gitstats

   1 #!/usr/bin/env python
   2 # Copyright (c) 2007-2012 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 os.environ['LC_ALL'] = 'C'
  18
  19 GNUPLOT_COMMON = 'set terminal png transparent size 640,240\nset size 1.0,1.0\n'
  20 ON_LINUX = (platform.system() == 'Linux')
  21 WEEKDAYS = ('Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun')
  22
  23 exectime_internal = 0.0
  24 exectime_external = 0.0
  25 time_start = time.time()
  26
  27 # By default, gnuplot is searched from path, but can be overridden with the
  28 # environment variable "GNUPLOT"
  29 gnuplot_cmd = 'gnuplot'
  30 if 'GNUPLOT' in os.environ:
  31         gnuplot_cmd = os.environ['GNUPLOT']
  32
  33 conf = {
  34         'max_domains': 10,
  35         'max_ext_length': 10,
  36         'style': 'gitstats.css',
  37         'max_authors': 20,
  38         'authors_top': 5,
  39         'commit_begin': '',
  40         'commit_end': 'HEAD',
  41         'linear_linestats': 1,
  42         'project_name': '',
  43         'merge_authors': {}
  44 }
  45
  46 def getpipeoutput(cmds, quiet = False):
  47         global exectime_external
  48         start = time.time()
  49         if not quiet and ON_LINUX and os.isatty(1):
  50                 print '>> ' + ' | '.join(cmds),
  51                 sys.stdout.flush()
  52         p0 = subprocess.Popen(cmds[0], stdout = subprocess.PIPE, shell = True)
  53         p = p0
  54         for x in cmds[1:]:
  55                 p = subprocess.Popen(x, stdin = p0.stdout, stdout = subprocess.PIPE, shell = True)
  56                 p0 = p
  57         output = p.communicate()[0]
  58         end = time.time()
  59         if not quiet:
  60                 if ON_LINUX and os.isatty(1):
  61                         print '\r',
  62                 print '[%.5f] >> %s' % (end - start, ' | '.join(cmds))
  63         exectime_external += (end - start)
  64         return output.rstrip('\n')
  65
  66 def getcommitrange(defaultrange = 'HEAD', end_only = False):
  67         if len(conf['commit_end']) > 0:
  68                 if end_only or len(conf['commit_begin']) == 0:
  69                         return conf['commit_end']
  70                 return '%s..%s' % (conf['commit_begin'], conf['commit_end'])
  71         return defaultrange
  72
  73 def getkeyssortedbyvalues(dict):
  74         return map(lambda el : el[1], sorted(map(lambda el : (el[1], el[0]), dict.items())))
  75
  76 # dict['author'] = { 'commits': 512 } - ...key(dict, 'commits')
  77 def getkeyssortedbyvaluekey(d, key):
  78         return map(lambda el : el[1], sorted(map(lambda el : (d[el][key], el), d.keys())))
  79
  80 def getstatsummarycounts(line):
  81         numbers = re.findall('\d+', line)
  82         if   len(numbers) == 1:
  83                 # neither insertions nor deletions: may probably only happen for "0 files changed"
  84                 numbers.append(0);
  85                 numbers.append(0);
  86         elif len(numbers) == 2 and line.find('(+)') != -1:
  87                 numbers.append(0);    # only insertions were printed on line
  88         elif len(numbers) == 2 and line.find('(-)') != -1:
  89                 numbers.insert(1, 0); # only deletions were printed on line
  90         return numbers
  91
  92 VERSION = 0
  93 def getversion():
  94         global VERSION
  95         if VERSION == 0:
  96                 gitstats_repo = os.path.dirname(os.path.abspath(__file__))
  97                 VERSION = getpipeoutput(["git --git-dir=%s/.git --work-tree=%s rev-parse --short %s" %
  98                         (gitstats_repo, gitstats_repo, getcommitrange('HEAD').split('\n')[0])])
  99         return VERSION
 100
 101 def getgitversion():
 102         return getpipeoutput(['git --version']).split('\n')[0]
 103
 104 def getgnuplotversion():
 105         return getpipeoutput(['%s --version' % gnuplot_cmd]).split('\n')[0]
 106
 107 class DataCollector:
 108         """Manages data collection from a revision control repository."""
 109         def __init__(self):
 110                 self.stamp_created = time.time()
 111                 self.cache = {}
 112                 self.total_authors = 0
 113                 self.activity_by_hour_of_day = {} # hour -> commits
 114                 self.activity_by_day_of_week = {} # day -> commits
 115                 self.activity_by_month_of_year = {} # month [1-12] -> commits
 116                 self.activity_by_hour_of_week = {} # weekday -> hour -> commits
 117                 self.activity_by_hour_of_day_busiest = 0
 118                 self.activity_by_hour_of_week_busiest = 0
 119                 self.activity_by_year_week = {} # yy_wNN -> commits
 120                 self.activity_by_year_week_peak = 0
 121
 122                 self.authors = {} # name -> {commits, first_commit_stamp, last_commit_stamp, last_active_day, active_days, lines_added, lines_removed}
 123
 124                 self.total_commits = 0
 125                 self.total_files = 0
 126                 self.authors_by_commits = 0
 127
 128                 # domains
 129                 self.domains = {} # domain -> commits
 130
 131                 # author of the month
 132                 self.author_of_month = {} # month -> author -> commits
 133                 self.author_of_year = {} # year -> author -> commits
 134                 self.commits_by_month = {} # month -> commits
 135                 self.commits_by_year = {} # year -> commits
 136                 self.lines_added_by_month = {} # month -> lines added
 137                 self.lines_added_by_year = {} # year -> lines added
 138                 self.lines_removed_by_month = {} # month -> lines removed
 139                 self.lines_removed_by_year = {} # year -> lines removed
 140                 self.first_commit_stamp = 0
 141                 self.last_commit_stamp = 0
 142                 self.last_active_day = None
 143                 self.active_days = set()
 144
 145                 # lines
 146                 self.total_lines = 0
 147                 self.total_lines_added = 0
 148                 self.total_lines_removed = 0
 149
 150                 # size
 151                 self.total_size = 0
 152
 153                 # timezone
 154                 self.commits_by_timezone = {} # timezone -> commits
 155
 156                 # tags
 157                 self.tags = {}
 158
 159                 self.files_by_stamp = {} # stamp -> files
 160
 161                 # extensions
 162                 self.extensions = {} # extension -> files, lines
 163
 164                 # line statistics
 165                 self.changes_by_date = {} # stamp -> { files, ins, del }
 166
 167         ##
 168         # This should be the main function to extract data from the repository.
 169         def collect(self, dir):
 170                 self.dir = dir
 171                 if len(conf['project_name']) == 0:
 172                         self.projectname = os.path.basename(os.path.abspath(dir))
 173                 else:
 174                         self.projectname = conf['project_name']
 175
 176         ##
 177         # Load cacheable data
 178         def loadCache(self, cachefile):
 179                 if not os.path.exists(cachefile):
 180                         return
 181                 print 'Loading cache...'
 182                 f = open(cachefile, 'rb')
 183                 try:
 184                         self.cache = pickle.loads(zlib.decompress(f.read()))
 185                 except:
 186                         # temporary hack to upgrade non-compressed caches
 187                         f.seek(0)
 188                         self.cache = pickle.load(f)
 189                 f.close()
 190
 191         ##
 192         # Produce any additional statistics from the extracted data.
 193         def refine(self):
 194                 pass
 195
 196         ##
 197         # : get a dictionary of author
 198         def getAuthorInfo(self, author):
 199                 return None
 200
 201         def getActivityByDayOfWeek(self):
 202                 return {}
 203
 204         def getActivityByHourOfDay(self):
 205                 return {}
 206
 207         # : get a dictionary of domains
 208         def getDomainInfo(self, domain):
 209                 return None
 210
 211         ##
 212         # Get a list of authors
 213         def getAuthors(self):
 214                 return []
 215
 216         def getFirstCommitDate(self):
 217                 return datetime.datetime.now()
 218
 219         def getLastCommitDate(self):
 220                 return datetime.datetime.now()
 221
 222         def getStampCreated(self):
 223                 return self.stamp_created
 224
 225         def getTags(self):
 226                 return []
 227
 228         def getTotalAuthors(self):
 229                 return -1
 230
 231         def getTotalCommits(self):
 232                 return -1
 233
 234         def getTotalFiles(self):
 235                 return -1
 236
 237         def getTotalLOC(self):
 238                 return -1
 239
 240         ##
 241         # Save cacheable data
 242         def saveCache(self, cachefile):
 243                 print 'Saving cache...'
 244                 tempfile = cachefile + '.tmp'
 245                 f = open(tempfile, 'wb')
 246                 #pickle.dump(self.cache, f)
 247                 data = zlib.compress(pickle.dumps(self.cache))
 248                 f.write(data)
 249                 f.close()
 250                 try:
 251                         os.remove(cachefile)
 252                 except OSError:
 253                         pass
 254                 os.rename(tempfile, cachefile)
 255
 256 class GitDataCollector(DataCollector):
 257         def collect(self, dir):
 258                 DataCollector.collect(self, dir)
 259
 260                 self.total_authors += int(getpipeoutput(['git shortlog -s %s' % getcommitrange(), 'wc -l']))
 261                 #self.total_lines = int(getoutput('git-ls-files -z |xargs -0 cat |wc -l'))
 262
 263                 # tags
 264                 lines = getpipeoutput(['git show-ref --tags']).split('\n')
 265                 for line in lines:
 266                         if len(line) == 0:
 267                                 continue
 268                         (hash, tag) = line.split(' ')
 269
 270                         tag = tag.replace('refs/tags/', '')
 271                         output = getpipeoutput(['git log "%s" --pretty=format:"%%at %%aN" -n 1' % hash])
 272                         if len(output) > 0:
 273                                 parts = output.split(' ')
 274                                 stamp = 0
 275                                 try:
 276                                         stamp = int(parts[0])
 277                                 except ValueError:
 278                                         stamp = 0
 279                                 self.tags[tag] = { 'stamp': stamp, 'hash' : hash, 'date' : datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), 'commits': 0, 'authors': {} }
 280
 281                 # collect info on tags, starting from latest
 282                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), self.tags.items()))))
 283                 prev = None
 284                 for tag in reversed(tags_sorted_by_date_desc):
 285                         cmd = 'git shortlog -s "%s"' % tag
 286                         if prev != None:
 287                                 cmd += ' "^%s"' % prev
 288                         output = getpipeoutput([cmd])
 289                         if len(output) == 0:
 290                                 continue
 291                         prev = tag
 292                         for line in output.split('\n'):
 293                                 parts = re.split('\s+', line, 2)
 294                                 commits = int(parts[1])
 295                                 author = parts[2]
 296                                 if author in conf['merge_authors']:
 297                                         author = conf['merge_authors'][author]
 298                                 self.tags[tag]['commits'] += commits
 299                                 self.tags[tag]['authors'][author] = commits
 300
 301                 # Collect revision statistics
 302                 # Outputs "<stamp> <date> <time> <timezone> <author> '<' <mail> '>'"
 303                 lines = getpipeoutput(['git rev-list --pretty=format:"%%at %%ai %%aN <%%aE>" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).split('\n')
 304                 for line in lines:
 305                         parts = line.split(' ', 4)
 306                         author = ''
 307                         try:
 308                                 stamp = int(parts[0])
 309                         except ValueError:
 310                                 stamp = 0
 311                         timezone = parts[3]
 312                         author, mail = parts[4].split('<', 1)
 313                         author = author.rstrip()
 314                         if author in conf['merge_authors']:
 315                                 author = conf['merge_authors'][author]
 316                         mail = mail.rstrip('>')
 317                         domain = '?'
 318                         if mail.find('@') != -1:
 319                                 domain = mail.rsplit('@', 1)[1]
 320                         date = datetime.datetime.fromtimestamp(float(stamp))
 321
 322                         # First and last commit stamp (may be in any order because of cherry-picking and patches)
 323                         if stamp > self.last_commit_stamp:
 324                                 self.last_commit_stamp = stamp
 325                         if self.first_commit_stamp == 0 or stamp < self.first_commit_stamp:
 326                                 self.first_commit_stamp = stamp
 327
 328                         # activity
 329                         # hour
 330                         hour = date.hour
 331                         self.activity_by_hour_of_day[hour] = self.activity_by_hour_of_day.get(hour, 0) + 1
 332                         # most active hour?
 333                         if self.activity_by_hour_of_day[hour] > self.activity_by_hour_of_day_busiest:
 334                                 self.activity_by_hour_of_day_busiest = self.activity_by_hour_of_day[hour]
 335
 336                         # day of week
 337                         day = date.weekday()
 338                         self.activity_by_day_of_week[day] = self.activity_by_day_of_week.get(day, 0) + 1
 339
 340                         # domain stats
 341                         if domain not in self.domains:
 342                                 self.domains[domain] = {}
 343                         # commits
 344                         self.domains[domain]['commits'] = self.domains[domain].get('commits', 0) + 1
 345
 346                         # hour of week
 347                         if day not in self.activity_by_hour_of_week:
 348                                 self.activity_by_hour_of_week[day] = {}
 349                         self.activity_by_hour_of_week[day][hour] = self.activity_by_hour_of_week[day].get(hour, 0) + 1
 350                         # most active hour?
 351                         if self.activity_by_hour_of_week[day][hour] > self.activity_by_hour_of_week_busiest:
 352                                 self.activity_by_hour_of_week_busiest = self.activity_by_hour_of_week[day][hour]
 353
 354                         # month of year
 355                         month = date.month
 356                         self.activity_by_month_of_year[month] = self.activity_by_month_of_year.get(month, 0) + 1
 357
 358                         # yearly/weekly activity
 359                         yyw = date.strftime('%Y-%W')
 360                         self.activity_by_year_week[yyw] = self.activity_by_year_week.get(yyw, 0) + 1
 361                         if self.activity_by_year_week_peak < self.activity_by_year_week[yyw]:
 362                                 self.activity_by_year_week_peak = self.activity_by_year_week[yyw]
 363
 364                         # author stats
 365                         if author not in self.authors:
 366                                 self.authors[author] = {}
 367                         # commits, note again that commits may be in any date order because of cherry-picking and patches
 368                         if 'last_commit_stamp' not in self.authors[author]:
 369                                 self.authors[author]['last_commit_stamp'] = stamp
 370                         if stamp > self.authors[author]['last_commit_stamp']:
 371                                 self.authors[author]['last_commit_stamp'] = stamp
 372                         if 'first_commit_stamp' not in self.authors[author]:
 373                                 self.authors[author]['first_commit_stamp'] = stamp
 374                         if stamp < self.authors[author]['first_commit_stamp']:
 375                                 self.authors[author]['first_commit_stamp'] = stamp
 376
 377                         # author of the month/year
 378                         yymm = date.strftime('%Y-%m')
 379                         if yymm in self.author_of_month:
 380                                 self.author_of_month[yymm][author] = self.author_of_month[yymm].get(author, 0) + 1
 381                         else:
 382                                 self.author_of_month[yymm] = {}
 383                                 self.author_of_month[yymm][author] = 1
 384                         self.commits_by_month[yymm] = self.commits_by_month.get(yymm, 0) + 1
 385
 386                         yy = date.year
 387                         if yy in self.author_of_year:
 388                                 self.author_of_year[yy][author] = self.author_of_year[yy].get(author, 0) + 1
 389                         else:
 390                                 self.author_of_year[yy] = {}
 391                                 self.author_of_year[yy][author] = 1
 392                         self.commits_by_year[yy] = self.commits_by_year.get(yy, 0) + 1
 393
 394                         # authors: active days
 395                         yymmdd = date.strftime('%Y-%m-%d')
 396                         if 'last_active_day' not in self.authors[author]:
 397                                 self.authors[author]['last_active_day'] = yymmdd
 398                                 self.authors[author]['active_days'] = set([yymmdd])
 399                         elif yymmdd != self.authors[author]['last_active_day']:
 400                                 self.authors[author]['last_active_day'] = yymmdd
 401                                 self.authors[author]['active_days'].add(yymmdd)
 402
 403                         # project: active days
 404                         if yymmdd != self.last_active_day:
 405                                 self.last_active_day = yymmdd
 406                                 self.active_days.add(yymmdd)
 407
 408                         # timezone
 409                         self.commits_by_timezone[timezone] = self.commits_by_timezone.get(timezone, 0) + 1
 410
 411                 # TODO Optimize this, it's the worst bottleneck
 412                 # outputs "<stamp> <files>" for each revision
 413                 revlines = getpipeoutput(['git rev-list --pretty=format:"%%at %%T" %s' % getcommitrange('HEAD'), 'grep -v ^commit']).strip().split('\n')
 414                 lines = []
 415                 for revline in revlines:
 416                         time, rev = revline.split(' ')
 417                         linecount = self.getFilesInCommit(rev)
 418                         lines.append('%d %d' % (int(time), linecount))
 419
 420                 self.total_commits += len(lines)
 421                 for line in lines:
 422                         parts = line.split(' ')
 423                         if len(parts) != 2:
 424                                 continue
 425                         (stamp, files) = parts[0:2]
 426                         try:
 427                                 self.files_by_stamp[int(stamp)] = int(files)
 428                         except ValueError:
 429                                 print 'Warning: failed to parse line "%s"' % line
 430
 431                 # extensions and size of files
 432                 lines = getpipeoutput(['git ls-tree -r -l -z %s' % getcommitrange('HEAD', end_only = True)]).split('\000')
 433                 for line in lines:
 434                         if len(line) == 0:
 435                                 continue
 436                         parts = re.split('\s+', line, 5)
 437                         if parts[0] == '160000' and parts[3] == '-':
 438                                 # skip submodules
 439                                 continue
 440                         sha1 = parts[2]
 441                         size = int(parts[3])
 442                         fullpath = parts[4]
 443
 444                         self.total_size += size
 445                         self.total_files += 1
 446
 447                         filename = fullpath.split('/')[-1] # strip directories
 448                         if filename.find('.') == -1 or filename.rfind('.') == 0:
 449                                 ext = ''
 450                         else:
 451                                 ext = filename[(filename.rfind('.') + 1):]
 452                         if len(ext) > conf['max_ext_length']:
 453                                 ext = ''
 454
 455                         if ext not in self.extensions:
 456                                 self.extensions[ext] = {'files': 0, 'lines': 0}
 457
 458                         self.extensions[ext]['files'] += 1
 459                         try:
 460                                 self.extensions[ext]['lines'] += self.getLinesInBlob(sha1)
 461                         except:
 462                                 print 'Warning: Could not count lines for file "%s"' % line
 463
 464                 # line statistics
 465                 # outputs:
 466                 #  N files changed, N insertions (+), N deletions(-)
 467                 # <stamp> <author>
 468                 self.changes_by_date = {} # stamp -> { files, ins, del }
 469                 # computation of lines of code by date is better done
 470                 # on a linear history.
 471                 extra = ''
 472                 if conf['linear_linestats']:
 473                         extra = '--first-parent -m'
 474                 lines = getpipeoutput(['git log --shortstat %s --pretty=format:"%%at %%aN" %s' % (extra, getcommitrange('HEAD'))]).split('\n')
 475                 lines.reverse()
 476                 files = 0; inserted = 0; deleted = 0; total_lines = 0
 477                 author = None
 478                 for line in lines:
 479                         if len(line) == 0:
 480                                 continue
 481
 482                         # <stamp> <author>
 483                         if re.search('files? changed', line) == None:
 484                                 pos = line.find(' ')
 485                                 if pos != -1:
 486                                         try:
 487                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 488                                                 if author in conf['merge_authors']:
 489                                                         author = conf['merge_authors'][author]
 490                                                 self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted, 'lines': total_lines }
 491
 492                                                 date = datetime.datetime.fromtimestamp(stamp)
 493                                                 yymm = date.strftime('%Y-%m')
 494                                                 self.lines_added_by_month[yymm] = self.lines_added_by_month.get(yymm, 0) + inserted
 495                                                 self.lines_removed_by_month[yymm] = self.lines_removed_by_month.get(yymm, 0) + deleted
 496
 497                                                 yy = date.year
 498                                                 self.lines_added_by_year[yy] = self.lines_added_by_year.get(yy,0) + inserted
 499                                                 self.lines_removed_by_year[yy] = self.lines_removed_by_year.get(yy, 0) + deleted
 500
 501                                                 files, inserted, deleted = 0, 0, 0
 502                                         except ValueError:
 503                                                 print 'Warning: unexpected line "%s"' % line
 504                                 else:
 505                                         print 'Warning: unexpected line "%s"' % line
 506                         else:
 507                                 numbers = getstatsummarycounts(line)
 508
 509                                 if len(numbers) == 3:
 510                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 511                                         total_lines += inserted
 512                                         total_lines -= deleted
 513                                         self.total_lines_added += inserted
 514                                         self.total_lines_removed += deleted
 515
 516                                 else:
 517                                         print 'Warning: failed to handle line "%s"' % line
 518                                         (files, inserted, deleted) = (0, 0, 0)
 519                                 #self.changes_by_date[stamp] = { 'files': files, 'ins': inserted, 'del': deleted }
 520                 self.total_lines += total_lines
 521
 522                 # Per-author statistics
 523
 524                 # defined for stamp, author only if author commited at this timestamp.
 525                 self.changes_by_date_by_author = {} # stamp -> author -> lines_added
 526
 527                 # Similar to the above, but never use --first-parent
 528                 # (we need to walk through every commit to know who
 529                 # committed what, not just through mainline)
 530                 lines = getpipeoutput(['git log --shortstat --date-order --pretty=format:"%%at %%aN" %s' % (getcommitrange('HEAD'))]).split('\n')
 531                 lines.reverse()
 532                 files = 0; inserted = 0; deleted = 0
 533                 author = None
 534                 stamp = 0
 535                 for line in lines:
 536                         if len(line) == 0:
 537                                 continue
 538
 539                         # <stamp> <author>
 540                         if re.search('files? changed', line) == None:
 541                                 pos = line.find(' ')
 542                                 if pos != -1:
 543                                         try:
 544                                                 oldstamp = stamp
 545                                                 (stamp, author) = (int(line[:pos]), line[pos+1:])
 546                                                 if author in conf['merge_authors']:
 547                                                         author = conf['merge_authors'][author]
 548                                                 if oldstamp > stamp:
 549                                                         # clock skew, keep old timestamp to avoid having ugly graph
 550                                                         stamp = oldstamp
 551                                                 if author not in self.authors:
 552                                                         self.authors[author] = { 'lines_added' : 0, 'lines_removed' : 0, 'commits' : 0}
 553                                                 self.authors[author]['commits'] = self.authors[author].get('commits', 0) + 1
 554                                                 self.authors[author]['lines_added'] = self.authors[author].get('lines_added', 0) + inserted
 555                                                 self.authors[author]['lines_removed'] = self.authors[author].get('lines_removed', 0) + deleted
 556                                                 if stamp not in self.changes_by_date_by_author:
 557                                                         self.changes_by_date_by_author[stamp] = {}
 558                                                 if author not in self.changes_by_date_by_author[stamp]:
 559                                                         self.changes_by_date_by_author[stamp][author] = {}
 560                                                 self.changes_by_date_by_author[stamp][author]['lines_added'] = self.authors[author]['lines_added']
 561                                                 self.changes_by_date_by_author[stamp][author]['commits'] = self.authors[author]['commits']
 562                                                 files, inserted, deleted = 0, 0, 0
 563                                         except ValueError:
 564                                                 print 'Warning: unexpected line "%s"' % line
 565                                 else:
 566                                         print 'Warning: unexpected line "%s"' % line
 567                         else:
 568                                 numbers = getstatsummarycounts(line);
 569
 570                                 if len(numbers) == 3:
 571                                         (files, inserted, deleted) = map(lambda el : int(el), numbers)
 572                                 else:
 573                                         print 'Warning: failed to handle line "%s"' % line
 574                                         (files, inserted, deleted) = (0, 0, 0)
 575
 576         def refine(self):
 577                 # authors
 578                 # name -> {place_by_commits, commits_frac, date_first, date_last, timedelta}
 579                 self.authors_by_commits = getkeyssortedbyvaluekey(self.authors, 'commits')
 580                 self.authors_by_commits.reverse() # most first
 581                 for i, name in enumerate(self.authors_by_commits):
 582                         self.authors[name]['place_by_commits'] = i + 1
 583
 584                 for name in self.authors.keys():
 585                         a = self.authors[name]
 586                         a['commits_frac'] = (100 * float(a['commits'])) / self.getTotalCommits()
 587                         date_first = datetime.datetime.fromtimestamp(a['first_commit_stamp'])
 588                         date_last = datetime.datetime.fromtimestamp(a['last_commit_stamp'])
 589                         delta = date_last - date_first
 590                         a['date_first'] = date_first.strftime('%Y-%m-%d')
 591                         a['date_last'] = date_last.strftime('%Y-%m-%d')
 592                         a['timedelta'] = delta
 593                         if 'lines_added' not in a: a['lines_added'] = 0
 594                         if 'lines_removed' not in a: a['lines_removed'] = 0
 595
 596         def getActiveDays(self):
 597                 return self.active_days
 598
 599         def getActivityByDayOfWeek(self):
 600                 return self.activity_by_day_of_week
 601
 602         def getActivityByHourOfDay(self):
 603                 return self.activity_by_hour_of_day
 604
 605         def getAuthorInfo(self, author):
 606                 return self.authors[author]
 607
 608         def getAuthors(self, limit = None):
 609                 res = getkeyssortedbyvaluekey(self.authors, 'commits')
 610                 res.reverse()
 611                 return res[:limit]
 612
 613         def getCommitDeltaDays(self):
 614                 return (self.last_commit_stamp / 86400 - self.first_commit_stamp / 86400) + 1
 615
 616         def getDomainInfo(self, domain):
 617                 return self.domains[domain]
 618
 619         def getDomains(self):
 620                 return self.domains.keys()
 621
 622         def getFilesInCommit(self, rev):
 623                 try:
 624                         res = self.cache['files_in_tree'][rev]
 625                 except:
 626                         res = int(getpipeoutput(['git ls-tree -r --name-only "%s"' % rev, 'wc -l']).split('\n')[0])
 627                         if 'files_in_tree' not in self.cache:
 628                                 self.cache['files_in_tree'] = {}
 629                         self.cache['files_in_tree'][rev] = res
 630
 631                 return res
 632
 633         def getFirstCommitDate(self):
 634                 return datetime.datetime.fromtimestamp(self.first_commit_stamp)
 635
 636         def getLastCommitDate(self):
 637                 return datetime.datetime.fromtimestamp(self.last_commit_stamp)
 638
 639         def getLinesInBlob(self, sha1):
 640                 try:
 641                         res = self.cache['lines_in_blob'][sha1]
 642                 except:
 643                         res = int(getpipeoutput(['git cat-file blob %s' % sha1, 'wc -l']).split()[0])
 644                         if 'lines_in_blob' not in self.cache:
 645                                 self.cache['lines_in_blob'] = {}
 646                         self.cache['lines_in_blob'][sha1] = res
 647                 return res
 648
 649         def getTags(self):
 650                 lines = getpipeoutput(['git show-ref --tags', 'cut -d/ -f3'])
 651                 return lines.split('\n')
 652
 653         def getTagDate(self, tag):
 654                 return self.revToDate('tags/' + tag)
 655
 656         def getTotalAuthors(self):
 657                 return self.total_authors
 658
 659         def getTotalCommits(self):
 660                 return self.total_commits
 661
 662         def getTotalFiles(self):
 663                 return self.total_files
 664
 665         def getTotalLOC(self):
 666                 return self.total_lines
 667
 668         def getTotalSize(self):
 669                 return self.total_size
 670
 671         def revToDate(self, rev):
 672                 stamp = int(getpipeoutput(['git log --pretty=format:%%at "%s" -n 1' % rev]))
 673                 return datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d')
 674
 675 class ReportCreator:
 676         """Creates the actual report based on given data."""
 677         def __init__(self):
 678                 pass
 679
 680         def create(self, data, path):
 681                 self.data = data
 682                 self.path = path
 683
 684 def html_linkify(text):
 685         return text.lower().replace(' ', '_')
 686
 687 def html_header(level, text):
 688         name = html_linkify(text)
 689         return '\n<h%d><a href="#%s" name="%s">%s</a></h%d>\n\n' % (level, name, name, text, level)
 690
 691 class HTMLReportCreator(ReportCreator):
 692         def create(self, data, path):
 693                 ReportCreator.create(self, data, path)
 694                 self.title = data.projectname
 695
 696                 # copy static files. Looks in the binary directory, ../share/gitstats and /usr/share/gitstats
 697                 binarypath = os.path.dirname(os.path.abspath(__file__))
 698                 secondarypath = os.path.join(binarypath, '..', 'share', 'gitstats')
 699                 basedirs = [binarypath, secondarypath, '/usr/share/gitstats']
 700                 for file in ('gitstats.css', 'sortable.js', 'arrow-up.gif', 'arrow-down.gif', 'arrow-none.gif'):
 701                         for base in basedirs:
 702                                 src = base + '/' + file
 703                                 if os.path.exists(src):
 704                                         shutil.copyfile(src, path + '/' + file)
 705                                         break
 706                         else:
 707                                 print 'Warning: "%s" not found, so not copied (searched: %s)' % (file, basedirs)
 708
 709                 f = open(path + "/index.html", 'w')
 710                 format = '%Y-%m-%d %H:%M:%S'
 711                 self.printHeader(f)
 712
 713                 f.write('<h1>GitStats - %s</h1>' % data.projectname)
 714
 715                 self.printNav(f)
 716
 717                 f.write('<dl>')
 718                 f.write('<dt>Project name</dt><dd>%s</dd>' % (data.projectname))
 719                 f.write('<dt>Generated</dt><dd>%s (in %d seconds)</dd>' % (datetime.datetime.now().strftime(format), time.time() - data.getStampCreated()))
 720                 f.write('<dt>Generator</dt><dd><a href="http://gitstats.sourceforge.net/">GitStats</a> (version %s), %s, %s</dd>' % (getversion(), getgitversion(), getgnuplotversion()))
 721                 f.write('<dt>Report Period</dt><dd>%s to %s</dd>' % (data.getFirstCommitDate().strftime(format), data.getLastCommitDate().strftime(format)))
 722                 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())))
 723                 f.write('<dt>Total Files</dt><dd>%s</dd>' % data.getTotalFiles())
 724                 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))
 725                 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()))
 726                 f.write('<dt>Authors</dt><dd>%s (average %.1f commits per author)</dd>' % (data.getTotalAuthors(), (1.0 * data.getTotalCommits()) / data.getTotalAuthors()))
 727                 f.write('</dl>')
 728
 729                 f.write('</body>\n</html>')
 730                 f.close()
 731
 732                 ###
 733                 # Activity
 734                 f = open(path + '/activity.html', 'w')
 735                 self.printHeader(f)
 736                 f.write('<h1>Activity</h1>')
 737                 self.printNav(f)
 738
 739                 #f.write('<h2>Last 30 days</h2>')
 740
 741                 #f.write('<h2>Last 12 months</h2>')
 742
 743                 # Weekly activity
 744                 WEEKS = 32
 745                 f.write(html_header(2, 'Weekly activity'))
 746                 f.write('<p>Last %d weeks</p>' % WEEKS)
 747
 748                 # generate weeks to show (previous N weeks from now)
 749                 now = datetime.datetime.now()
 750                 deltaweek = datetime.timedelta(7)
 751                 weeks = []
 752                 stampcur = now
 753                 for i in range(0, WEEKS):
 754                         weeks.insert(0, stampcur.strftime('%Y-%W'))
 755                         stampcur -= deltaweek
 756
 757                 # top row: commits & bar
 758                 f.write('<table class="noborders"><tr>')
 759                 for i in range(0, WEEKS):
 760                         commits = 0
 761                         if weeks[i] in data.activity_by_year_week:
 762                                 commits = data.activity_by_year_week[weeks[i]]
 763
 764                         percentage = 0
 765                         if weeks[i] in data.activity_by_year_week:
 766                                 percentage = float(data.activity_by_year_week[weeks[i]]) / data.activity_by_year_week_peak
 767                         height = max(1, int(200 * percentage))
 768                         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))
 769
 770                 # bottom row: year/week
 771                 f.write('</tr><tr>')
 772                 for i in range(0, WEEKS):
 773                         f.write('<td>%s</td>' % (WEEKS - i))
 774                 f.write('</tr></table>')
 775
 776                 # Hour of Day
 777                 f.write(html_header(2, 'Hour of Day'))
 778                 hour_of_day = data.getActivityByHourOfDay()
 779                 f.write('<table><tr><th>Hour</th>')
 780                 for i in range(0, 24):
 781                         f.write('<th>%d</th>' % i)
 782                 f.write('</tr>\n<tr><th>Commits</th>')
 783                 fp = open(path + '/hour_of_day.dat', 'w')
 784                 for i in range(0, 24):
 785                         if i in hour_of_day:
 786                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 787                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%d</td>' % (r, hour_of_day[i]))
 788                                 fp.write('%d %d\n' % (i, hour_of_day[i]))
 789                         else:
 790                                 f.write('<td>0</td>')
 791                                 fp.write('%d 0\n' % i)
 792                 fp.close()
 793                 f.write('</tr>\n<tr><th>%</th>')
 794                 totalcommits = data.getTotalCommits()
 795                 for i in range(0, 24):
 796                         if i in hour_of_day:
 797                                 r = 127 + int((float(hour_of_day[i]) / data.activity_by_hour_of_day_busiest) * 128)
 798                                 f.write('<td style="background-color: rgb(%d, 0, 0)">%.2f</td>' % (r, (100.0 * hour_of_day[i]) / totalcommits))
 799                         else:
 800                                 f.write('<td>0.00</td>')
 801                 f.write('</tr></table>')
 802                 f.write('<img src="hour_of_day.png" alt="Hour of Day" />')
 803                 fg = open(path + '/hour_of_day.dat', 'w')
 804                 for i in range(0, 24):
 805                         if i in hour_of_day:
 806                                 fg.write('%d %d\n' % (i + 1, hour_of_day[i]))
 807                         else:
 808                                 fg.write('%d 0\n' % (i + 1))
 809                 fg.close()
 810
 811                 # Day of Week
 812                 f.write(html_header(2, 'Day of Week'))
 813                 day_of_week = data.getActivityByDayOfWeek()
 814                 f.write('<div class="vtable"><table>')
 815                 f.write('<tr><th>Day</th><th>Total (%)</th></tr>')
 816                 fp = open(path + '/day_of_week.dat', 'w')
 817                 for d in range(0, 7):
 818                         commits = 0
 819                         if d in day_of_week:
 820                                 commits = day_of_week[d]
 821                         fp.write('%d %s %d\n' % (d + 1, WEEKDAYS[d], commits))
 822                         f.write('<tr>')
 823                         f.write('<th>%s</th>' % (WEEKDAYS[d]))
 824                         if d in day_of_week:
 825                                 f.write('<td>%d (%.2f%%)</td>' % (day_of_week[d], (100.0 * day_of_week[d]) / totalcommits))
 826                         else:
 827                                 f.write('<td>0</td>')
 828                         f.write('</tr>')
 829                 f.write('</table></div>')
 830                 f.write('<img src="day_of_week.png" alt="Day of Week" />')
 831                 fp.close()
 832
 833                 # Hour of Week
 834                 f.write(html_header(2, 'Hour of Week'))
 835                 f.write('<table>')
 836
 837                 f.write('<tr><th>Weekday</th>')
 838                 for hour in range(0, 24):
 839                         f.write('<th>%d</th>' % (hour))
 840                 f.write('</tr>')
 841
 842                 for weekday in range(0, 7):
 843                         f.write('<tr><th>%s</th>' % (WEEKDAYS[weekday]))
 844                         for hour in range(0, 24):
 845                                 try:
 846                                         commits = data.activity_by_hour_of_week[weekday][hour]
 847                                 except KeyError:
 848                                         commits = 0
 849                                 if commits != 0:
 850                                         f.write('<td')
 851                                         r = 127 + int((float(commits) / data.activity_by_hour_of_week_busiest) * 128)
 852                                         f.write(' style="background-color: rgb(%d, 0, 0)"' % r)
 853                                         f.write('>%d</td>' % commits)
 854                                 else:
 855                                         f.write('<td></td>')
 856                         f.write('</tr>')
 857
 858                 f.write('</table>')
 859
 860                 # Month of Year
 861                 f.write(html_header(2, 'Month of Year'))
 862                 f.write('<div class="vtable"><table>')
 863                 f.write('<tr><th>Month</th><th>Commits (%)</th></tr>')
 864                 fp = open (path + '/month_of_year.dat', 'w')
 865                 for mm in range(1, 13):
 866                         commits = 0
 867                         if mm in data.activity_by_month_of_year:
 868                                 commits = data.activity_by_month_of_year[mm]
 869                         f.write('<tr><td>%d</td><td>%d (%.2f %%)</td></tr>' % (mm, commits, (100.0 * commits) / data.getTotalCommits()))
 870                         fp.write('%d %d\n' % (mm, commits))
 871                 fp.close()
 872                 f.write('</table></div>')
 873                 f.write('<img src="month_of_year.png" alt="Month of Year" />')
 874
 875                 # Commits by year/month
 876                 f.write(html_header(2, 'Commits by year/month'))
 877                 f.write('<div class="vtable"><table><tr><th>Month</th><th>Commits</th><th>Lines added</th><th>Lines removed</th></tr>')
 878                 for yymm in reversed(sorted(data.commits_by_month.keys())):
 879                         f.write('<tr><td>%s</td><td>%d</td><td>%d</td><td>%d</td></tr>' % (yymm, data.commits_by_month.get(yymm,0), data.lines_added_by_month.get(yymm,0), data.lines_removed_by_month.get(yymm,0)))
 880                 f.write('</table></div>')
 881                 f.write('<img src="commits_by_year_month.png" alt="Commits by year/month" />')
 882                 fg = open(path + '/commits_by_year_month.dat', 'w')
 883                 for yymm in sorted(data.commits_by_month.keys()):
 884                         fg.write('%s %s\n' % (yymm, data.commits_by_month[yymm]))
 885                 fg.close()
 886
 887                 # Commits by year
 888                 f.write(html_header(2, 'Commits by Year'))
 889                 f.write('<div class="vtable"><table><tr><th>Year</th><th>Commits (% of all)</th><th>Lines added</th><th>Lines removed</th></tr>')
 890                 for yy in reversed(sorted(data.commits_by_year.keys())):
 891                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d</td><td>%d</td></tr>' % (yy, data.commits_by_year.get(yy,0), (100.0 * data.commits_by_year.get(yy,0)) / data.getTotalCommits(), data.lines_added_by_year.get(yy,0), data.lines_removed_by_year.get(yy,0)))
 892                 f.write('</table></div>')
 893                 f.write('<img src="commits_by_year.png" alt="Commits by Year" />')
 894                 fg = open(path + '/commits_by_year.dat', 'w')
 895                 for yy in sorted(data.commits_by_year.keys()):
 896                         fg.write('%d %d\n' % (yy, data.commits_by_year[yy]))
 897                 fg.close()
 898
 899                 # Commits by timezone
 900                 f.write(html_header(2, 'Commits by Timezone'))
 901                 f.write('<table><tr>')
 902                 f.write('<th>Timezone</th><th>Commits</th>')
 903                 max_commits_on_tz = max(data.commits_by_timezone.values())
 904                 for i in sorted(data.commits_by_timezone.keys(), key = lambda n : int(n)):
 905                         commits = data.commits_by_timezone[i]
 906                         r = 127 + int((float(commits) / max_commits_on_tz) * 128)
 907                         f.write('<tr><th>%s</th><td style="background-color: rgb(%d, 0, 0)">%d</td></tr>' % (i, r, commits))
 908                 f.write('</tr></table>')
 909
 910                 f.write('</body></html>')
 911                 f.close()
 912
 913                 ###
 914                 # Authors
 915                 f = open(path + '/authors.html', 'w')
 916                 self.printHeader(f)
 917
 918                 f.write('<h1>Authors</h1>')
 919                 self.printNav(f)
 920
 921                 # Authors :: List of authors
 922                 f.write(html_header(2, 'List of Authors'))
 923
 924                 f.write('<table class="authors sortable" id="authors">')
 925                 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>')
 926                 for author in data.getAuthors(conf['max_authors']):
 927                         info = data.getAuthorInfo(author)
 928                         f.write('<tr><td>%s</td><td>%d (%.2f%%)</td><td>%d</td><td>%d</td><td>%s</td><td>%s</td><td>%s</td><td>%d</td><td>%d</td></tr>' % (author, info['commits'], info['commits_frac'], info['lines_added'], info['lines_removed'], info['date_first'], info['date_last'], info['timedelta'], len(info['active_days']), info['place_by_commits']))
 929                 f.write('</table>')
 930
 931                 allauthors = data.getAuthors()
 932                 if len(allauthors) > conf['max_authors']:
 933                         rest = allauthors[conf['max_authors']:]
 934                         f.write('<p class="moreauthors">These didn\'t make it to the top: %s</p>' % ', '.join(rest))
 935
 936                 f.write(html_header(2, 'Cumulated Added Lines of Code per Author'))
 937                 f.write('<img src="lines_of_code_by_author.png" alt="Lines of code per Author" />')
 938                 if len(allauthors) > conf['max_authors']:
 939                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 940
 941                 f.write(html_header(2, 'Commits per Author'))
 942                 f.write('<img src="commits_by_author.png" alt="Commits per Author" />')
 943                 if len(allauthors) > conf['max_authors']:
 944                         f.write('<p class="moreauthors">Only top %d authors shown</p>' % conf['max_authors'])
 945
 946                 fgl = open(path + '/lines_of_code_by_author.dat', 'w')
 947                 fgc = open(path + '/commits_by_author.dat', 'w')
 948
 949                 lines_by_authors = {} # cumulated added lines by
 950                 # author. to save memory,
 951                 # changes_by_date_by_author[stamp][author] is defined
 952                 # only at points where author commits.
 953                 # lines_by_authors allows us to generate all the
 954                 # points in the .dat file.
 955
 956                 # Don't rely on getAuthors to give the same order each
 957                 # time. Be robust and keep the list in a variable.
 958                 commits_by_authors = {} # cumulated added lines by
 959
 960                 self.authors_to_plot = data.getAuthors(conf['max_authors'])
 961                 for author in self.authors_to_plot:
 962                         lines_by_authors[author] = 0
 963                         commits_by_authors[author] = 0
 964                 for stamp in sorted(data.changes_by_date_by_author.keys()):
 965                         fgl.write('%d' % stamp)
 966                         fgc.write('%d' % stamp)
 967                         for author in self.authors_to_plot:
 968                                 if author in data.changes_by_date_by_author[stamp].keys():
 969                                         lines_by_authors[author] = data.changes_by_date_by_author[stamp][author]['lines_added']
 970                                         commits_by_authors[author] = data.changes_by_date_by_author[stamp][author]['commits']
 971                                 fgl.write(' %d' % lines_by_authors[author])
 972                                 fgc.write(' %d' % commits_by_authors[author])
 973                         fgl.write('\n')
 974                         fgc.write('\n')
 975                 fgl.close()
 976                 fgc.close()
 977
 978                 # Authors :: Author of Month
 979                 f.write(html_header(2, 'Author of Month'))
 980                 f.write('<table class="sortable" id="aom">')
 981                 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'])
 982                 for yymm in reversed(sorted(data.author_of_month.keys())):
 983                         authordict = data.author_of_month[yymm]
 984                         authors = getkeyssortedbyvalues(authordict)
 985                         authors.reverse()
 986                         commits = data.author_of_month[yymm][authors[0]]
 987                         next = ', '.join(authors[1:conf['authors_top']+1])
 988                         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)))
 989
 990                 f.write('</table>')
 991
 992                 f.write(html_header(2, 'Author of Year'))
 993                 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'])
 994                 for yy in reversed(sorted(data.author_of_year.keys())):
 995                         authordict = data.author_of_year[yy]
 996                         authors = getkeyssortedbyvalues(authordict)
 997                         authors.reverse()
 998                         commits = data.author_of_year[yy][authors[0]]
 999                         next = ', '.join(authors[1:conf['authors_top']+1])
1000                         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)))
1001                 f.write('</table>')
1002
1003                 # Domains
1004                 f.write(html_header(2, 'Commits by Domains'))
1005                 domains_by_commits = getkeyssortedbyvaluekey(data.domains, 'commits')
1006                 domains_by_commits.reverse() # most first
1007                 f.write('<div class="vtable"><table>')
1008                 f.write('<tr><th>Domains</th><th>Total (%)</th></tr>')
1009                 fp = open(path + '/domains.dat', 'w')
1010                 n = 0
1011                 for domain in domains_by_commits:
1012                         if n == conf['max_domains']:
1013                                 break
1014                         commits = 0
1015                         n += 1
1016                         info = data.getDomainInfo(domain)
1017                         fp.write('%s %d %d\n' % (domain, n , info['commits']))
1018                         f.write('<tr><th>%s</th><td>%d (%.2f%%)</td></tr>' % (domain, info['commits'], (100.0 * info['commits'] / totalcommits)))
1019                 f.write('</table></div>')
1020                 f.write('<img src="domains.png" alt="Commits by Domains" />')
1021                 fp.close()
1022
1023                 f.write('</body></html>')
1024                 f.close()
1025
1026                 ###
1027                 # Files
1028                 f = open(path + '/files.html', 'w')
1029                 self.printHeader(f)
1030                 f.write('<h1>Files</h1>')
1031                 self.printNav(f)
1032
1033                 f.write('<dl>\n')
1034                 f.write('<dt>Total files</dt><dd>%d</dd>' % data.getTotalFiles())
1035                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1036                 try:
1037                         f.write('<dt>Average file size</dt><dd>%.2f bytes</dd>' % (float(data.getTotalSize()) / data.getTotalFiles()))
1038                 except ZeroDivisionError:
1039                         pass
1040                 f.write('</dl>\n')
1041
1042                 # Files :: File count by date
1043                 f.write(html_header(2, 'File count by date'))
1044
1045                 # use set to get rid of duplicate/unnecessary entries
1046                 files_by_date = set()
1047                 for stamp in sorted(data.files_by_stamp.keys()):
1048                         files_by_date.add('%s %d' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1049
1050                 fg = open(path + '/files_by_date.dat', 'w')
1051                 for line in sorted(list(files_by_date)):
1052                         fg.write('%s\n' % line)
1053                 #for stamp in sorted(data.files_by_stamp.keys()):
1054                 #       fg.write('%s %d\n' % (datetime.datetime.fromtimestamp(stamp).strftime('%Y-%m-%d'), data.files_by_stamp[stamp]))
1055                 fg.close()
1056
1057                 f.write('<img src="files_by_date.png" alt="Files by Date" />')
1058
1059                 #f.write('<h2>Average file size by date</h2>')
1060
1061                 # Files :: Extensions
1062                 f.write(html_header(2, 'Extensions'))
1063                 f.write('<table class="sortable" id="ext"><tr><th>Extension</th><th>Files (%)</th><th>Lines (%)</th><th>Lines/file</th></tr>')
1064                 for ext in sorted(data.extensions.keys()):
1065                         files = data.extensions[ext]['files']
1066                         lines = data.extensions[ext]['lines']
1067                         try:
1068                                 loc_percentage = (100.0 * lines) / data.getTotalLOC()
1069                         except ZeroDivisionError:
1070                                 loc_percentage = 0
1071                         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, loc_percentage, lines / files))
1072                 f.write('</table>')
1073
1074                 f.write('</body></html>')
1075                 f.close()
1076
1077                 ###
1078                 # Lines
1079                 f = open(path + '/lines.html', 'w')
1080                 self.printHeader(f)
1081                 f.write('<h1>Lines</h1>')
1082                 self.printNav(f)
1083
1084                 f.write('<dl>\n')
1085                 f.write('<dt>Total lines</dt><dd>%d</dd>' % data.getTotalLOC())
1086                 f.write('</dl>\n')
1087
1088                 f.write(html_header(2, 'Lines of Code'))
1089                 f.write('<img src="lines_of_code.png" />')
1090
1091                 fg = open(path + '/lines_of_code.dat', 'w')
1092                 for stamp in sorted(data.changes_by_date.keys()):
1093                         fg.write('%d %d\n' % (stamp, data.changes_by_date[stamp]['lines']))
1094                 fg.close()
1095
1096                 f.write('</body></html>')
1097                 f.close()
1098
1099                 ###
1100                 # tags.html
1101                 f = open(path + '/tags.html', 'w')
1102                 self.printHeader(f)
1103                 f.write('<h1>Tags</h1>')
1104                 self.printNav(f)
1105
1106                 f.write('<dl>')
1107                 f.write('<dt>Total tags</dt><dd>%d</dd>' % len(data.tags))
1108                 if len(data.tags) > 0:
1109                         f.write('<dt>Average commits per tag</dt><dd>%.2f</dd>' % (1.0 * data.getTotalCommits() / len(data.tags)))
1110                 f.write('</dl>')
1111
1112                 f.write('<table class="tags">')
1113                 f.write('<tr><th>Name</th><th>Date</th><th>Commits</th><th>Authors</th></tr>')
1114                 # sort the tags by date desc
1115                 tags_sorted_by_date_desc = map(lambda el : el[1], reversed(sorted(map(lambda el : (el[1]['date'], el[0]), data.tags.items()))))
1116                 for tag in tags_sorted_by_date_desc:
1117                         authorinfo = []
1118                         self.authors_by_commits = getkeyssortedbyvalues(data.tags[tag]['authors'])
1119                         for i in reversed(self.authors_by_commits):
1120                                 authorinfo.append('%s (%d)' % (i, data.tags[tag]['authors'][i]))
1121                         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)))
1122                 f.write('</table>')
1123
1124                 f.write('</body></html>')
1125                 f.close()
1126
1127                 self.createGraphs(path)
1128
1129         def createGraphs(self, path):
1130                 print 'Generating graphs...'
1131
1132                 # hour of day
1133                 f = open(path + '/hour_of_day.plot', 'w')
1134                 f.write(GNUPLOT_COMMON)
1135                 f.write(
1136 """
1137 set output 'hour_of_day.png'
1138 unset key
1139 set xrange [0.5:24.5]
1140 set xtics 4
1141 set grid y
1142 set ylabel "Commits"
1143 plot 'hour_of_day.dat' using 1:2:(0.5) w boxes fs solid
1144 """)
1145                 f.close()
1146
1147                 # day of week
1148                 f = open(path + '/day_of_week.plot', 'w')
1149                 f.write(GNUPLOT_COMMON)
1150                 f.write(
1151 """
1152 set output 'day_of_week.png'
1153 unset key
1154 set xrange [0.5:7.5]
1155 set xtics 1
1156 set grid y
1157 set ylabel "Commits"
1158 plot 'day_of_week.dat' using 1:3:(0.5):xtic(2) w boxes fs solid
1159 """)
1160                 f.close()
1161
1162                 # Domains
1163                 f = open(path + '/domains.plot', 'w')
1164                 f.write(GNUPLOT_COMMON)
1165                 f.write(
1166 """
1167 set output 'domains.png'
1168 unset key
1169 unset xtics
1170 set yrange [0:]
1171 set grid y
1172 set ylabel "Commits"
1173 plot 'domains.dat' using 2:3:(0.5) with boxes fs solid, '' using 2:3:1 with labels rotate by 45 offset 0,1
1174 """)
1175                 f.close()
1176
1177                 # Month of Year
1178                 f = open(path + '/month_of_year.plot', 'w')
1179                 f.write(GNUPLOT_COMMON)
1180                 f.write(
1181 """
1182 set output 'month_of_year.png'
1183 unset key
1184 set xrange [0.5:12.5]
1185 set xtics 1
1186 set grid y
1187 set ylabel "Commits"
1188 plot 'month_of_year.dat' using 1:2:(0.5) w boxes fs solid
1189 """)
1190                 f.close()
1191
1192                 # commits_by_year_month
1193                 f = open(path + '/commits_by_year_month.plot', 'w')
1194                 f.write(GNUPLOT_COMMON)
1195                 f.write(
1196 """
1197 set output 'commits_by_year_month.png'
1198 unset key
1199 set xdata time
1200 set timefmt "%Y-%m"
1201 set format x "%Y-%m"
1202 set xtics rotate
1203 set bmargin 5
1204 set grid y
1205 set ylabel "Commits"
1206 plot 'commits_by_year_month.dat' using 1:2:(0.5) w boxes fs solid
1207 """)
1208                 f.close()
1209
1210                 # commits_by_year
1211                 f = open(path + '/commits_by_year.plot', 'w')
1212                 f.write(GNUPLOT_COMMON)
1213                 f.write(
1214 """
1215 set output 'commits_by_year.png'
1216 unset key
1217 set xtics 1 rotate
1218 set grid y
1219 set ylabel "Commits"
1220 set yrange [0:]
1221 plot 'commits_by_year.dat' using 1:2:(0.5) w boxes fs solid
1222 """)
1223                 f.close()
1224
1225                 # Files by date
1226                 f = open(path + '/files_by_date.plot', 'w')
1227                 f.write(GNUPLOT_COMMON)
1228                 f.write(
1229 """
1230 set output 'files_by_date.png'
1231 unset key
1232 set xdata time
1233 set timefmt "%Y-%m-%d"
1234 set format x "%Y-%m-%d"
1235 set grid y
1236 set ylabel "Files"
1237 set xtics rotate
1238 set ytics autofreq
1239 set bmargin 6
1240 plot 'files_by_date.dat' using 1:2 w steps
1241 """)
1242                 f.close()
1243
1244                 # Lines of Code
1245                 f = open(path + '/lines_of_code.plot', 'w')
1246                 f.write(GNUPLOT_COMMON)
1247                 f.write(
1248 """
1249 set output 'lines_of_code.png'
1250 unset key
1251 set xdata time
1252 set timefmt "%s"
1253 set format x "%Y-%m-%d"
1254 set grid y
1255 set ylabel "Lines"
1256 set xtics rotate
1257 set bmargin 6
1258 plot 'lines_of_code.dat' using 1:2 w lines
1259 """)
1260                 f.close()
1261
1262                 # Lines of Code Added per author
1263                 f = open(path + '/lines_of_code_by_author.plot', 'w')
1264                 f.write(GNUPLOT_COMMON)
1265                 f.write(
1266 """
1267 set terminal png transparent size 640,480
1268 set output 'lines_of_code_by_author.png'
1269 set key left top
1270 set xdata time
1271 set timefmt "%s"
1272 set format x "%Y-%m-%d"
1273 set grid y
1274 set ylabel "Lines"
1275 set xtics rotate
1276 set bmargin 6
1277 plot """
1278 )
1279                 i = 1
1280                 plots = []
1281                 for a in self.authors_to_plot:
1282                         i = i + 1
1283                         plots.append("""'lines_of_code_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1284                 f.write(", ".join(plots))
1285                 f.write('\n')
1286
1287                 f.close()
1288
1289                 # Commits per author
1290                 f = open(path + '/commits_by_author.plot', 'w')
1291                 f.write(GNUPLOT_COMMON)
1292                 f.write(
1293 """
1294 set terminal png transparent size 640,480
1295 set output 'commits_by_author.png'
1296 set key left top
1297 set xdata time
1298 set timefmt "%s"
1299 set format x "%Y-%m-%d"
1300 set grid y
1301 set ylabel "Commits"
1302 set xtics rotate
1303 set bmargin 6
1304 plot """
1305 )
1306                 i = 1
1307                 plots = []
1308                 for a in self.authors_to_plot:
1309                         i = i + 1
1310                         plots.append("""'commits_by_author.dat' using 1:%d title "%s" w lines""" % (i, a.replace("\"", "\\\"")))
1311                 f.write(", ".join(plots))
1312                 f.write('\n')
1313
1314                 f.close()
1315
1316                 os.chdir(path)
1317                 files = glob.glob(path + '/*.plot')
1318                 for f in files:
1319                         out = getpipeoutput([gnuplot_cmd + ' "%s"' % f])
1320                         if len(out) > 0:
1321                                 print out
1322
1323         def printHeader(self, f, title = ''):
1324                 f.write(
1325 """<?xml version="1.0" encoding="UTF-8"?>
1326 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
1327 <html xmlns="http://www.w3.org/1999/xhtml">
1328 <head>
1329         <title>GitStats - %s</title>
1330         <link rel="stylesheet" href="%s" type="text/css" />
1331         <meta name="generator" content="GitStats %s" />
1332         <script type="text/javascript" src="sortable.js"></script>
1333 </head>
1334 <body>
1335 """ % (self.title, conf['style'], getversion()))
1336
1337         def printNav(self, f):
1338                 f.write("""
1339 <div class="nav">
1340 <ul>
1341 <li><a href="index.html">General</a></li>
1342 <li><a href="activity.html">Activity</a></li>
1343 <li><a href="authors.html">Authors</a></li>
1344 <li><a href="files.html">Files</a></li>
1345 <li><a href="lines.html">Lines</a></li>
1346 <li><a href="tags.html">Tags</a></li>
1347 </ul>
1348 </div>
1349 """)
1350
1351
1352 class GitStats:
1353         def run(self, args_orig):
1354                 optlist, args = getopt.getopt(args_orig, 'c:')
1355                 for o,v in optlist:
1356                         if o == '-c':
1357                                 key, value = v.split('=', 1)
1358                                 if key not in conf:
1359                                         raise KeyError('no such key "%s" in config' % key)
1360                                 if isinstance(conf[key], int):
1361                                         conf[key] = int(value)
1362                                 elif isinstance(conf[key], dict):
1363                                         kk,vv = value.split(',', 1)
1364                                         conf[key][kk] = vv
1365                                 else:
1366                                         conf[key] = value
1367
1368                 if len(args) < 2:
1369                         print """
1370 Usage: gitstats [options] <gitpath..> <outputpath>
1371
1372 Options:
1373 -c key=value     Override configuration value
1374
1375 Default config values:
1376 %s
1377 """ % conf
1378                         sys.exit(0)
1379
1380                 outputpath = os.path.abspath(args[-1])
1381                 rundir = os.getcwd()
1382
1383                 try:
1384                         os.makedirs(outputpath)
1385                 except OSError:
1386                         pass
1387                 if not os.path.isdir(outputpath):
1388                         print 'FATAL: Output path is not a directory or does not exist'
1389                         sys.exit(1)
1390
1391                 if not getgnuplotversion():
1392                         print 'gnuplot not found'
1393                         sys.exit(1)
1394
1395                 print 'Output path: %s' % outputpath
1396                 cachefile = os.path.join(outputpath, 'gitstats.cache')
1397
1398                 data = GitDataCollector()
1399                 data.loadCache(cachefile)
1400
1401                 for gitpath in args[0:-1]:
1402                         print 'Git path: %s' % gitpath
1403
1404                         os.chdir(gitpath)
1405
1406                         print 'Collecting data...'
1407                         data.collect(gitpath)
1408
1409                 print 'Refining data...'
1410                 data.saveCache(cachefile)
1411                 data.refine()
1412
1413                 os.chdir(rundir)
1414
1415                 print 'Generating report...'
1416                 report = HTMLReportCreator()
1417                 report.create(data, outputpath)
1418
1419                 time_end = time.time()
1420                 exectime_internal = time_end - time_start
1421                 print 'Execution time %.5f secs, %.5f secs (%.2f %%) in external commands)' % (exectime_internal, exectime_external, (100.0 * exectime_external) / exectime_internal)
1422                 if sys.stdin.isatty():
1423                         print 'You may now run:'
1424                         print
1425                         print '   sensible-browser \'%s\'' % os.path.join(outputpath, 'index.html').replace("'", "'\\''")
1426                         print
1427
1428 if __name__=='__main__':
1429         g = GitStats()
1430         g.run(sys.argv[1:])
1431