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