poky/scripts/oe-git-archive

#!/usr/bin/env python3
#
# Helper script for committing data to git and pushing upstream
#
# Copyright (c) 2017, Intel Corporation.
#
# This program is free software; you can redistribute it and/or modify it
# under the terms and conditions of the GNU General Public License,
# version 2, as published by the Free Software Foundation.
#
# This program is distributed in the hope it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
# FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
# more details.
#
import argparse
import glob
import json
import logging
import math
import os
import re
import sys
from collections import namedtuple, OrderedDict
from datetime import datetime, timedelta, tzinfo
from operator import attrgetter

# Import oe and bitbake libs
scripts_path = os.path.dirname(os.path.realpath(__file__))
sys.path.append(os.path.join(scripts_path, 'lib'))
import scriptpath
scriptpath.add_bitbake_lib_path()
scriptpath.add_oe_lib_path()

from oeqa.utils.git import GitRepo, GitError
from oeqa.utils.metadata import metadata_from_bb


# Setup logging
logging.basicConfig(level=logging.INFO, format="%(levelname)s: %(message)s")
log = logging.getLogger()


class ArchiveError(Exception):
    """Internal error handling of this script"""


def format_str(string, fields):
    """Format string using the given fields (dict)"""
    try:
        return string.format(**fields)
    except KeyError as err:
        raise ArchiveError("Unable to expand string '{}': unknown field {} "
                           "(valid fields are: {})".format(
                               string, err, ', '.join(sorted(fields.keys()))))


def init_git_repo(path, no_create, bare):
    """Initialize local Git repository"""
    path = os.path.abspath(path)
    if os.path.isfile(path):
        raise ArchiveError("Invalid Git repo at {}: path exists but is not a "
                           "directory".format(path))
    if not os.path.isdir(path) or not os.listdir(path):
        if no_create:
            raise ArchiveError("No git repo at {}, refusing to create "
                               "one".format(path))
        if not os.path.isdir(path):
            try:
                os.mkdir(path)
            except (FileNotFoundError, PermissionError) as err:
                raise ArchiveError("Failed to mkdir {}: {}".format(path, err))
        if not os.listdir(path):
            log.info("Initializing a new Git repo at %s", path)
            repo = GitRepo.init(path, bare)
    try:
        repo = GitRepo(path, is_topdir=True)
    except GitError:
        raise ArchiveError("Non-empty directory that is not a Git repository "
                           "at {}\nPlease specify an existing Git repository, "
                           "an empty directory or a non-existing directory "
                           "path.".format(path))
    return repo


def git_commit_data(repo, data_dir, branch, message, exclude, notes):
    """Commit data into a Git repository"""
    log.info("Committing data into to branch %s", branch)
    tmp_index = os.path.join(repo.git_dir, 'index.oe-git-archive')
    try:
        # Create new tree object from the data
        env_update = {'GIT_INDEX_FILE': tmp_index,
                      'GIT_WORK_TREE': os.path.abspath(data_dir)}
        repo.run_cmd('add .', env_update)

        # Remove files that are excluded
        if exclude:
            repo.run_cmd(['rm', '--cached'] + [f for f in exclude], env_update)

        tree = repo.run_cmd('write-tree', env_update)

        # Create new commit object from the tree
        parent = repo.rev_parse(branch)
        git_cmd = ['commit-tree', tree, '-m', message]
        if parent:
            git_cmd += ['-p', parent]
        commit = repo.run_cmd(git_cmd, env_update)

        # Create git notes
        for ref, filename in notes:
            ref = ref.format(branch_name=branch)
            repo.run_cmd(['notes', '--ref', ref, 'add',
                          '-F', os.path.abspath(filename), commit])

        # Update branch head
        git_cmd = ['update-ref', 'refs/heads/' + branch, commit]
        if parent:
            git_cmd.append(parent)
        repo.run_cmd(git_cmd)

        # Update current HEAD, if we're on branch 'branch'
        if not repo.bare and repo.get_current_branch() == branch:
            log.info("Updating %s HEAD to latest commit", repo.top_dir)
            repo.run_cmd('reset --hard')

        return commit
    finally:
        if os.path.exists(tmp_index):
            os.unlink(tmp_index)


def expand_tag_strings(repo, name_pattern, msg_subj_pattern, msg_body_pattern,
                       keywords):
    """Generate tag name and message, with support for running id number"""
    keyws = keywords.copy()
    # Tag number is handled specially: if not defined, we autoincrement it
    if 'tag_number' not in keyws:
        # Fill in all other fields than 'tag_number'
        keyws['tag_number'] = '{tag_number}'
        tag_re = format_str(name_pattern, keyws)
        # Replace parentheses for proper regex matching
        tag_re = tag_re.replace('(', '\(').replace(')', '\)') + '$'
        # Inject regex group pattern for 'tag_number'
        tag_re = tag_re.format(tag_number='(?P<tag_number>[0-9]{1,5})')

        keyws['tag_number'] = 0
        for existing_tag in repo.run_cmd('tag').splitlines():
            match = re.match(tag_re, existing_tag)

            if match and int(match.group('tag_number')) >= keyws['tag_number']:
                keyws['tag_number'] = int(match.group('tag_number')) + 1

    tag_name = format_str(name_pattern, keyws)
    msg_subj= format_str(msg_subj_pattern.strip(), keyws)
    msg_body = format_str(msg_body_pattern, keyws)
    return tag_name, msg_subj + '\n\n' + msg_body


def parse_args(argv):
    """Parse command line arguments"""
    parser = argparse.ArgumentParser(
            description="Commit data to git and push upstream",
            formatter_class=argparse.ArgumentDefaultsHelpFormatter)

    parser.add_argument('--debug', '-D', action='store_true',
                        help="Verbose logging")
    parser.add_argument('--git-dir', '-g', required=True,
                        help="Local git directory to use")
    parser.add_argument('--no-create', action='store_true',
                        help="If GIT_DIR is not a valid Git repository, do not "
                             "try to create one")
    parser.add_argument('--bare', action='store_true',
                        help="Initialize a bare repository when creating a "
                             "new one")
    parser.add_argument('--push', '-p', nargs='?', default=False, const=True,
                        help="Push to remote")
    parser.add_argument('--branch-name', '-b',
                        default='{hostname}/{branch}/{machine}',
                        help="Git branch name (pattern) to use")
    parser.add_argument('--no-tag', action='store_true',
                        help="Do not create Git tag")
    parser.add_argument('--tag-name', '-t',
                        default='{hostname}/{branch}/{machine}/{commit_count}-g{commit}/{tag_number}',
                        help="Tag name (pattern) to use")
    parser.add_argument('--commit-msg-subject',
                        default='Results of {branch}:{commit} on {hostname}',
                        help="Subject line (pattern) to use in the commit message")
    parser.add_argument('--commit-msg-body',
                        default='branch: {branch}\ncommit: {commit}\nhostname: {hostname}',
                        help="Commit message body (pattern)")
    parser.add_argument('--tag-msg-subject',
                        default='Test run #{tag_number} of {branch}:{commit} on {hostname}',
                        help="Subject line (pattern) of the tag message")
    parser.add_argument('--tag-msg-body',
                        default='',
                        help="Tag message body (pattern)")
    parser.add_argument('--exclude', action='append', default=[],
                        help="Glob to exclude files from the commit. Relative "
                             "to DATA_DIR. May be specified multiple times")
    parser.add_argument('--notes', nargs=2, action='append', default=[],
                        metavar=('GIT_REF', 'FILE'),
                        help="Add a file as a note under refs/notes/GIT_REF. "
                             "{branch_name} in GIT_REF will be expanded to the "
                             "actual target branch name (specified by "
                             "--branch-name). This option may be specified "
                             "multiple times.")
    parser.add_argument('data_dir', metavar='DATA_DIR',
                        help="Data to commit")
    return parser.parse_args(argv)

def get_nested(d, list_of_keys):
    try:
        for k in list_of_keys:
            d = d[k]
        return d
    except KeyError:
        return ""

def main(argv=None):
    """Script entry point"""
    args = parse_args(argv)
    if args.debug:
        log.setLevel(logging.DEBUG)

    try:
        if not os.path.isdir(args.data_dir):
            raise ArchiveError("Not a directory: {}".format(args.data_dir))

        data_repo = init_git_repo(args.git_dir, args.no_create, args.bare)

        # Get keywords to be used in tag and branch names and messages
        metadata = metadata_from_bb()
        keywords = {'hostname': get_nested(metadata, ['hostname']),
                    'branch': get_nested(metadata, ['layers', 'meta', 'branch']),
                    'commit': get_nested(metadata, ['layers', 'meta', 'commit']),
                    'commit_count': get_nested(metadata, ['layers', 'meta', 'commit_count']),
                    'machine': get_nested(metadata, ['config', 'MACHINE'])}

        # Expand strings early in order to avoid getting into inconsistent
        # state (e.g. no tag even if data was committed)
        commit_msg = format_str(args.commit_msg_subject.strip(), keywords)
        commit_msg += '\n\n' + format_str(args.commit_msg_body, keywords)
        branch_name = format_str(args.branch_name, keywords)
        tag_name = None
        if not args.no_tag and args.tag_name:
            tag_name, tag_msg = expand_tag_strings(data_repo, args.tag_name,
                                                   args.tag_msg_subject,
                                                   args.tag_msg_body, keywords)

        # Commit data
        commit = git_commit_data(data_repo, args.data_dir, branch_name,
                                 commit_msg, args.exclude, args.notes)

        # Create tag
        if tag_name:
            log.info("Creating tag %s", tag_name)
            data_repo.run_cmd(['tag', '-a', '-m', tag_msg, tag_name, commit])

        # Push data to remote
        if args.push:
            cmd = ['push', '--tags']
            # If no remote is given we push with the default settings from
            # gitconfig
            if args.push is not True:
                notes_refs = ['refs/notes/' + ref.format(branch_name=branch_name)
                                for ref, _ in args.notes]
                cmd.extend([args.push, branch_name] + notes_refs)
            log.info("Pushing data to remote")
            data_repo.run_cmd(cmd)

    except ArchiveError as err:
        log.error(str(err))
        return 1

    return 0

if __name__ == "__main__":
    sys.exit(main())