def process_section(self, section_name, b_verbose=False):
# this sections' repositories are under this folder
section_folder = os.path.abspath(self.config['folders'][section_name])
# get conflict list of the section
self.conflict_list = collect_conflicts(section_folder)
# filter blank
write_this = [conflict for conflict in self.conflict_list if conflict]
# no point of writing if no conflict
if write_this:
with open(section_name + '_' +
self.config['operation']['conflict_output_file'],
'wt',
encoding='utf-8') as f_out:
for conflict in write_this:
f_out.write(pprint.pformat(conflict) + '\n')
# return value
result = self.conflict_list
# reset conflict storage
self.conflict_list = unique_list.unique_list()
return result
def __init__(self, config_filename=False):
# prepare configuration
self.config = configparser.ConfigParser()
if config_filename:
config_filename = config_filename
else:
config_filename = 'reference.cfg'
if os.path.exists(config_filename):
self.config.read(config_filename)
else:
raise IOError("Can't find config file %s" % config_filename)
# folder
self.folder = os.path.abspath(self.config['operation']['folder'])
print(self.folder)
# comments
self.comments = unique_list.unique_list()
# initialize from the file
if os.path.exists(self.config['operation']['comment_output_file']):
with open(self.config['operation']['comment_output_file'], 'rt', encoding='utf-8') as f_in:
start_list = f_in.readlines()
for item in start_list:
self.comments.add(item.strip())
def __init__(self, config_filename=False):
# prepare configuration
self.config = configparser.ConfigParser()
if config_filename:
config_filename = config_filename
else:
config_filename = 'detect_conflict.cfg'
if os.path.exists(config_filename):
self.config.read(config_filename)
else:
raise IOError("Can't find config file %s" % config_filename)
# to store conflict info
self.conflict_list = unique_list.unique_list()
def __init__(self):
super(RepoEvalPoundByteCounterExcludingRef, self).__init__()
if os.path.exists(self.reference_cfg_filename):
self.comments_ref = unique_list.unique_list()
self.config_ref = configparser.ConfigParser()
self.config_ref.read(self.reference_cfg_filename)
reference_comment_filename = self.config_ref['operation']['comment_output_file']
with open(reference_comment_filename, 'r', encoding='utf-8') as f_in:
for line in f_in:
self.comments_ref.add(line.strip())
else:
self.init_reference_cfg_file()
def collect_comments_recursively(path, comments=unique_list.unique_list(), b_verbose=False):
for root_path, _, file_names in os.walk(path):
if not ignore.is_ignore_path(root_path):
for file_name in file_names:
if ignore.is_python(file_name):
full_path = os.path.join(root_path, file_name)
for line in read_python.get_comments_list_from_filename(full_path):
comments.add(line)
if b_verbose:
print('len(comments) =', len(comments))
else:
if b_verbose:
print('ignore file %s' % file_name)
else:
if b_verbose:
print('ignore path %s' % root_path)
return comments
def __init__(self):
# header row
self.column = unique_list.unique_list()
# table body
self.index = {}
def convert_git_log_to_table(self, git_log):
# column titles == unique file names in the git log
column_set = unique_list.unique_list()
commits_list = []
eval_dict = {}
# one big git log -> list of commits
git_log_split_blocks = git_log.split(self.commit_split_token)
# remove empty string
# TODO : consider git_log.strip(self.commit_split_token).split(self.commit_split_token)
# to avoid a special case
if not git_log_split_blocks[0]:
del git_log_split_blocks[0]
temporary_file_list = unique_list.unique_list(['no commit yet'])
last_commit_dict = {'subject': 'no commit yet'}
# commit loop
for git_log_block in git_log_split_blocks:
# TODO : refactor into a method?
# one commit example:
# < commit information >
# <int add> <int del> <filename 0>
# <int add> <int del> <filename 1>
# <int add> <int del> <filename 2>
# <int add> <int del> <filename 3>
# <blank line>
git_log_lines = git_log_block.splitlines()
# using git log output as input to python
last_commit_dict = self.get_commit_dict(git_log_lines[0])
if isinstance(last_commit_dict, (str, bytes)):
# if double quote is wrapping the output from the git log
last_commit_dict = self.get_commit_dict(
git_log_lines[0].strip('"'))
if isinstance(last_commit_dict, (str, bytes)):
# if still a str or bytes, try again
last_commit_dict = self.get_commit_dict(last_commit_dict)
assert isinstance(last_commit_dict, dict), f"git_log_lines[0] = {repr(git_log_lines[0])}\n" \
f"convert_git_log_to_table(): last_commit_dict = {last_commit_dict}" \
f"convert_git_log_to_table(): isinstance(last_commit_dict, dict) = {isinstance(last_commit_dict, dict)}" \
f"convert_git_log_to_table(): type(last_commit_dict) = {type(last_commit_dict)}"
# filer using email address
# TODO : consider refactoring into a function
if last_commit_dict['email'] in self.exclude_email_tuple:
last_commit_dict = {}
continue
# keep note of commits
commits_list.append(last_commit_dict)
# reset storage for 'files in commit'
temporary_file_list = unique_list.unique_list()
line = 'could be a merge commit'
# process file list
for line in git_log_lines[1:]:
# lines below the new commit
line_strip = line.strip()
if line_strip:
# a file in the last_commit_dict
# get path/filename
add__delete__key = line_strip.split()
column_key = add__delete__key[2]
path, filename = os.path.split(column_key)
# check ignore list
if not (self.is_ignore_path(path) or self.is_ignore_filename(filename)):
# if n files in one commit, evaluation would be 1/n
# TODO : simplify data structures
# for header of the table
column_set.add(column_key)
# to count the number of
temporary_file_list.add(column_key)
# end of files list
if temporary_file_list:
point = fractions.Fraction(1, len(temporary_file_list))
# build commit count table
# loop over files in commit
for files_in_commit in temporary_file_list:
eval_dict[files_in_commit] = eval_dict.get(
files_in_commit, 0) + point
# if it is still not a dict
if isinstance(last_commit_dict, str):
# try to convert it again
print(f'last_commit_dict = {repr(last_commit_dict)}')
last_commit_dict = ast.literal_eval(last_commit_dict)
if not isinstance(last_commit_dict, dict):
raise ValueError(
f'last_commit_dict = {repr(last_commit_dict)}')
last_commit_dict['files'] = temporary_file_list
else:
print(
f'\n{self.get_class_name()}.convert_git_log_to_table() : end of file list but temporary_file_list empty')
print(f'line = {line}')
print('last_commit_dict =', repr(last_commit_dict))
# total number of commits
# to make it the first element
# Sorting feature may use this feature
column_key = ' total'
column_set.add(column_key)
eval_dict[column_key] = len(commits_list)
return column_set, eval_dict