示例#1
0
    def _find_commit(self, commit_id: str) -> Commit:
        logging.debug(f'running query for commit: [{commit_id}]')
        query = self._commit_query(commit_id)
        response_json = self.graph_ql_client.execute_query(query)
        logging.debug(f'query complete for commit: [{commit_id}]')

        node = response_json["data"]["node"]
        commit = Commit(node["id"])
        commit.repository_id = self.repository.id
        if node["author"] is not None:
            if node["author"]["user"] is not None:
                commit.author = self._find_author_by_id(
                    node["author"]["user"]["id"])
        commit.authored_by_committer = node["authoredByCommitter"]
        commit.total_authors = node["authors"]["totalCount"]
        if node["committer"] is not None:
            if node["committer"]["user"] is not None:
                commit.committer = self._find_author_by_id(
                    node["committer"]["user"]["id"])
        if node["onBehalfOf"] is not None:
            commit.for_organization_id = node["onBehalfOf"]["id"]
        commit.create_datetime = base.to_datetime_from_str(
            node["committedDate"])
        if node["pushedDate"] is not None:
            commit.push_datetime = base.to_datetime_from_str(
                node["pushedDate"])
        commit.total_comments = node["comments"]["totalCount"]
        commit.total_associated_pull_requests = \
            node["associatedPullRequests"]["totalCount"]
        commit.total_check_suites = node["checkSuites"]["totalCount"]
        commit.message_headline = node["messageHeadline"]
        commit.message_body = node["messageBody"]
        commit.additions = node["additions"]
        commit.deletions = node["deletions"]
        commit.changed_files = node["changedFiles"]

        # iterate over the commit tree
        commit.tree_id = node["tree"]["id"]
        for entry_node in node["tree"]["entries"]:
            entry = CommitEntry()
            entry.id = entry_node["object"]["id"]
            entry.name = entry_node["name"]
            entry.extension = entry_node["extension"]
            entry.path = entry_node["path"]
            entry.is_generated = entry_node["isGenerated"]
            entry.mode = entry_node["mode"]
            entry.type = entry_node["type"]
            if entry_node["submodule"] is not None:
                entry.submodule_name = entry_node["submodule"]["name"]
            commit.entries.append(entry)

        if node["status"] is not None:
            commit.state = node["status"]["state"]

        logging.debug(f'query complete for commit: [{commit_id}]')
        return commit
示例#2
0
    def run(self):
        """
        loads the comments for the pull request reviews for a specific repository
        :return: None
        """
        pull_request_reviews_reviewed: int = 0

        pull_request_review_count = self._get_objects_saved_count(
            self.repository, base.ObjectType.PULL_REQUEST_REVIEW)

        pull_request_reviews = self._get_collection().find({
            'repository_id':
            self.repository.id,
            'object_type':
            base.ObjectType.PULL_REQUEST_REVIEW.name
        })
        for review in pull_request_reviews:
            pull_request_review_id: str = review['id']
            pull_request_id: str = review['pull_request_id']
            comments_expected: int = review['total_comments']
            comments: [PullRequestReviewComment] = []
            comment_ids: [str] = []
            comment_cursor: str = None
            pull_request_reviews_reviewed += 1

            while comments_expected > self._get_actual_comments(
                    pull_request_review_id):
                logging.debug(
                    f'running query for comments for pull request review [{pull_request_review_id}] '
                    f'against {self.repository}')
                query = self._pull_request_review_comments_query(
                    pull_request_review_id, comment_cursor)
                response_json = self.graph_ql_client.execute_query(query)
                logging.debug(
                    f'query complete for comments for pull request review [{pull_request_review_id}] '
                    f'against {self.repository}')

                # iterate over each comment returned (we return 100 at a time)
                for edge in response_json["data"]["node"]["comments"]["edges"]:

                    comment_cursor = edge["cursor"]
                    comment = PullRequestReviewComment(edge["node"]["id"])
                    comments.append(comment)
                    comment_ids.append(comment.id)
                    comment.repository_id = self.repository.id
                    comment.pull_request_id = pull_request_id
                    comment.pull_request_review_id = pull_request_review_id
                    comment.state = edge["node"]["state"]

                    # get the body text
                    comment.body_text = edge["node"]["bodyText"]

                    # get the counts for the sub items to comment
                    comment.total_reactions = edge["node"]["reactions"][
                        "totalCount"]
                    comment.total_edits = edge["node"]["userContentEdits"][
                        "totalCount"]

                    # parse the datetime
                    comment.create_datetime = base.to_datetime_from_str(
                        edge["node"]["createdAt"])

                    # set the diffHunk
                    comment.diff_hunk = edge["node"]["diffHunk"]

                    # set the path
                    comment.path = edge["node"]["path"]

                    # set the original position
                    if edge["node"]["originalPosition"] is not None:
                        comment.original_position = edge["node"][
                            "originalPosition"]

                    # set the position
                    if edge["node"]["position"] is not None:
                        comment.position = edge["node"]["position"]

                    # set if the comment has been minimized
                    if edge["node"]["isMinimized"] is not None and edge[
                            "node"]["isMinimized"] is True:
                        comment.minimized_status = edge["node"][
                            "minimizedReason"]

                    # author can be None.  Who knew?
                    if edge["node"]["author"] is not None:
                        comment.author = self._find_author_by_login(
                            edge["node"]["author"]["login"])
                    comment.author_association = edge["node"][
                        'authorAssociation']

                    # get original commit id
                    if edge["node"]["originalCommit"] is not None:
                        comment.original_commit_id = edge["node"][
                            "originalCommit"]["id"]

                    # get commit id
                    if edge["node"]["commit"] is not None:
                        comment.commit_id = edge["node"]["commit"]["id"]

                    # get author of comment we are replying to
                    if edge["node"]["replyTo"] is not None:
                        comment.reply_to_comment_id = edge["node"]["replyTo"][
                            "id"]

                    # check to see if pull request comment is in the database
                    found_request = \
                        self._get_collection().find_one({'id': comment.id, 'object_type':
                                                        base.ObjectType.PULL_REQUEST_REVIEW_COMMENT.name})
                    if found_request is None:
                        self._get_collection().insert_one(
                            comment.to_dictionary())

            self._get_collection().update_one(
                {'id': pull_request_review_id},
                {'$set': {
                    'comment_ids': comment_ids
                }})

            logging.debug(
                f'pull request reviews reviewed for {self.repository} '
                f'{pull_request_reviews_reviewed}/{pull_request_review_count}')

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'comments returned for {self.repository} '
            f'returned: [{actual_count}], expected: [{expected_count}]')
    def run(self):
        """
        loads the reviews for the pull requests for a specific repository
        :return: None
        """
        pull_request_reviewed: int = 0

        pull_request_count = self._get_objects_saved_count(
            self.repository, base.ObjectType.PULL_REQUEST)

        pull_requests = self._get_collection().find({
            'repository_id':
            self.repository.id,
            'object_type':
            base.ObjectType.PULL_REQUEST.name
        })
        for pr in pull_requests:
            pull_request_id: str = pr['id']
            reviews_expected: int = pr['total_reviews']
            review_ids: [str] = []
            review_cursor: str = None
            pull_request_reviewed += 1

            while reviews_expected > self._get_actual_reviews(pull_request_id):
                logging.debug(
                    f'running query for reviews for pull request [{pull_request_id}] against {self.repository}'
                )
                query = self._pull_request_reviews_query(
                    pull_request_id, review_cursor)
                response_json = self.graph_ql_client.execute_query(query)
                logging.debug(
                    f'query complete for reviews for pull request [{pull_request_id}] against {self.repository}'
                )

                # iterate over each review returned (we return 100 at a time)
                for edge in response_json["data"]["node"]["reviews"]["edges"]:

                    review_cursor = edge["cursor"]
                    review = PullRequestReview(edge["node"]["id"],
                                               pull_request_id)
                    review_ids.append(review.id)
                    review.repository_id = self.repository.id
                    review.body_text = edge["node"]["bodyText"]
                    review.commit_id = edge["node"]["commit"]["id"]
                    review.total_comments = edge["node"]["comments"][
                        "totalCount"]
                    review.total_edits = edge["node"]["userContentEdits"][
                        "totalCount"]
                    review.total_reactions = edge["node"]["reactions"][
                        "totalCount"]
                    review.total_for_teams = edge["node"]["onBehalfOf"][
                        "totalCount"]
                    review.create_datetime = base.to_datetime_from_str(
                        edge["node"]["createdAt"])
                    review.state = edge["node"]["state"]

                    # author can be None.  Who knew?
                    if edge["node"]["author"] is not None:
                        review.author = self._find_author_by_login(
                            edge["node"]["author"]["login"])
                    review.author_association = edge["node"][
                        "authorAssociation"]

                    # check to see if pull request review is in the database
                    found_request = self._get_collection().find_one({
                        'id':
                        review.id,
                        'object_type':
                        'PULL_REQUEST_REVIEW'
                    })
                    if found_request is None:
                        self._get_collection().insert_one(
                            review.to_dictionary())

                self._get_collection().update_one(
                    {'id': pull_request_id},
                    {'$set': {
                        'review_ids': review_ids
                    }})

            logging.debug(
                f'pull requests reviewed for {self.repository} {pull_request_reviewed}/{pull_request_count}'
            )

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'reviews returned for {self.repository} '
            f'returned: [{actual_count}], expected: [{expected_count}]')
    def run(self):
        """
        loads the check suite ids from the commits for a specific repository
        :return: None
        """
        commits_reviewed: int = 0

        commit_count = self._get_objects_saved_count(self.repository,
                                                     base.ObjectType.COMMIT)

        commits = self._get_collection().find({
            'repository_id':
            self.repository.id,
            'object_type':
            base.ObjectType.COMMIT.name
        })
        for item in commits:
            commit_id: str = item['id']
            check_suites_expected: int = item['total_check_suites']
            check_suite_ids: [str] = []
            check_suite_cursor: str = None
            commits_reviewed += 1

            if check_suites_expected > len(item['check_suite_ids']):
                while check_suites_expected > len(check_suite_ids):
                    logging.debug(
                        f'running query for check suite ids for commit [{commit_id}] against {self.repository}'
                    )
                    query = self._commit_check_suite_query(
                        commit_id, check_suite_cursor)
                    response_json = self.graph_ql_client.execute_query(query)
                    logging.debug(
                        f'query complete for check suite ids for commit [{commit_id}] against {self.repository}'
                    )

                    # iterate over each check suite returned (we return 100 at a time)
                    for edge in response_json["data"]["node"]["checkSuites"][
                            "edges"]:
                        check_suite_cursor = edge["cursor"]
                        check_suite = CheckSuite(edge["node"]["id"])
                        check_suite.commit_id = edge["node"]["commit"]["id"]
                        check_suite.repository_id = edge["node"]["repository"][
                            "id"]
                        if edge["node"]["app"] is not None:
                            check_suite.application_id = edge["node"]["app"][
                                "id"]
                        if edge["node"]["branch"] is not None:
                            check_suite.branch_id = edge["node"]["branch"][
                                "id"]
                        check_suite.conclusion = edge["node"]["conclusion"]
                        # check_suite.push_id = edge["node"]["push"]["id"]
                        check_suite.state = edge["node"]["status"]

                        # load the counts
                        check_suite.total_check_runs = edge["node"][
                            "checkRuns"]["totalCount"]
                        check_suite.total_matching_pull_requests = edge[
                            "node"]["matchingPullRequests"]["totalCount"]

                        # parse the datetime
                        check_suite.create_datetime = base.to_datetime_from_str(
                            edge["node"]["createdAt"])

                        check_suite_ids.append(check_suite.id)

                        # check to see if pull request comment is in the database
                        found_request = \
                            self._get_collection().find_one({'id': check_suite.id,
                                                             'object_type': base.ObjectType.CHECK_SUITE.name})
                        if found_request is None:
                            self._get_collection().insert_one(
                                check_suite.to_dictionary())

                self._get_collection().update_one(
                    {'id': commit_id},
                    {'$set': {
                        'check_suite_ids': check_suite_ids
                    }})

            logging.debug(
                f'commits reviewed for {self.repository} {commits_reviewed}/{commit_count}'
            )

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'check suites returned for {self.repository} returned: [{actual_count}], expected: [{expected_count}]'
        )
示例#5
0
    def run(self):
        """
        loads the comments for the commits for a specific repository
        :return: None
        """
        commits_reviewed: int = 0

        commit_count = self._get_objects_saved_count(self.repository, base.ObjectType.COMMIT)

        commits = self._get_collection().find({'repository_id': self.repository.id,
                                               'object_type': base.ObjectType.COMMIT.name})
        for item in commits:
            commit_id: str = item['id']
            comments_expected: int = item['total_comments']
            comment_ids: [str] = []
            comment_cursor: str = None
            commits_reviewed += 1

            if comments_expected > len(item['comment_ids']):
                while comments_expected > len(comment_ids):
                    logging.debug(
                        f'running query for comments for commit [{commit_id}] against {self.repository}'
                    )
                    query = self._commit_comment_query(commit_id, comment_cursor)
                    response_json = self.graph_ql_client.execute_query(query)
                    logging.debug(
                        f'query complete for comments for commit [{commit_id}] against {self.repository}'
                    )

                    # iterate over each participant returned (we return 100 at a time)
                    for edge in response_json["data"]["node"]["comments"]["edges"]:
                        comment_cursor = edge["cursor"]
                        commit_comment = CommitComment(edge["node"]["id"])
                        comment_ids.append(commit_comment.id)
                        commit_comment.repository_id = self.repository.id
                        commit_comment.commit_id = commit_id

                        # get the counts for the sub items to comment
                        commit_comment.total_reactions = edge["node"]["reactions"]["totalCount"]
                        commit_comment.total_edits = edge["node"]["userContentEdits"]["totalCount"]

                        # get the body text
                        commit_comment.body_text = edge["node"]["bodyText"]

                        # parse the datetime
                        commit_comment.create_datetime = base.to_datetime_from_str(edge["node"]["createdAt"])

                        # set the path
                        commit_comment.path = edge["node"]["path"]

                        # author can be None.  Who knew?
                        if edge["node"]["author"] is not None:
                            commit_comment.author = self._find_author_by_login(edge["node"]["author"]["login"])
                        commit_comment.author_association = edge["node"]['authorAssociation']

                        # set the position
                        if edge["node"]["position"] is not None:
                            commit_comment.position = edge["node"]["position"]

                        # set if the comment has been minimized
                        if edge["node"]["isMinimized"] is not None and edge["node"]["isMinimized"] is True:
                            commit_comment.minimized_status = edge["node"]["minimizedReason"]

                        # check to see if pull request comment is in the database
                        found_request = \
                            self._get_collection().find_one({'id': commit_comment.id,
                                                             'object_type': base.ObjectType.COMMIT_COMMENT.name})
                        if found_request is None:
                            self._get_collection().insert_one(commit_comment.to_dictionary())

                self._get_collection().update_one({'id': commit_id},
                                                  {'$set': {'comment_ids': comment_ids}})

            logging.debug(f'pull requests reviewed for {self.repository} {commits_reviewed}/{commit_count}')

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'participants returned for {self.repository} returned: [{actual_count}], expected: [{expected_count}]'
        )
示例#6
0
    def run(self):
        """
        loads the pull request for a specific repository
        :return: None
        """
        if self._get_expected_results() != self._get_actual_results():
            pull_requests_loaded: int = 0
            pull_request_cursor: str = None

            # continue executing gets against git until we have all the PRs
            while self.repository.total_pull_requests > pull_requests_loaded:
                logging.debug(
                    f'running query for pull requests against {self.repository}'
                )
                query = self._pull_request_query(pull_request_cursor)
                response_json = self.graph_ql_client.execute_query(query)
                logging.debug(
                    f'query complete for pull requests against {self.repository}'
                )

                # iterate over each pull request returned (we return 20 at a time)
                for edge in response_json["data"]["repository"][
                        "pullRequests"]["edges"]:
                    pull_requests_loaded += 1
                    pull_request_cursor = edge["cursor"]
                    pull_request_id = edge["node"]["id"]

                    # check to see if pull request is in the database
                    found_request = self._get_collection().find_one({
                        'id':
                        pull_request_id,
                        'object_type':
                        base.ObjectType.PULL_REQUEST.name
                    })
                    if found_request is None:

                        pr = PullRequest(pull_request_id, self.repository.id)
                        pr.body_text = edge["node"]["bodyText"]
                        pr.state = edge["node"]["state"]

                        # load the counts
                        pr.total_reviews = edge["node"]["reviews"][
                            "totalCount"]
                        pr.total_comments = edge["node"]["comments"][
                            "totalCount"]
                        pr.total_participants = edge["node"]["participants"][
                            "totalCount"]
                        pr.total_edits = edge["node"]["userContentEdits"][
                            "totalCount"]
                        pr.total_reactions = edge["node"]["reactions"][
                            "totalCount"]
                        pr.total_commits = edge["node"]["commits"][
                            "totalCount"]

                        # author can be None.  Who knew?
                        if edge["node"]["author"] is not None:
                            pr.author = self._find_author_by_login(
                                edge["node"]["author"]["login"])
                        pr.author_association = pr.author_login = edge["node"][
                            "authorAssociation"]

                        # parse the datetime
                        pr.create_datetime = base.to_datetime_from_str(
                            edge["node"]["createdAt"])

                        logging.debug(f'inserting record for {pr}')
                        self._get_collection().insert_one(pr.to_dictionary())
                        logging.debug(f'insert complete for {pr}')
                    else:
                        logging.debug(
                            f'Pull Request [id: {pull_request_id}] already found in database'
                        )

                logging.debug(
                    f'pull requests found for {self.repository} '
                    f'{pull_requests_loaded}/{self.repository.total_pull_requests}'
                )

            actual_count: int = self._get_actual_results()
            logging.debug(
                f'pull requests returned for {self.repository} returned: [{actual_count}], '
                f'expected: [{self.repository.total_pull_requests}]')
示例#7
0
    def run(self):
        """
        loads the reactions for a document type for a specific repository
        :return: None
        """
        items_reviewed: int = 0

        item_count = self._get_objects_saved_count(self.repository,
                                                   self.object_type)

        items = self._get_collection().find({
            'repository_id':
            self.repository.id,
            'object_type':
            self.object_type.name
        })
        for item in items:
            item_id: str = item['id']
            reactions_expected: int = item['total_reactions']
            reactions: [Reaction] = []
            reaction_cursor: str = None
            items_reviewed += 1

            if reactions_expected > len(item['reactions']):
                while reactions_expected > len(reactions):
                    logging.debug(
                        f'running query for reactions for {self.object_type.name} [{item_id}] against {self.repository}'
                    )
                    query = self._reactions_query(item_id, reaction_cursor)
                    response_json = self.graph_ql_client.execute_query(query)
                    logging.debug(
                        f'query complete for reactions for {self.object_type.name} [{item_id}] '
                        f'against {self.repository}')

                    # iterate over each reaction returned (we return 100 at a time)
                    for edge in response_json["data"]["node"]["reactions"][
                            "edges"]:
                        reaction_cursor = edge["cursor"]
                        reaction = Reaction(edge["node"]["id"])
                        reaction.content = edge["node"]["content"]
                        reaction.create_datetime = base.to_datetime_from_str(
                            edge["node"]["createdAt"])

                        # author can be None.  Who knew?
                        if edge["node"]["user"] is not None:
                            reaction.author = self._find_author_by_id(
                                edge["node"]["user"]["id"])
                        reactions.append(reaction)

                reaction_dictionaries = list(map(base.to_dictionary,
                                                 reactions))
                self._get_collection().update_one(
                    {'id': item_id},
                    {'$set': {
                        'reactions': reaction_dictionaries
                    }})

            logging.debug(
                f'{self.object_type.name} reviewed for {self.repository} {items_reviewed}/{item_count}'
            )

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'reactions returned for {self.repository} returned: [{actual_count}], expected: [{expected_count}]'
        )
示例#8
0
    def run(self):
        """
        loads the edits for a specific object type for a specific repository
        :return: None
        """
        item_reviewed: int = 0

        item_count = self._get_objects_saved_count(self.repository,
                                                   self.object_type)

        items = self._get_collection().find({
            'repository_id':
            self.repository.id,
            'object_type':
            self.object_type.name
        })
        for item in items:
            item_reviewed += 1
            item_id: str = item['id']
            edits_expected: int = item['total_edits']
            edits: [ContentEdit] = []
            edit_cursor: str = None
            has_delete: bool = False

            if edits_expected > len(item['edits']):
                while edits_expected > len(edits):
                    logging.debug(
                        f'running query for edits for {self.object_type.name} [{item_id}] '
                        f'against {self.repository}')
                    query = self._edits_query(item_id, edit_cursor)
                    response_json = self.graph_ql_client.execute_query(query)
                    logging.debug(
                        f'query complete for edits for {self.object_type.name} [{item_id}] '
                        f'against {self.repository}')

                    # iterate over each edit returned (we return 100 at a time)
                    for edge in response_json["data"]["node"][
                            "userContentEdits"]["edges"]:
                        edit_cursor = edge["cursor"]
                        edit = ContentEdit(edge["node"]["id"])
                        edits.append(edit)
                        edit.edit_datetime = base.to_datetime_from_str(
                            edge["node"]["editedAt"])
                        if edge["node"]["editor"] is not None:
                            edit.editor = self._find_author_by_login(
                                edge["node"]["editor"]["login"])
                        # checking to see if this is an edit
                        if edge["node"]["diff"] is not None:
                            edit.difference = edge["node"]["diff"]

                        if edge["node"]["deletedAt"] is not None:
                            if edge["node"]["deletedBy"] is not None:
                                edit.editor = self._find_author_by_login(
                                    edge["node"]["deletedBy"]["login"])
                            edit.edit_datetime = base.to_datetime_from_str(
                                edge["node"]["deletedAt"])
                            edit.is_delete = True
                            has_delete = True

                edit_dictionaries = list(map(base.to_dictionary, edits))
                self._get_collection().update_one({'id': item_id}, {
                    '$set': {
                        'edits': edit_dictionaries,
                        'is_deleted': has_delete
                    }
                })
            logging.debug(
                f'{self.object_type.name} reviewed for {self.repository} {item_reviewed}/{item_count}'
            )

        actual_count: int = self._get_actual_results()
        expected_count: int = self._get_expected_results()
        logging.debug(
            f'edits returned for {self.repository} returned: [{actual_count}], expected: [{expected_count}]'
        )