 688                         f.write('<td>%s</td>' % (WEEKS - i))
 689                 f.write('</tr></table>')
 690
 691                 # Hour of Day
 692                 f.write(html_header(2, 'Hour of Day'))
 693                 hour_of_day = data.getActivityByHourOfDay()
 694                 f.write('<table><tr><th>Hour</th>')
 695                 for i in range(0, 24):
 696                         f.write('<th>%d</th>' % i)
 697                 f.write('</tr>\n<tr><th>Commits</th>')
 698                 fp = open(path + '/hour_of_day.dat', 'w')
 699                 for i in range(0, 24):
 700                         if i in hour_of_day:
 701                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 702                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 703                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 704                         else:
 705                                 f.write('<td>0</td>')
 706                                 fp.write('%d 0\n' % i)
 707                 fp.close()
 708                 f.write('</tr>\n<tr><th>%</th>')
 709                 totalcommits = data.getTotalCommits()
 710                 for i in range(0, 24):
 711                         if i in hour_of_day:
 712                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 713                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 714                         else:
 715                                 f.write('<td>0.00</td>')
 716                 f.write('</tr></table>')
 717                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 718                 fg = open(path + '/hour_of_day.dat', 'w')
 719                 for i in range(0, 24):
 720                         if i in hour_of_day:
 721                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 722                         else:
 723                                 fg.write('%d 0\n' % (i + 1))
 724                 fg.close()
 725
 726                 # Day of Week
 727                 f.write(html_header(2, 'Day of Week'))
 728                 day_of_week = data.getActivityByDayOfWeek()
 729                 f.write('<div class="vtable"><table>')
 730                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 731                 fp = open(path + '/day_of_week.dat', 'w')
 732                 for d in range(0, 7):
 733                         commits = 0
 734                         if d in day_of_week:
 735                                 commits = day_of_week[d]
 736                         fp.write('%d %s %d\n' % (d + 1, WEEKDAYS[d], commits))
 737                         f.write('<tr>')
 738                         f.write('<th>%s</th>' % (WEEKDAYS[d]))
 739                         if d in day_of_week:
 740                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 741                         else:
 742                                 f.write('<td>0</td>')
 743                         f.write('</tr>')
 744                 f.write('</table></div>')
 745                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 746                 fp.close()
 747
 748                 # Hour of Week
 749                 f.write(html_header(2, 'Hour of Week'))
 750                 f.write('<table>')
 751
 752                 f.write('<tr><th>Weekday</th>')
 753                 for hour in range(0, 24):
 754                         f.write('<th>%d</th>' % (hour))
 755                 f.write('</tr>')
 756
 757                 for weekday in range(0, 7):
 758                         f.write('<tr><th>%s</th>' % (WEEKDAYS[weekday]))
 759                         for hour in range(0, 24):
 760                                 try:
 761                                         commits = data.activity_by_hour_of_week[weekday][hour]
 762                                 except KeyError:
 763                                         commits = 0
 764                                 if commits != 0:
 765                                         f.write('<td')
 766                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 767                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 768                                         f.write('>%d</td>' % commits)
 769                                 else:
 770                                         f.write('<td></td>')
 771                         f.write('</tr>')
 772
 773                 f.write('</table>')
 774
 775                 # Month of Year
 776                 f.write(html_header(2, 'Month of Year'))
 777                 f.write('<div class="vtable"><table>')
 778                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 779                 fp = open (path + '/month_of_year.dat', 'w')
 780                 for mm in range(1, 13):
 781                         commits = 0
 782                         if mm in data.activity_by_month_of_year:
 783                                 commits = data.activity_by_month_of_year[mm]
 784                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 785                         fp.write('%d %d\n' % (mm, commits))
 786                 fp.close()
 787                 f.write('</table></div>')
 788                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 789
 790                 # Commits by year/month
 791                 f.write(html_header(2, 'Commits by year/month'))
 792                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th></tr>')
 793                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 794                         f.write('<tr><td>%s</td><td>%d</td></tr>' % (yymm, data.commits_by_month[yymm]))
 795                 f.write('</table></div>')
 796                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 797                 fg = open(path + '/commits_by_year_month.dat', 'w')
 798                 for yymm in sorted(data.commits_by_month.keys()):
 799                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 800                 fg.close()
 801
 802                 # Commits by year
 803                 f.write(html_header(2, 'Commits by Year'))
 804                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th></tr>')
 805                 for yy in reversed(sorted(data.commits_by_year.keys())):
 806                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td></tr>' % (yy, data.commits_by_year[yy], (100.0 * data.commits_by_year[yy]) / data.getTotalCommits()))
 807                 f.write('</table></div>')
 808                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 809                 fg = open(path + '/commits_by_year.dat', 'w')
 810                 for yy in sorted(data.commits_by_year.keys()):
 811                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 812                 fg.close()
 813
 814                 # Commits by timezone
 815                 f.write(html_header(2, 'Commits by Timezone'))
 816                 f.write('<table><tr>')
 817                 f.write('<th>Timezone</th><th>Commits</th>')
 818                 max_commits_on_tz = max(data.commits_by_timezone.values())
 819                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 820                         commits = data.commits_by_timezone[i]
 821                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 822                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 823                 f.write('</tr></table>')
 824
 825                 f.write('</body></html>')
 826                 f.close()
 827
 828                 ###
 829                 # Authors
 830                 f = open(path + '/authors.html', 'w')
 831                 self.printHeader(f)
 832
 833                 f.write('<h1>Authors</h1>')
 834                 self.printNav(f)
 835
 836                 # Authors :: List of authors
 837                 f.write(html_header(2, 'List of Authors'))
 838
 839                 f.write('<table class="authors sortable" id="authors">')
 840                 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>')
 841                 for author in data.getAuthors(conf['max_authors']):
 842                         info = data.getAuthorInfo(author)
 843                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d</td><td>%d</td><td>%s</td><td>%s</td><td>%s</td><td>%d</td><td>%d</td></tr>' % (author, info['commits'], info['commits_frac'], info['lines_added'], info['lines_removed'], info['date_first'], info['date_last'], info['timedelta'], info['active_days'], info['place_by_commits']))
 844                 f.write('</table>')
 845
 846                 allauthors = data.getAuthors()
 847                 if len(allauthors) > conf['max_authors']:
 848                         rest = allauthors[conf['max_authors']:]
 849                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 850
 851                 f.write(html_header(2, 'Cumulated Added Lines of Code per Author'))
 852                 f.write('<img src="lines_of_code_by_author.png" alt="Lines of code per Author" />')
 853                 if len(allauthors) > conf['max_authors']:
 854                         rest = allauthors[conf['max_authors']:]
 855                         f.write('<p class="moreauthors">These didn\'t make it to the graph: %s</p>' % ', '.join(rest))
 856
 857                 fg = open(path + '/lines_of_code_by_author.dat', 'w')
 858                 lines_by_authors = {} # cumulated added lines by
 859                 # author. to save memory,
 860                 # changes_by_date_by_author[stamp][author] is defined
 861                 # only at points where author commits.
 862                 # lines_by_authors allows us to generate all the
 863                 # points in the .dat file.
 864
 865                 # Don't rely on getAuthors to give the same order each
 866                 # time. Be robust and keep the list in a variable.
 867                 self.authors_to_plot = data.getAuthors(conf['max_authors'])
 868                 for author in self.authors_to_plot:
 869                         lines_by_authors[author] = 0
 870                 for stamp in sorted(data.changes_by_date_by_author.keys()):
 871                         fg.write('%d' % stamp)
 872                         for author in self.authors_to_plot:
 873                                 if author in data.changes_by_date_by_author[stamp].keys():
 874                                         lines_by_authors[author] = data.changes_by_date_by_author[stamp][author]
 875                                 fg.write(' %d' % lines_by_authors[author])
 876                         fg.write('\n');
 877                 fg.close()
 878
 879                 # Authors :: Author of Month
 880                 f.write(html_header(2, 'Author of Month'))
 881                 f.write('<table class="sortable" id="aom">')
 882                 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'])
 883                 for yymm in reversed(sorted(data.author_of_month.keys())):
 884                         authordict = data.author_of_month[yymm]
 885                         authors = getkeyssortedbyvalues(authordict)
 886                         authors.reverse()
 887                         commits = data.author_of_month[yymm][authors[0]]
 888                         next = ', '.join(authors[1:conf['authors_top']+1])
 889                         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)))
 890
 891                 f.write('</table>')
 892
 893                 f.write(html_header(2, 'Author of Year'))
 894                 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'])
 895                 for yy in reversed(sorted(data.author_of_year.keys())):
 896                         authordict = data.author_of_year[yy]
 897                         authors = getkeyssortedbyvalues(authordict)
 898                         authors.reverse()
 899                         commits = data.author_of_year[yy][authors[0]]
 900                         next = ', '.join(authors[1:conf['authors_top']+1])
 901                         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)))
 902                 f.write('</table>')
 903
 904                 # Domains
 905                 f.write(html_header(2, 'Commits by Domains'))
 906                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
 907                 domains_by_commits.reverse() # most first
 908                 f.write('<div class="vtable"><table>')
 909                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
 910                 fp = open(path + '/domains.dat', 'w')
 911                 n = 0
 912                 for domain in domains_by_commits:
 913                         if n == conf['max_domains']:
 914                                 break
 915                         commits = 0
 916                         n += 1
 917                         info = data.getDomainInfo(domain)
 918                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
 919                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
 920                 f.write('</table></div>')
 921                 f.write('<img src="domains.png" alt="Commits by Domains" />')
 922                 fp.close()
 923
 924                 f.write('</body></html>')
 925                 f.close()
 926
 927                 ###
 928                 # Files
 929                 f = open(path + '/files.html', 'w')
 930                 self.printHeader(f)
 931                 f.write('<h1>Files</h1>')
 932                 self.printNav(f)
 933
 934                 f.write('<dl>\n')
 935                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
 936                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
 937                 f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % ((100.0 * data.getTotalLOC()) / data.getTotalFiles()))
 938                 f.write('</dl>\n')
 939
 940                 # Files :: File count by date
 941                 f.write(html_header(2, 'File count by date'))
 942
 943                 # use set to get rid of duplicate/unnecessary entries
 944                 files_by_date = set()
 945                 for stamp in sorted(data.files_by_stamp.keys()):
 946                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
 947
 948                 fg = open(path + '/files_by_date.dat', 'w')
 949                 for line in sorted(list(files_by_date)):
 950                         fg.write('%s\n' % line)
 951                 #for stamp in sorted(data.files_by_stamp.keys()):
 952                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
 953                 fg.close()
 954
 955                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
 956
 957                 #f.write('<h2>Average file size by date</h2>')
 958
 959                 # Files :: Extensions
 960                 f.write(html_header(2, 'Extensions'))
 961                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
 962                 for ext in sorted(data.extensions.keys()):
 963                         files = data.extensions[ext]['files']
 964                         lines = data.extensions[ext]['lines']
 965                         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))
 966                 f.write('</table>')
 967
 968                 f.write('</body></html>')
 969                 f.close()
 970
 971                 ###
 972                 # Lines
 973                 f = open(path + '/lines.html', 'w')
 974                 self.printHeader(f)
 975                 f.write('<h1>Lines</h1>')
 976                 self.printNav(f)
 977
 978                 f.write('<dl>\n')
 979                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
 980                 f.write('</dl>\n')
 981
 982                 f.write(html_header(2, 'Lines of Code'))
 983                 f.write('<img src="lines_of_code.png" />')
 984
 985                 fg = open(path + '/lines_of_code.dat', 'w')
 986                 for stamp in sorted(data.changes_by_date.keys()):
 987                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
 988                 fg.close()
 989
 990                 f.write('</body></html>')
 991                 f.close()
 992
 993                 ###
 994                 # tags.html
 995                 f = open(path + '/tags.html', 'w')
 996                 self.printHeader(f)
 997                 f.write('<h1>Tags</h1>')
 998                 self.printNav(f)
 999
1000                 f.write('<dl>')
1001                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
1002                 if len(data.tags) > 0:
1003                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
1004                 f.write('</dl>')
1005
1006                 f.write('<table class="tags">')
1007                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
1008                 # sort the tags by date desc
1009                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
1010                 for tag in tags_sorted_by_date_desc:
1011                         authorinfo = []
1012                         authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
1013                         for i in reversed(authors_by_commits):
1014                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
1015                         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)))
1016                 f.write('</table>')
1017
1018                 f.write('</body></html>')
1019                 f.close()
1020
1021                 self.createGraphs(path)
1022
1023         def createGraphs(self, path):
1024                 print 'Generating graphs...'
1025
1026                 # hour of day
1027                 f = open(path + '/hour_of_day.plot', 'w')
1028                 f.write(GNUPLOT_COMMON)
1029                 f.write(
1030 """
1031 set output 'hour_of_day.png'
1032 unset key
1033 set xrange [0.5:24.5]
1034 set xtics 4
1035 set grid y
1036 set ylabel "Commits"
1037 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
1038 """)
1039                 f.close()
1040
1041                 # day of week
1042                 f = open(path + '/day_of_week.plot', 'w')
1043                 f.write(GNUPLOT_COMMON)
1044                 f.write(
1045 """
1046 set output 'day_of_week.png'
1047 unset key
1048 set xrange [0.5:7.5]
1049 set xtics 1
1050 set grid y
1051 set ylabel "Commits"
1052 plot 'day_of_week.dat' using 1:3:(0.5):xtic(2) w boxes fs solid
1053 """)
1054                 f.close()
1055
1056                 # Domains
1057                 f = open(path + '/domains.plot', 'w')
1058                 f.write(GNUPLOT_COMMON)
1059                 f.write(
1060 """
1061 set output 'domains.png'
1062 unset key
1063 unset xtics
1064 set yrange [0:]
1065 set grid y
1066 set ylabel "Commits"
1067 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
1068 """)
1069                 f.close()
1070
1071                 # Month of Year
1072                 f = open(path + '/month_of_year.plot', 'w')
1073                 f.write(GNUPLOT_COMMON)
1074                 f.write(
1075 """
1076 set output 'month_of_year.png'
1077 unset key
1078 set xrange [0.5:12.5]
1079 set xtics 1
1080 set grid y
1081 set ylabel "Commits"
1082 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
1083 """)
1084                 f.close()
1085
1086                 # commits_by_year_month
1087                 f = open(path + '/commits_by_year_month.plot', 'w')
1088                 f.write(GNUPLOT_COMMON)
1089                 f.write(
1090 """
1091 set output 'commits_by_year_month.png'
1092 unset key
1093 set xdata time
1094 set timefmt "%Y-%m"
1095 set format x "%Y-%m"
1096 set xtics rotate
1097 set bmargin 5
1098 set grid y
1099 set ylabel "Commits"
1100 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1101 """)
1102                 f.close()
1103
1104                 # commits_by_year
1105                 f = open(path + '/commits_by_year.plot', 'w')
1106                 f.write(GNUPLOT_COMMON)
1107                 f.write(
1108 """
1109 set output 'commits_by_year.png'
1110 unset key
1111 set xtics 1 rotate
1112 set grid y
1113 set ylabel "Commits"
1114 set yrange [0:]
1115 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1116 """)
1117                 f.close()
1118
1119                 # Files by date
1120                 f = open(path + '/files_by_date.plot', 'w')
1121                 f.write(GNUPLOT_COMMON)
1122                 f.write(
1123 """
1124 set output 'files_by_date.png'
1125 unset key
1126 set xdata time
1127 set timefmt "%Y-%m-%d"
1128 set format x "%Y-%m-%d"
1129 set grid y
1130 set ylabel "Files"
1131 set xtics rotate
1132 set ytics autofreq
1133 set bmargin 6
1134 plot 'files_by_date.dat' using 1:2 w steps
1135 """)
1136                 f.close()
1137
1138                 # Lines of Code
1139                 f = open(path + '/lines_of_code.plot', 'w')
1140                 f.write(GNUPLOT_COMMON)
1141                 f.write(
1142 """
1143 set output 'lines_of_code.png'
1144 unset key
1145 set xdata time
1146 set timefmt "%s"
1147 set format x "%Y-%m-%d"
1148 set grid y
1149 set ylabel "Lines"
1150 set xtics rotate
1151 set bmargin 6
1152 plot 'lines_of_code.dat' using 1:2 w lines
1153 """)
1154                 f.close()
1155
1156                 # Lines of Code Added per author
1157                 f = open(path + '/lines_of_code_by_author.plot', 'w')
1158                 f.write(GNUPLOT_COMMON)
1159                 f.write(
1160 """
1161 set terminal png transparent size 640,480
1162 set output 'lines_of_code_by_author.png'
1163 set key left top
1164 set xdata time
1165 set timefmt "%s"
1166 set format x "%Y-%m-%d"
1167 set grid y
1168 set ylabel "Lines"
1169 set xtics rotate
1170 set bmargin 6
1171 plot """
1172 )
1173                 i = 1
1174                 plots = []
1175                 for a in self.authors_to_plot:
1176                         i = i + 1
1177                         plots.append("""'lines_of_code_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1178                 f.write(", ".join(plots))
1179                 f.write('\n')
1180
1181                 f.close()
1182
1183                 os.chdir(path)
1184                 files = glob.glob(path + '/*.plot')
1185                 for f in files:
1186                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1187                         if len(out) > 0:
1188                                 print out
1189
1190         def printHeader(self, f, title = ''):
1191                 f.write(
1192 """<?xml version="1.0" encoding="UTF-8"?>
1193 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1194 <html xmlns="http://www.w3.org/1999/xhtml">
1195 <head>
1196         <title>GitStats - %s</title>
1197         <link rel="stylesheet" href="%s" type="text/css" />
1198         <meta name="generator" content="GitStats %s" />
1199         <script type="text/javascript" src="sortable.js"></script>
1200 </head>
1201 <body>
1202 """ % (self.title, conf['style'], getversion()))
1203
1204         def printNav(self, f):
1205                 f.write("""
1206 <div class="nav">
1207 <ul>
1208 <li><a href="index.html">General</a></li>
1209 <li><a href="activity.html">Activity</a></li>
1210 <li><a href="authors.html">Authors</a></li>
1211 <li><a href="files.html">Files</a></li>
1212 <li><a href="lines.html">Lines</a></li>
1213 <li><a href="tags.html">Tags</a></li>
1214 </ul>
1215 </div>
1216 """)
1217
1218
1219 class GitStats:
1220         def run(self, args_orig):
1221                 optlist, args = getopt.getopt(args_orig, 'c:')
1222                 for o,v in optlist:
1223                         if o == '-c':
1224                                 key, value = v.split('=', 1)
1225                                 if key not in conf:
1226                                         raise KeyError('no such key "%s" in config' % key)
1227                                 if isinstance(conf[key], int):
1228                                         conf[key] = int(value)
1229                                 else:
1230                                         conf[key] = value
1231
1232                 if len(args) < 2:
1233                         print """
1234 Usage: gitstats [options] <gitpath> <outputpath>
1235
1236 Options:
1237 -c key=value     Override configuration value
1238
1239 Default config values:
1240 %s
1241 """ % conf
1242                         sys.exit(0)
1243
1244                 gitpath = args[0]
1245                 outputpath = os.path.abspath(args[1])
1246                 rundir = os.getcwd()
1247
1248                 try:
1249                         os.makedirs(outputpath)
1250                 except OSError:
1251                         pass
1252                 if not os.path.isdir(outputpath):
1253                         print 'FATAL: Output path is not a directory or does not exist'
1254                         sys.exit(1)
1255
1256                 print 'Git path: %s' % gitpath
1257                 print 'Output path: %s' % outputpath
1258
1259                 os.chdir(gitpath)
1260
1261                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1262
1263                 print 'Collecting data...'
1264                 data = GitDataCollector()
1265                 data.loadCache(cachefile)
1266                 data.collect(gitpath)
1267                 print 'Refining data...'
1268                 data.saveCache(cachefile)
1269                 data.refine()
1270
1271                 os.chdir(rundir)
1272
1273                 print 'Generating report...'
1274                 report = HTMLReportCreator()
1275                 report.create(data, outputpath)
1276
1277                 time_end = time.time()
1278                 exectime_internal = time_end - time_start
1279                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1280
1281 if __name__=='__main__':
1282         g = GitStats()
1283         g.run(sys.argv[1:])
1284