Made the CSV file aggregating data by weeks or months

When using -w option together with the -x file option, the data exported
in the CSV file are aggregated by weeks instead of months (the default).
This is useful to extract meaningful stats on short periods.

diff --git a/README b/README
index 26cf81f75b9151b60488d3e3f835aa50ee7f6cf1..a5bdf638d1cf5fcf8c6a2dcb3cd8ac3e0024a0f7 100644 (file)
--- a/README
+++ b/README
@@ -55,6 +55,11 @@ be:
        -u      Group all unknown developers under the "(Unknown)"
                employer. 
 
+    -x file  Export raw statistics as CSV.
+
+    -w  Aggregate the data by weeks instead of months in the
+        CSV file when -x is used.
+
        -z      Dump out the hacker database to "database.dump".
 
 A typical command line used to generate the "who write 2.6.x" LWN articles

diff --git a/csv.py b/csv.py
index 34ea10aba6b91f0b0f751707feaccc350b848b6f..cec1f063b02d83765f4bd74b1635fd0807e9ce46 100644 (file)
--- a/csv.py
+++ b/csv.py
@@ -15,8 +15,10 @@ class CSVStat:
 
 PeriodCommitHash = { }
 
-def AccumulatePatch (p):
+def AccumulatePatch (p, Aggregate):
     date = "%.2d-%.2d-01"%(p.date.year, p.date.month)
+    if (Aggregate == 'week'):
+        date = "%.2d-%.2d"%(p.date.isocalendar()[0], p.date.isocalendar()[1])
     authdatekey = "%s-%s"%(p.author.name, date)
     if authdatekey not in PeriodCommitHash:
         empl = p.author.emailemployer (p.email, p.date)

diff --git a/gitdm b/gitdm
index 47d19669f152e6f25daf69321e320c883a88a8cc..70f813885f63ce903534f549b3f59bb7d20a3474 100755 (executable)
--- a/gitdm
+++ b/gitdm
@@ -36,6 +36,7 @@ AkpmOverLt = 0
 DumpDB = 0
 CFName = 'gitdm.config'
 DirName = ''
+Aggregate = 'month'
 
 #
 # Options:
@@ -52,14 +53,15 @@ DirName = ''
 # -s           Ignore author SOB lines
 # -u           Map unknown employers to '(Unknown)'
 # -x file.csv   Export raw statistics as CSV
+# -w        Aggregrate the raw statistics by weeks instead of months
 # -z           Dump out the hacker database at completion
 
 def ParseOpts ():
     global MapUnknown, DevReports
     global DateStats, AuthorSOBs, FileFilter, AkpmOverLt, DumpDB
-    global CFName, CSVFile, DirName
+    global CFName, CSVFile, DirName, Aggregate
 
-    opts, rest = getopt.getopt (sys.argv[1:], 'ab:dc:Dh:l:o:r:sux:z')
+    opts, rest = getopt.getopt (sys.argv[1:], 'a:b:dc:Dh:l:o:r:suwx:z')
     for opt in opts:
         if opt[0] == '-a':
             AkpmOverLt = 1
@@ -87,6 +89,8 @@ def ParseOpts ():
         elif opt[0] == '-x':
             CSVFile = open (opt[1], 'w')
             print "open output file " + opt[1] + "\n"
+        elif opt [0] == '-w':
+            Aggregate = 'week'
         elif opt[0] == '-z':
             DumpDB = 1
         
@@ -370,7 +374,7 @@ while (1):
         for hacker in p.reports:
             hacker.addreport (p)
     CSCount += 1
-    csv.AccumulatePatch (p)
+    csv.AccumulatePatch (p, Aggregate)
 print >> sys.stderr, 'Grabbing changesets...done       '
 
 if DumpDB: