csvdump.py

   1 #
   2 # aggregate per-month statistics for people
   3 #
   4 import sys, datetime
   5 import csv
   6
   7 class CSVStat:
   8     def __init__ (self, name, email, employer, date):
   9         self.name = name
  10         self.email = email
  11         self.employer = employer
  12         self.added = self.removed = self.changesets = 0
  13         self.date = date
  14     def accumulate (self, p):
  15         self.added = self.added + p.added
  16         self.removed = self.removed + p.removed
  17         self.changesets += 1
  18
  19 PeriodCommitHash = { }
  20
  21 def AccumulatePatch (p, Aggregate):
  22     if (Aggregate == 'week'):
  23         date = "%.2d-%.2d"%(p.date.isocalendar()[0], p.date.isocalendar()[1])
  24     elif (Aggregate == 'year'):
  25         date = "%.2d"%(p.date.year)
  26     else:
  27         date = "%.2d-%.2d-01"%(p.date.year, p.date.month)
  28     authdatekey = "%s-%s"%(p.author.name, date)
  29     if authdatekey not in PeriodCommitHash:
  30         empl = p.author.emailemployer (p.email, p.date)
  31         stat = CSVStat (p.author.name, p.email, empl, date)
  32         PeriodCommitHash[authdatekey] = stat
  33     else:
  34         stat = PeriodCommitHash[authdatekey]
  35     stat.accumulate (p)
  36
  37 ChangeSets = []
  38 FileTypes = []
  39
  40 def store_patch(patch):
  41     if not patch.merge:
  42         employer = patch.author.emailemployer(patch.email, patch.date)
  43         employer = employer.name.replace('"', '.').replace ('\\', '.')
  44         author = patch.author.name.replace ('"', '.').replace ('\\', '.')
  45         author = patch.author.name.replace ("'", '.')
  46         try:
  47             domain = patch.email.split('@')[1]
  48         except:
  49             domain = patch.email
  50         ChangeSets.append([patch.commit, str(patch.date),
  51                            patch.email, domain, author, employer,
  52                            patch.added, patch.removed])
  53         for (filetype, (added, removed)) in patch.filetypes.iteritems():
  54             FileTypes.append([patch.commit, filetype, added, removed])
  55
  56
  57 def save_csv (prefix='data'):
  58     # Dump the ChangeSets
  59     if len(ChangeSets) > 0:
  60         fd = open('%s-changesets.csv' % prefix, 'w')
  61         writer = csv.writer (fd, quoting=csv.QUOTE_NONNUMERIC)
  62         writer.writerow (['Commit', 'Date', 'Domain',
  63                           'Email', 'Name', 'Affliation',
  64                           'Added', 'Removed'])
  65         for commit in ChangeSets:
  66             writer.writerow(commit)
  67
  68     # Dump the file types
  69     if len(FileTypes) > 0:
  70         fd = open('%s-filetypes.csv' % prefix, 'w')
  71         writer = csv.writer (fd, quoting=csv.QUOTE_NONNUMERIC)
  72
  73         writer.writerow (['Commit', 'Type', 'Added', 'Removed'])
  74         for commit in FileTypes:
  75             writer.writerow(commit)
  76
  77
  78
  79 def OutputCSV (file):
  80     if file is None:
  81         return
  82     writer = csv.writer (file, quoting=csv.QUOTE_NONNUMERIC)
  83     writer.writerow (['Name', 'Email', 'Affliation', 'Date',
  84                       'Added', 'Removed', 'Changesets'])
  85     for date, stat in PeriodCommitHash.items():
  86         # sanitise names " is common and \" sometimes too
  87         empl_name = stat.employer.name.replace ('"', '.').replace ('\\', '.')
  88         author_name = stat.name.replace ('"', '.').replace ('\\', '.')
  89         writer.writerow ([author_name, stat.email, empl_name, stat.date,
  90                           stat.added, stat.removed, stat.changesets])
  91
  92 __all__ = [  'AccumulatePatch', 'OutputCSV', 'store_patch' ]