def code_sim_wrap(g, open_issue_url, close_issue_url):
# use jplag offline, due to moss upload limit
open_iss_ob = util.get_issue(g, open_issue_url)
close_iss_ob = util.get_issue(g, close_issue_url)
code_sim = issue_based_compare(g, open_iss_ob, close_iss_ob, mode='jplag')
logger.debug(f"Code-sim: {code_sim}")
return code_sim
def test_diff(self):
from persontoken import MY_TOKEN
g = Github(MY_TOKEN)
issue = util.get_issue(g, 'https://github.com/owlcs/owlapi/issues/936')
result = last_commits(g, issue)
print(result)
def main():
config = configparser.ConfigParser()
config.read('jira_search_replace.ini')
config.read('auth.ini')
issues = get_issues(config)
for issue in issues:
issue = util.get_issue(config['instance'], issue['key'])
do_search_replace(issue, config)
print('UPDATE COUNT:', update_count)
def _build_issue_dict(g, issue_html_url) -> dict:
issue_ob = util.get_issue(g, issue_html_url)
issue_dict = dict()
issue_dict['html_url'] = issue_ob.html_url
issue_dict['title'] = issue_ob.title
if issue_ob.body is None:
issue_dict['body'] = ""
else:
issue_dict['body'] = issue_ob.body
issue_dict['id'] = issue_ob.id
issue_dict['number'] = issue_ob.number
issue_dict['state'] = issue_ob.state
issue_dict['comments'] = issue_ob.comments # reply count
issue_dict['body_len'] = nlp_util.word_count(issue_dict['body'])
issue_dict['fix'] = contain_fix(g, issue_ob)
iss_body_tokens = nlp_util.tokenize(issue_dict['body'])
iss_body_tokens = nlp_util.remove_meanless(iss_body_tokens)
issue_dict['body_tokens'] = iss_body_tokens
title_tokens = nlp_util.nlp_process(issue_dict['title'])
body_list_tokens = nlp_util.nlp_process(issue_dict['body'])
hot_k = nlp_util.get_hot_keys()
hit_count_hot_title = search_rank.get_key_sea_count(hot_k, title_tokens, unique=False)
issue_dict['hit_count_hot_title'] = hit_count_hot_title
hit_count_hot_body = search_rank.get_key_sea_count(hot_k, body_list_tokens, unique=False)
issue_dict['hit_count_hot_body'] = hit_count_hot_body
labels = []
for _it in issue_ob.labels:
name = _it.name if _it.name else ""
description = _it.description if _it.description else ""
labels.append((name.lower(), description.lower()))
issue_dict['labels'] = labels
con_label = nlp_util.get_concern_label()
labels_c = []
for _it in issue_dict['labels']:
name, descrip = _it
for _c in con_label:
if _c in name or _c in descrip:
labels_c.append(_it)
issue_dict['labels_c'] = labels_c
return issue_dict
def test_once(self):
from persontoken import MY_TOKEN
g = Github(MY_TOKEN)
# issob = util.get_issue(g, 'https://github.com/json-path/JsonPath/issues/549')
# issob = util.get_issue(g, 'https://github.com/deathmarine/Luyten/issues/253')
issob = util.get_issue(g, 'https://github.com/arthur-star/test/issues/2')
curr_q = fast_query(issob.title, issob.body)
try_pair = [
(True, False, 'body'), # stacktrace in body
(False, True, 'title'), # condition in title
(False, False, 'title'), # title in title
(False, False, 'other') # title (no field constraint)
]
for _fi, pair in enumerate(try_pair):
trace, condition, pos = pair
query_list = form_query(curr_q, None, trace=trace, condition=condition)
query_chars = " ".join(query_list)
query_chars = add_pos(query_chars, pos)
logger.debug(f"query_chars, {query_chars}")
def mian():
from persontoken import MY_TOKEN
g = Github(MY_TOKEN)
done_open_urls = util.read_csv('./hist.txt')
done_open_urls = util.get_col(done_open_urls, 0)
_f = open("data2.csv", 'a+', encoding='utf-8-sig', newline='')
_f2 = open("hist.txt", 'a+', encoding='utf-8', newline='')
try:
csvwriter = csv.writer(_f, delimiter=',')
for i, open_url in enumerate(open_urls):
if open_url in done_open_urls:
continue
this_row = [""] * 5
# open_url = 'https://github.com/json-path/JsonPath/issues/460'
logger.info("-" * 100)
logger.info(open_url)
open_iss_ob = util.get_issue(g, open_url)
this_row[0] = open_url
# stacktrace / condition / title
repo_name = open_iss_ob.repository.full_name
extra_rm = nlp_util.full_name_token(
repo_name) # remove the number and its name
curr_q = fast_query(open_iss_ob.title, open_iss_ob.body)
logger.info(f"curr_q, {curr_q}")
# check sequence stacktrace, condition, title
try_pair = [
(True, False, 'body'), # stacktrace in body
(False, True, 'title'), # condition in title
(False, False, 'title'), # title in title
(False, False, 'other') # title (no field constraint)
]
try_hist = []
all_fail = True
for _fi, pair in enumerate(try_pair):
trace, condition, pos = pair
query_list = form_query(curr_q,
extra_rm,
trace=trace,
condition=condition)
query_chars = " ".join(query_list)
query_chars = add_pos(query_chars, pos)
logger.debug(f"query_chars, {query_chars}")
if query_list:
close_iss = run_close_query(g,
query_chars,
is_android,
depth=10,
fallback_size=5)
try_hist.append(query_chars)
this_row[1] = query_chars
if close_iss["info"] == 'NORMAL':
all_fail = False
else:
if close_iss["info"] == 'FALLBACK':
all_fail = False
logger.info(
f"[try {_fi}] FALLBACK failed query [Too few results], {query_chars}"
)
elif close_iss["info"] == 'EMPTY':
logger.info(
f"[try {_fi}] FALLBACK failed query [Zero results], {query_chars}"
)
if close_iss["iss"]:
# open url, open info online, open info offline, close url, close info (rank property)
# 0 , 1 , 2 , 3 , 4
rank_list = []
for _c in close_iss["iss"]:
close_url, close_info = _c
if is_android:
plugin = android_plugin
else:
plugin = java_plugin
# open url, open info online, open info offline, close url, close info (rank property)
# 0 , 1 , 2 , 3 , 4
open_off = plugin(g, open_url)
close_off = plugin(g, close_url)
logger.debug(f"open {open_url}")
logger.debug(
f"open offline rank, {open_off}=={len(open_off.keys())}"
)
logger.debug(f"close {close_url}")
logger.debug(
f"close offline rank, {close_off}=={len(close_off.keys())}"
)
assert len(open_off.keys()) == len(
close_off.keys())
all_empty = True
join_off = dict()
for _k in open_off.keys():
join = set(open_off[_k]) & set(close_off[_k])
join_off[_k] = list(join)
if join:
all_empty = False
close_info.insert(0, f"Off-SIM-{_k}")
logger.debug(f"join_off, {join_off}")
if all_empty:
this_row[2] = "empty offline"
else:
this_row[2] = json.dumps(join_off)
this_row[3] = close_url
flag = code_sim_wrap(g, open_url, close_url)
if flag:
close_info.insert(0, "Code-SIM")
this_row[4] = json.dumps(close_info)
rank_list.append(deepcopy(this_row))
rank_list = rank_issue(rank_list)
if rank_list:
csvwriter.writerows(rank_list)
if close_iss["info"] == 'NORMAL':
break
if all_fail:
write_list = []
this_row[4] = 'NONE close issue'
for col1 in try_hist:
this_row[1] = col1
write_list.append(deepcopy(this_row))
csvwriter.writerows(write_list)
print(open_url, file=_f2)
_f.flush()
_f2.flush()
except Exception as e:
logger.error(f"{open_url}, skip")
print(f"{open_url}, skip", file=_f2)
raise e
finally:
_f.close()
_f2.close()
def test_last_commits(self):
from persontoken import MY_TOKEN
g = Github(MY_TOKEN)
issue = util.get_issue(g, 'https://github.com/confluentinc/ksql/issues/5062')
result = last_commits(g, issue)
print(result)
if __name__ == '__main__':
from persontoken import MY_TOKEN
g = Github(MY_TOKEN)
# extract_bugfix("")
# results = run_open_query(g, "crash", "crash.csv")
# results = run_open_query(g, "permission Android")
# print(run_close_query(g, "google action button page", depth=20))
# https://api.github.com/repos/arquivo/pwa-technologies/issues/931
# https://api.github.com/repos/wordpress-mobile/WordPress-Android/issues/9685
# https://api.github.com/repos/tlaplus/tlaplus/issues/477
ss = util.SS(port=7890)
# repo = g.get_repo("tlaplus/tlaplus")
# print(repo.get_topics())
# print(get_topic("tlaplus/tlaplus"))
#
# print(is_android_repo(g, 'nextcloud/android'))
# print(is_android_repo(g, 'arduino/Arduino'))
import util
issue_ob = util.get_issue(g, 'https://github.com/apache/dubbo/issues/6489')
print(contain_fix(g, issue_ob))
# a = 'org.springframework.beans.factory.BeanCreationException'
# a = "a" * 50 + "....." + "ac" * 30
# a = "-" * 77
# print(cut_text(a, 54))