gistfile1.py

#!/usr/bin/env python

from mercurial import ui, hg, cmdutil, match
from collections import defaultdict
import json
import sys

if len (sys.argv) == 1:
    print(  "\nThis script determines contributor numbers between a certain range of commits." +
            "\nOrder of arguments is important:" + 
            "\n\n contrib.py [local mercurial repo] [employee emails (flat)]\n")
    exit()


repo = hg.repository( ui.ui(), sys.argv[1] )
employees = {}

res_emp = list()
res_vol = list()
res_tot = list()
res_emp_cont = list()
res_vol_cont = list()
res_tot_cont = list()
res_vol_perc = list()
res_leverage_people = list()
res_leverage_patches = list()

hg_revnums =[ 113672, 117931, 122315, 126855, 131682, 136272, 141367, 147372,
              152943, 159494, 166629, 173936, 180520, 187591, 195409, 202941,
              210096, 217976, 223387, 230280] 


def sanitize(s):
    return s.replace(u"\u201c", '"').replace(u"\u201d", '"').replace(u"\u2018", "'").replace(u"\u2019", "'") 

emails = list()

with open(sys.argv[2]) as f:
    lines = f.readlines()
    if not lines[-1].split():
        lines.pop()
    emails += map(lambda x: x.split()[0], lines)

def count_contribs(from_rev, to_rev):

    authors = defaultdict(int)
    pats = ()
    opts = {'rev': [to_rev + ':' + from_rev]}
    matchfn = match.match(repo.root, repo.getcwd(), pats)
    def prep(ctx, fns):
        rev = ctx.rev()
        if len(repo.changelog.parentrevs(rev)) == 2:
            return

    for rev in cmdutil.walkchangerevs(repo, matchfn, opts, prep):
        author = str(rev.user()).decode('utf-8')
        authors[author] += 1
    
    employee_authors = set()
    volunteer_authors = set()
    partials = set()
    
    for author in authors.keys():  #All entries in authors.keys and emails should be unique at this point.
        # check to see if it's an @mozilla address 
        if 0 < author.find("@mozilla."): # might be a problem of people own mozilla.whoever.wherever domains
            employee_authors.add(author)
        #    print("Employee:    " + author).encode('utf-8')
        else: 
        # ok, now see if it's on the list of known bugmail or employee addresses
            for email in emails:
                if author in email:
                    employee_authors.add(author)
        #            print("Employee:    " + author).encode('utf-8')
                    break
        # OK, welp. If we're here, they're probably a contributor.
            else: 
                volunteer_authors.add(author)
        #        print("Contributor: " + author).encode('utf-8')


    emp_contributions = sum(map(lambda x: authors[x], filter(lambda x: x in employee_authors, authors)))
    vol_contributions = sum(map(lambda x: authors[x], filter(lambda x: x in volunteer_authors, authors)))
    res_emp.append( len(employee_authors) )
    res_vol.append( len(volunteer_authors) )
    res_tot.append( len(employee_authors) + len(volunteer_authors) ) 
    res_emp_cont.append(emp_contributions)
    res_vol_cont.append(vol_contributions)
    res_tot_cont.append(emp_contributions + vol_contributions)  
    res_vol_perc.append(100 * vol_contributions / (emp_contributions + vol_contributions) ) 
    res_leverage_people.append(float(len(volunteer_authors)) / float(len(employee_authors)))
    res_leverage_patches.append( float(vol_contributions) / float(len(employee_authors)))

for c in range(1, len(hg_revnums) - 1) :
    count_contribs(str(hg_revnums[c-1]+1), str(hg_revnums[c]) )

print "employees: " + str(res_emp) + ","
print "volunteers: " + str(res_vol) + ","
print "total: " + str(res_tot) + ","
print "employee_commits: " + str(res_emp_cont) + ","
print "volunteer_commits: " + str(res_vol_cont) + ","
print "total_commits: " + str(res_tot_cont) + ","
print "volunteer_commit_percentage: " + str(res_vol_perc) + ","
print "leverage_people: " + str(res_leverage_people) + ","
print "leverage_patches: " + str(res_leverage_patches)