def print_prs(project_flag):
"""
Calculates the number of merged and closed pull requests, as well as the high
frequency pull requesters, and prints to stdout. Takes in a flag, defined by
constants.py, that determines whether to look at stack repos or ml repos.
"""
constants = return_constants(project_flag)
repos = constants["repos"]
directory = constants["pulls-fpath"]
for repo_dict in repos:
repo = repo_dict["name"]
owner = repo_dict["user"]
print repo, owner
# grab pull requests
pulls = json.load(
open(os.path.join(directory, "{}_pulls.json".format(repo))))
print 'num_pulls,{}'.format(len(pulls))
# sort
merged = 0
closed = 0
open_pulls = 0
merged_users = {}
closed_users = {}
for p in pulls:
merged_at = p["merged_at"]
closed_at = p["closed_at"]
username = p["user"]["login"]
# make an exception for systemml
if closed_at != None and repo == systemml_name:
merged_at = 1 if is_merged_systemml(p) else None
# count
if merged_at != None:
merged += 1
merged_users[username] = merged_users.get(username, 0) + 1
elif closed_at != None:
closed += 1
closed_users[username] = closed_users.get(username, 0) + 1
else:
open_pulls += 1
# sort all users who have had merged prs
merged_users_str = sort_by_val_return_str(merged_users)
# sort all users that have closed prs (unmerged)
closed_users_str = sort_by_val_return_str(closed_users)
# statistics for that repo
print 'num_merged,{}'.format(merged)
print 'merged_users,{}'.format(merged_users_str)
print 'num_open,{}'.format(open_pulls)
print 'num_closed,{}'.format(closed)
print 'closed_users,{}'.format(closed_users_str)
print ''
def print_prs(project_flag):
"""
Calculates the number of merged and closed pull requests, as well as the high
frequency pull requesters, and prints to stdout. Takes in a flag, defined by
constants.py, that determines whether to look at stack repos or ml repos.
"""
constants = return_constants(project_flag)
repos = constants["repos"]
directory = constants["pulls-fpath"]
for repo_dict in repos:
repo = repo_dict["name"]
owner = repo_dict["user"]
print repo, owner
# grab pull requests
pulls = json.load(open(os.path.join(directory, "{}_pulls.json".format(repo))))
print 'num_pulls,{}'.format(len(pulls))
# sort
merged = 0
closed = 0
open_pulls = 0
merged_users = {}
closed_users = {}
for p in pulls:
merged_at = p["merged_at"]
closed_at = p["closed_at"]
username = p["user"]["login"]
# make an exception for systemml
if closed_at != None and repo == systemml_name:
merged_at = 1 if is_merged_systemml(p) else None
# count
if merged_at != None:
merged += 1
merged_users[username] = merged_users.get(username, 0) + 1
elif closed_at != None:
closed += 1
closed_users[username] = closed_users.get(username, 0) + 1
else:
open_pulls += 1
# sort all users who have had merged prs
merged_users_str = sort_by_val_return_str(merged_users)
# sort all users that have closed prs (unmerged)
closed_users_str = sort_by_val_return_str(closed_users)
# statistics for that repo
print 'num_merged,{}'.format(merged)
print 'merged_users,{}'.format(merged_users_str)
print 'num_open,{}'.format(open_pulls)
print 'num_closed,{}'.format(closed)
print 'closed_users,{}'.format(closed_users_str)
print ''
def count(flag):
global constants_dict
constants_dict = return_constants(flag)
for repo in constants_dict["repos"]:
#print fn
# obtain a dictionary of commiters: # commits :
commit_dict = count_commits_per_user(return_filename(repo))
#print len(commit_dict.keys())
output_file = "committer_csvs/{}-{}-dict.csv".format(repo["name"], repo["user"])
with open(output_file, 'w') as csv_file:
writer = csv.writer(csv_file)
print "\n {} - {} \n".format(repo["name"], repo["user"])
for k, v in commit_dict.items():
writer.writerow([k, v])
print k + "," + str(v)
def get_pull_requests(project=ML):
constants = return_constants(project)
repo_list = constants["repos"]
output_dir = constants["pulls-fpath"]
# record the date of data retrieval
date_file = open(os.path.join(output_dir, "date.txt"), 'w')
date_file.write(CURRENT_DATE.strftime("%b %d, %Y"))
date_file.close()
for repo in repo_list:
pull_req = get_pulls(repo)
print "writing data..."
f = open(os.path.join(output_dir, "{}_pulls.json".format(repo["name"])), "w")
json.dump(pull_req, f)
f.close()
print "done writing data."
def get_pull_requests(project=ML):
constants = return_constants(project)
repo_list = constants["repos"]
output_dir = constants["pulls-fpath"]
# record the date of data retrieval
date_file = open(os.path.join(output_dir, "date.txt"), 'w')
date_file.write(CURRENT_DATE.strftime("%b %d, %Y"))
date_file.close()
for repo in repo_list:
pull_req = get_pulls(repo)
print "writing data..."
f = open(
os.path.join(output_dir, "{}_pulls.json".format(repo["name"])),
"w")
json.dump(pull_req, f)
f.close()
print "done writing data."
def herf(flag):
global constants_dict
constants_dict = return_constants(flag)
import doctest
doctest.testmod()
print 'Overall Herfindahl Indices'
for item in sorted(calculateOverall().items()):
print "{:<30}\t{}".format(item[0], item[1])
print '\nPer Year'
hiSortedByYear = sorted(calculateYear().items())
index = 0
sortedYears = sorted(hiSortedByYear[0][1].keys())
print "{:<12}".format("Year"),
for item in hiSortedByYear:
print "{:<30}".format(item[0]),
while(index < len(sortedYears)):
currYear = sortedYears[index]
print "\n{:<12}".format(currYear),
for item in hiSortedByYear:
print "{:<30}".format(item[1][currYear]),
index+=1
print ''
print '\nPer Month'
hiSortedByMonth = sorted(calculateMonth().items())
index = 0
sortedMonths = sorted(hiSortedByMonth[0][1].keys())
print "{:<12}".format("Month"),
for item in hiSortedByMonth:
print "{:<30}".format(item[0]),
while(index < len(sortedMonths)):
currMonth = sortedMonths[index]
print "\n{} {:<7}".format(currMonth/100, currMonth%100),
for item in hiSortedByMonth:
print "{:<30}".format(item[1][currMonth]),
index+=1
def herf(flag):
global constants_dict
constants_dict = return_constants(flag)
import doctest
doctest.testmod()
print 'Overall Herfindahl Indices'
for item in sorted(calculateOverall().items()):
print "{:<30}\t{}".format(item[0], item[1])
print '\nPer Year'
hiSortedByYear = sorted(calculateYear().items())
index = 0
sortedYears = sorted(hiSortedByYear[0][1].keys())
print "{:<12}".format("Year"),
for item in hiSortedByYear:
print "{:<30}".format(item[0]),
while (index < len(sortedYears)):
currYear = sortedYears[index]
print "\n{:<12}".format(currYear),
for item in hiSortedByYear:
print "{:<30}".format(item[1][currYear]),
index += 1
print ''
print '\nPer Month'
hiSortedByMonth = sorted(calculateMonth().items())
index = 0
sortedMonths = sorted(hiSortedByMonth[0][1].keys())
print "{:<12}".format("Month"),
for item in hiSortedByMonth:
print "{:<30}".format(item[0]),
while (index < len(sortedMonths)):
currMonth = sortedMonths[index]
print "\n{} {:<7}".format(currMonth / 100, currMonth % 100),
for item in hiSortedByMonth:
print "{:<30}".format(item[1][currMonth]),
index += 1
def get_commits(projects=ML):
constants = return_constants(projects) #change this if necessary
#directory
dirName = constants["commits-fpath"]
#getting correct repos
repo_list = constants['repos']
# record the date of data retrieval
date_file = open(os.path.join(dirName, "date.txt"), 'w')
date_file.write(CURRENT_DATE.strftime("%b %d, %Y"))
date_file.close()
#getting commits
for repo_data in repo_list:
commits = get_all_commits(repo_data)
print "writing data..."
output_file = open(os.path.join(dirName, return_filename(repo_data)), 'w')
json.dump(commits, output_file)
output_file.close()
print "done writing data."
def get_commits(projects=ML):
constants = return_constants(projects) #change this if necessary
#directory
dirName = constants["commits-fpath"]
#getting correct repos
repo_list = constants['repos']
# record the date of data retrieval
date_file = open(os.path.join(dirName, "date.txt"), 'w')
date_file.write(CURRENT_DATE.strftime("%b %d, %Y"))
date_file.close()
#getting commits
for repo_data in repo_list:
commits = get_all_commits(repo_data)
print "writing data..."
output_file = open(os.path.join(dirName, return_filename(repo_data)),
'w')
json.dump(commits, output_file)
output_file.close()
print "done writing data."
Currently not going to extend to stack since this code is not in use.
"""
import repo_info
import json
import sys, os
sys.path.insert(0, os.path.abspath(os.path.join('..')))
from Config.constants import ML, STACK, return_constants
dir = 'mlcontrib/'
# names = open(dir+"files.txt", 'r').read()
# mlFileNames = names.split()
# print mlFileNames
constants = return_constants(ML)
reponames = constants["repos"]
#matrix of projects (rows) and contributors (columns)
matrix = []
#collaborator name : index in matrix
allContributors = {}
#(project name, projectOwner) : index in matrix
projects = {}
#parentrepo : [list of repos forked from parent]
forkFamilyTree = {}
###