Commit our version of gitdm for release stats

This comes from git://git.lwn.net/gitdm.git

gitdm/.gitignore

+*.pyc
+*~

gitdm/COPYING

+The code in this directory can be distributed under the terms of the GNU
+General Public License, version 2.  

gitdm/ConfigFile.py

+#
+# Stuff for dealing with configuration files.
+#
+#
+# This code is part of the LWN git data miner.
+#
+# Copyright 2007-11 Eklektix, Inc.
+# Copyright 2007-11 Jonathan Corbet <corbet@lwn.net>
+#
+# This file may be distributed under the terms of the GNU General
+# Public License, version 2.
+#
+import sys, re, datetime, os.path
+import database
+
+#
+# Read a line and strip out junk.
+#
+def ReadConfigLine (file):
+    line = file.readline ()
+    if not line:
+        return None
+    line = line.split('#')[0] # Get rid of any comments
+    line = line.strip () # and extra white space
+    if len (line) == 0: # we got rid of everything
+        return ReadConfigLine (file)
+    return line
+
+#
+# Give up and die.
+#
+def croak (message):
+    sys.stderr.write (message + '\n')
+    sys.exit (1)
+
+#
+# Read a list of email aliases.
+#
+def ReadEmailAliases (name):
+    try:
+        file = open (name, 'r')
+    except IOError:
+        croak ('Unable to open email alias file %s' % (name))
+    line = ReadConfigLine (file)
+    while line:
+        m = re.match ('^("[^"]+"|\S+)\s+(.+)$', line)
+        if not m or len (m.groups ()) != 2:
+            croak ('Funky email alias line "%s"' % (line))
+        if m and m.group (2).find ('@') <= 0:
+            croak ('Non-addresses in email alias "%s"' % (line))
+        database.AddEmailAlias (m.group (1).replace ('"', ''), m.group (2))
+        line = ReadConfigLine (file)
+    file.close ()
+
+#
+# The Email/Employer map
+#
+EMMpat = re.compile (r'^([^\s]+)\s+([^<]+)\s*(<\s*(\d+-\d+-\d+)\s*)?$')
+
+def ReadEmailEmployers (name):
+    try:
+        file = open (name, 'r')
+    except IOError:
+        croak ('Unable to open email/employer file %s' % (name))
+    line = ReadConfigLine (file)
+    while line:
+        m = EMMpat.match (line)
+        if not m:
+            croak ('Funky email/employer line "%s"' % (line))
+        email = m.group (1)
+        company = m.group (2).strip ()
+        enddate = ParseDate (m.group (4))
+        database.AddEmailEmployerMapping (email, company, enddate)
+        line = ReadConfigLine (file)
+    file.close ()
+
+def ParseDate (cdate):
+    if not cdate:
+        return None
+    sdate = cdate.split ('-')
+    return datetime.date (int (sdate[0]), int (sdate[1]), int (sdate[2]))
+
+
+def ReadGroupMap (fname, employer):
+    try:
+        file = open (fname, 'r')
+    except IOError:
+        croak ('Unable to open group map file %s' % (fname))
+    line = ReadConfigLine (file)
+    while line:
+        database.AddEmailEmployerMapping (line, employer)
+        line = ReadConfigLine (file)
+    file.close ()
+
+#
+# Read in a virtual employer description.
+#
+def ReadVirtual (file, name):
+    ve = database.VirtualEmployer (name)
+    line = ReadConfigLine (file)
+    while line:
+        sl = line.split (None, 1)
+        first = sl[0]
+        if first == 'end':
+            ve.store ()
+            return
+        #
+        # Zap the "%" syntactic sugar if it's there
+        #
+        if first[-1] == '%':
+            first = first[:-1]
+        try:
+            percent = int (first)
+        except ValueError:
+            croak ('Bad split value "%s" for virtual empl %s' % (first, name))
+        if not (0 < percent <= 100):
+            croak ('Bad split value "%s" for virtual empl %s' % (first, name))
+        ve.addsplit (' '.join (sl[1:]), percent/100.0)
+        line = ReadConfigLine (file)
+    #
+    # We should never get here
+    #
+    croak ('Missing "end" line for virtual employer %s' % (name))
+
+#
+# Read an overall config file.
+#
+
+def ConfigFile (name, confdir):
+    try:
+        file = open (name, 'r')
+    except IOError:
+        croak ('Unable to open config file %s' % (name))
+    line = ReadConfigLine (file)
+    while line:
+        sline = line.split (None, 2)
+        if len (sline) < 2:
+            croak ('Funky config line: "%s"' % (line))
+        if sline[0] == 'EmailAliases':
+            ReadEmailAliases (os.path.join (confdir, sline[1]))
+        elif sline[0] == 'EmailMap':
+            ReadEmailEmployers (os.path.join (confdir, sline[1]))
+        elif sline[0] == 'GroupMap':
+            if len (sline) != 3:
+                croak ('Funky group map line "%s"' % (line))
+            ReadGroupMap (os.path.join (confdir, sline[1]), sline[2])
+        elif sline[0] == 'VirtualEmployer':
+            ReadVirtual (file, ' '.join (sline[1:]))
+        else:
+            croak ('Unrecognized config line: "%s"' % (line))
+        line = ReadConfigLine (file)
+        

gitdm/README

+The code in this directory makes up the "git data miner," a simple hack
+which attempts to figure things out from the revision history in a git
+repository. 
+
+
+INSTALLING GITDM
+
+gitdm is a python script and doesn't need to be proper installed like other
+normal programs. You just have to adjust your PATH variable, pointing it to
+the directory of gitdm or alternatively create a symbolic link of the script
+inside /usr/bin.
+
+Before actually run gitdm you may want also to update the configuration file
+(gitdm.config) with the needed information.
+
+
+RUNNING GITDM
+
+Run it like this:
+
+   git log -p -M [details] | gitdm [options]
+
+The [details] tell git which changesets are of interest; the [options] can
+be:
+
+	-a	If a patch contains signoff lines from both Andrew Morton 
+		and Linus Torvalds, omit Linus's.
+
+	-b dir	Specify the base directory to fetch the configuration files.
+
+	-c file Specify the name of the gitdm configuration file.  
+	   	By default, "./gitdm.config" is used.
+
+	-d	Omit the developer reports, giving employer information
+         	only. 
+
+	-D	Rather than create the usual statistics, create a 
+        file (datelc) providing lines changed per day, where the first column
+        displays the changes happened only on that day and the second sums
+        the day it happnened with the previous ones. This option is suitable
+        for feeding to a tool like gnuplot.
+
+	-h file	Generate HTML output to the given file
+
+	-l num	Only list the top <num> entries in each report.
+
+	-o file	Write text output to the given file (default is stdout).
+
+	-r pat	Only generate statistics for changes to files whose 
+	   	name matches the given regular expression.
+
+	-s	Ignore Signed-off-by lines which match the author of 
+		each patch.
+
+	-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
+looks like:
+
+    git log -p -M v2.6.19..v2.6.20 | \
+	gitdm -u -s -a -o results -h results.html
+
+
+CONFIGURATION FILE
+
+The main purpose of the configuration file is to direct the mapping of
+email addresses onto employers.  Please note that the config file parser is
+exceptionally stupid and unrobust at this point, but it gets the job done.  
+
+Blank lines and lines beginning with "#" are ignored.  Everything else
+specifies a file with some sort of mapping:
+
+EmailAliases file
+
+	Developers often post code under a number of different email
+	addresses, but it can be desirable to group them all together in
+	the statistics.  An EmailAliases file just contains a bunch of
+	lines of the form:
+
+		alias@address  canonical@address
+
+	Any patches originating from alias@address will be treated as if
+	they had come from canonical@address.
+
+    It may happen that some people set their git user data in the
+    following form: "joe.hacker@acme.org <Joe Hacker>". The 
+    "Joe Hacker" is then considered as the email... but gitdm says
+    it is a "Funky" email. An alias line in the following form can
+    be used to alias these commits aliased to the correct email
+    address:
+
+        "Joe Hacker" joe.hacker@acme.org
+
+
+EmailMap file
+
+	Map email addresses onto employers.  These files contain lines
+	like:
+
+		[user@]domain  employer  [< yyyy-mm-dd]
+
+	If the "user@" portion is missing, all email from the given domain
+	will be treated as being associated with the given employer.  If a
+	date is provided, the entry is only valid up to that date;
+	otherwise it is considered valid into the indefinite future.  This
+	feature can be useful for properly tracking developers' work when
+	they change employers but do not change email addresses.
+
+
+GroupMap file employer
+
+	This is a variant of EmailMap provided for convenience; it contains
+	email addresses only, all of which are associated with the given
+	employer.
+
+VirtualEmployer name
+    nn% employer1
+    ...
+end
+
+	This construct (which appears in the main configuration file)
+    	allows causes the creation of a fake employer with the given
+    	"name".  It directs that any contributions attributed to that
+    	employer should be split to other (real) employers using the given
+    	percentages.  The functionality works, but is primitive - there is,
+	for example, no check to ensure that the percentages add up to
+    	something rational.
+
+
+OTHER TOOLS
+
+A few other tools have been added to this repository:
+
+  treeplot
+	Reads a set of commits, then generates a graphviz file charting the
+	flow of patches into the mainline.  Needs to be smarter, but, then,
+	so does everything else in this directory.
+
+  findoldfiles
+	Simple brute-force crawler which outputs the names of any files
+	which have not been touched since the original (kernel) commit.
+
+  committags
+	I needed to be able to quickly associate a given commit with the
+	major release which contains it.  First attempt used 
+	"git tags --contains="; after it ran for a solid week, I concluded
+	there must be a better way.  This tool just reads through the repo,
+	remembering tags, and creating a Python dictionary containing the
+	association.  The result is an ugly 10mb pickle file, but, even so,
+	it's still a better way.
+
+  linetags
+	Crawls through a directory hierarchy, counting how many lines of
+	code are associated with each major release.  Needs the pickle file
+	from committags to get the job done.
+
+
+NOTES AND CREDITS
+
+Gitdm was written by Jonathan Corbet; many useful contributions have come
+from Greg Kroah-Hartman.
+
+Please note that this tool is provided in the hope that it will be useful,
+but it is not put forward as an example of excellence in design or
+implementation.  Hacking on gitdm tends to stop the moment it performs
+whatever task is required of it at the moment.  Patches to make it less
+hacky, less ugly, and more robust are welcome.
+
+Jonathan Corbet
+corbet@lwn.net

gitdm/committags

+#!/usr/bin/python
+#
+# Generate a database of commits and major versions they went into.
+#
+# committags [git-args]
+#
+# This code is part of the LWN git data miner.
+#
+# Copyright 2007-11 Eklektix, Inc.
+# Copyright 2007-11 Jonathan Corbet <corbet@lwn.net>
+#
+# This file may be distributed under the terms of the GNU General
+# Public License, version 2.
+#
+import sys
+import re
+import os
+import pickle
+
+git = 'git log --decorate '
+if len(sys.argv) > 1:
+    git += ' '.join(sys.argv[1:])
+input = os.popen(git, 'r')
+
+DB = { }
+Tag = 'None'
+
+tagline = re.compile(r'^commit ([\da-f]+) .*tag: (v2\.6\.\d\d)')
+commit = re.compile(r'^commit ([\da-f]+)')
+
+for line in input.readlines():
+    if not line.startswith('commit'):
+        continue  # This makes it go faster
+    m = tagline.search(line)
+    if m:
+        DB[m.group(1)] = Tag = m.group(2)
+    else:
+        m = commit.search(line)
+        if m:
+            DB[m.group(1)] = Tag
+
+print 'Found %d commits' % (len(DB.keys()))
+out = open('committags.db', 'w')
+pickle.dump(DB, out)
+out.close()

gitdm/csv.py

+#
+# aggregate per-month statistics for people
+#
+import sys, datetime
+
+class CSVStat:
+    def __init__ (self, name, employer, date):
+        self.name = name
+        self.employer = employer
+        self.added = self.removed = 0
+        self.date = date
+    def accumulate (self, p):
+        self.added = self.added + p.added
+        self.removed = self.removed + p.removed
+
+PeriodCommitHash = { }
+
+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)
+        stat = CSVStat (p.author.name, empl, date)
+        PeriodCommitHash[authdatekey] = stat
+    else:
+        stat = PeriodCommitHash[authdatekey]
+    stat.accumulate (p)
+
+def OutputCSV (file):
+    if file is None:
+        return
+    file.write ("Name\tAffliation\tDate\tAdded\tRemoved\n")
+    for date, stat in PeriodCommitHash.items():
+        # sanitise names " is common and \" sometimes too
+        empl_name = stat.employer.name.replace ("\"", ".").replace ("\\", ".")
+        author_name = stat.name.replace ("\"", ".").replace ("\\", ".")
+        file.write ("\"%s\"\t\"%s\"\t%s\t%d\t%d\n"%(author_name, empl_name, stat.date, \
+                                                    stat.added, stat.removed))

gitdm/database.py

+#
+# The "database".
+#
+# This code is part of the LWN git data miner.
+#
+# Copyright 2007-11 Eklektix, Inc.
+# Copyright 2007-11 Jonathan Corbet <corbet@lwn.net>
+#
+# This file may be distributed under the terms of the GNU General
+# Public License, version 2.
+#
+import sys, datetime
+
+
+class Hacker:
+    def __init__ (self, name, id, elist, email):
+        self.name = name
+        self.id = id
+        self.employer = [ elist ]
+        self.email = [ email ]
+        self.added = self.removed = 0
+        self.patches = [ ]
+        self.signoffs = [ ]
+        self.reviews = [ ]
+        self.tested = [ ]
+        self.reports = [ ]
+        self.testcred = self.repcred = 0
+
+    def addemail (self, email, elist):
+        self.email.append (email)
+        self.employer.append (elist)
+        HackersByEmail[email] = self
+
+    def emailemployer (self, email, date):
+        for i in range (0, len (self.email)):
+            if self.email[i] == email:
+                for edate, empl in self.employer[i]:
+                    if edate > date:
+                        return empl
+        print 'OOPS.  ', self.name, self.employer, self.email, email, date
+        return None # Should not happen
+
+    def addpatch (self, patch):
+        self.added += patch.added
+        self.removed += patch.removed
+        self.patches.append (patch)
+
+    #
+    # There's got to be a better way.
+    #
+    def addsob (self, patch):
+        self.signoffs.append (patch)
+    def addreview (self, patch):
+        self.reviews.append (patch)
+    def addtested (self, patch):
+        self.tested.append (patch)
+    def addreport (self, patch):
+        self.reports.append (patch)
+
+    def reportcredit (self, patch):
+        self.repcred += 1
+    def testcredit (self, patch):
+        self.testcred += 1
+
+HackersByName = { }
+HackersByEmail = { }
+HackersByID = { }
+MaxID = 0
+
+def StoreHacker (name, elist, email):
+    global MaxID
+
+    id = MaxID
+    MaxID += 1
+    h = Hacker (name, id, elist, email)
+    HackersByName[name] = h
+    HackersByEmail[email] = h
+    HackersByID[id] = h
+    return h
+
+def LookupEmail (addr):
+    try:
+        return HackersByEmail[addr]
+    except KeyError:
+        return None
+
+def LookupName (name):
+    try:
+        return HackersByName[name]
+    except KeyError:
+        return None
+
+def LookupID (id):
+    try:
+        return HackersByID[id]
+    except KeyError:
+        return None
+
+def AllHackers ():
+    return HackersByID.values ()
+#    return [h for h in HackersByID.values ()] #  if (h.added + h.removed) > 0]
+
+def DumpDB ():
+    out = open ('database.dump', 'w')
+    names = HackersByName.keys ()
+    names.sort ()
+    for name in names:
+        h = HackersByName[name]
+        out.write ('%4d %s %d p (+%d -%d) sob: %d\n' % (h.id, h.name,
+                                                        len (h.patches),
+                                                        h.added, h.removed,
+                                                        len (h.signoffs)))
+        for i in range (0, len (h.email)):
+            out.write ('\t%s -> \n' % (h.email[i]))
+            for date, empl in h.employer[i]:
+                out.write ('\t\t %d-%d-%d %s\n' % (date.year, date.month, date.day,
+                                                 empl.name))
+
+#
+# Employer info.
+#
+class Employer:
+    def __init__ (self, name):
+        self.name = name
+        self.added = self.removed = self.count = self.changed = 0
+        self.sobs = 0
+        self.hackers = [ ]
+
+    def AddCSet (self, patch):
+        self.added += patch.added
+        self.removed += patch.removed
+        self.changed += max(patch.added, patch.removed)
+        self.count += 1
+        if patch.author not in self.hackers:
+            self.hackers.append (patch.author)
+
+    def AddSOB (self):
+        self.sobs += 1
+
+Employers = { }
+
+def GetEmployer (name):
+    try:
+        return Employers[name]
+    except KeyError:
+        e = Employer (name)
+        Employers[name] = e
+        return e
+
+def AllEmployers ():
+    return Employers.values ()
+
+#
+# Certain obnoxious developers, who will remain nameless (because we
+# would never want to run afoul of Thomas) want their work split among
+# multiple companies.  Let's try to cope with that.  Let's also hope
+# this doesn't spread.
+#
+class VirtualEmployer (Employer):
+    def __init__ (self, name):
+        Employer.__init__ (self, name)
+        self.splits = [ ]
+
+    def addsplit (self, name, fraction):
+        self.splits.append ((name, fraction))
+
+    #
+    # Go through and (destructively) apply our credits to the
+    # real employer.  Only one level of weirdness is supported.
+    #
+    def applysplits (self):
+        for name, fraction in self.splits:
+            real = GetEmployer (name)
+            real.added += int (self.added*fraction)
+            real.removed += int (self.removed*fraction)
+            real.changed += int (self.changed*fraction)
+            real.count += int (self.count*fraction)
+        self.__init__ (name) # Reset counts just in case
+
+    def store (self):
+        if Employers.has_key (self.name):
+            print Employers[self.name]
+            sys.stderr.write ('WARNING: Virtual empl %s overwrites another\n'
+                              % (self.name))
+        if len (self.splits) == 0:
+            sys.stderr.write ('WARNING: Virtual empl %s has no splits\n'
+                              % (self.name))
+            # Should check that they add up too, but I'm lazy
+        Employers[self.name] = self
+
+#
+# Mix all the virtual employers into their real destinations.
+#
+def MixVirtuals ():
+    for empl in AllEmployers ():
+        if isinstance (empl, VirtualEmployer):
+            empl.applysplits ()
+
+#
+# The email map.
+#
+EmailAliases = { }
+
+def AddEmailAlias (variant, canonical):
+    if EmailAliases.has_key (variant):
+        sys.stderr.write ('Duplicate email alias for %s\n' % (variant))
+    EmailAliases[variant] = canonical
+
+def RemapEmail (email):
+    email = email.lower ()
+    try:
+        return EmailAliases[email]
+    except KeyError:
+        return email
+
+#
+# Email-to-employer mapping.
+#
+EmailToEmployer = { }
+nextyear = datetime.date.today () + datetime.timedelta (days = 365)
+
+def AddEmailEmployerMapping (email, employer, end = nextyear):
+    if end is None:
+        end = nextyear