# -*- coding: utf-8 -*- """Convert IDE TODOs to GitHub issues.""" import os import re from time import sleep from io import StringIO import itertools import operator from collections import defaultdict import sys from Client import Client from GitHubClient import GitHubClient from LineStatus import LineStatus from LocalClient import LocalClient from TodoParser import TodoParser def process_diff(diff, client=Client(), insert_issue_urls=False, parser=TodoParser(), output=sys.stdout): # Parse the diff for TODOs and create an Issue object for each. raw_issues = parser.parse(diff) # This is a simple, non-perfect check to filter out any TODOs that have just been moved. # It looks for items that appear in the diff as both an addition and deletion. # It is based on the assumption that TODOs will not have identical titles in identical files. # That is about as good as we can do for TODOs without issue URLs. issues_to_process = [] for values, similar_issues_iter in itertools.groupby(raw_issues, key=operator.attrgetter('title', 'file_name', 'markdown_language')): similar_issues = list(similar_issues_iter) if (len(similar_issues) == 2 and all(issue.issue_url is None for issue in similar_issues) and ((similar_issues[0].status == LineStatus.ADDED and similar_issues[1].status == LineStatus.DELETED) or (similar_issues[1].status == LineStatus.ADDED and similar_issues[0].status == LineStatus.DELETED))): print(f'Issue "{values[0]}" appears as both addition and deletion. ' f'Assuming this issue has been moved so skipping.', file=output) continue issues_to_process.extend(similar_issues) # If a TODO with an issue URL is updated, it may appear as both an addition and a deletion. # We need to ignore the deletion so it doesn't update then immediately close the issue. # First store TODOs based on their status. todos_status = defaultdict(lambda: {'added': False, 'deleted': False}) # Populate the status dictionary based on the issue URL. for raw_issue in issues_to_process: if raw_issue.issue_url: # Ensuring we're dealing with TODOs that have an issue URL. if raw_issue.status == LineStatus.ADDED: todos_status[raw_issue.issue_url]['added'] = True elif raw_issue.status == LineStatus.DELETED: todos_status[raw_issue.issue_url]['deleted'] = True # Determine which issues are both added and deleted. update_and_close_issues = set() for _issue_url, _status in todos_status.items(): if _status['added'] and _status['deleted']: update_and_close_issues.add(_issue_url) # Remove issues from issues_to_process if they are both to be updated and closed (i.e., ignore deletions). issues_to_process = [issue for issue in issues_to_process if not (issue.issue_url in update_and_close_issues and issue.status == LineStatus.DELETED)] # Cycle through the Issue objects and create or close a corresponding GitHub issue for each. for j, raw_issue in enumerate(sorted(reversed(sorted(issues_to_process, key = operator.attrgetter('start_line'))), key = operator.attrgetter('file_name'))): print(f"Processing issue {j + 1} of {len(issues_to_process)}: '{raw_issue.title}' @ {raw_issue.file_name}:{raw_issue.start_line}", file=output) if raw_issue.status == LineStatus.ADDED: status_code, new_issue_number = client.create_issue(raw_issue) if status_code == 201: print(f'Issue created: #{new_issue_number} @ {client.get_issue_url(new_issue_number)}', file=output) # Don't insert URLs for comments. Comments do not get updated. if insert_issue_urls and not (raw_issue.ref and raw_issue.ref.startswith('#')): line_number = raw_issue.start_line - 1 with open(raw_issue.file_name, 'r') as issue_file: file_lines = issue_file.readlines() # Get style of newlines used in this file, so that we # use the same type when writing the file back out. # Note: # - if only one newline type is detected, then # 'newlines' will be a string with that value # - if no newlines are detected, 'newlines' will # be 'None' and the platform-dependent default # will be used when terminating lines on write # - if multiple newline types are detected (e.g. # a mix of Windows- and Unix-style newlines in # the same file), then that is handled within # the following if block... newline_style = issue_file.newlines if isinstance(issue_file.newlines, tuple): # A tuple being returned indicates that a mix of # line ending styles was found in the file. In # order to not perturb the file any more than # intended (i.e. inserting the issue URL comment(s)) # we'll reread the file and keep the line endings. # On write, we'll tell writelines to not introduce # any explicit line endings. This modification # of the read and write behavior is handled by # passing '' to the newline argument of open(). # Note: the line ending of the issue URLs line(s) # itself will be that of the TODO line above it # and is handled later in this function. newline_style = '' # reread the file without stripping off line endings with open(raw_issue.file_name, 'r', newline=newline_style) as issue_file_reread: file_lines = issue_file_reread.readlines() else: newline_style = issue_file.newlines if line_number < len(file_lines): # Duplicate the line to retain the comment syntax. old_line = file_lines[line_number] remove = fr'(?i:{re.escape(raw_issue.identifier)}).*{re.escape(raw_issue.title)}.*?(\r|\r\n|\n)?$' insert = f'Issue URL: {client.get_issue_url(new_issue_number)}' # note that the '\1' capture group is the line ending character sequence and # will only be non-empty in the case of a mixed line-endings file new_line = re.sub('^.*' + remove, fr'{raw_issue.prefix + insert + raw_issue.suffix}\1', old_line) # make sure the above operation worked as intended if new_line != old_line: # Check if the URL line already exists, if so abort. if line_number == len(file_lines) - 1 or file_lines[line_number + 1] != new_line: file_lines.insert(line_number + 1, new_line) with open(raw_issue.file_name, 'w', newline=newline_style) as issue_file: issue_file.writelines(file_lines) print('Issue URL successfully inserted', file=output) else: print('ERROR: Issue URL was NOT successfully inserted', file=output) elif status_code == 200: print(f'Issue updated: #{new_issue_number} @ {client.get_issue_url(new_issue_number)}', file=output) else: print('Issue could not be created', file=output) elif raw_issue.status == LineStatus.DELETED and os.getenv('INPUT_CLOSE_ISSUES', 'true') == 'true': if raw_issue.ref and raw_issue.ref.startswith('#'): print('Issue looks like a comment, will not attempt to close.', file=output) continue status_code = client.close_issue(raw_issue) if status_code in [200, 201]: print('Issue closed', file=output) else: print('Issue could not be closed', file=output) # Stagger the requests to be on the safe side. sleep(1) return raw_issues if __name__ == "__main__": client: Client | None = None # Try to create a basic client for communicating with the remote version control server, automatically initialised with environment variables. try: # try to build a GitHub client client = GitHubClient() except EnvironmentError: # don't immediately give up pass # if needed, fall back to using a local client for testing client = client or LocalClient() # Get the diff from the last pushed commit. last_diff = client.get_last_diff() # process the diff if last_diff: # Check to see if we should insert the issue URL back into the linked TODO. insert_issue_urls = os.getenv('INPUT_INSERT_ISSUE_URLS', 'false') == 'true' process_diff(StringIO(last_diff), client, insert_issue_urls)