summaryrefslogtreecommitdiff
path: root/buildstream/plugins/sources/zip.py
blob: 03efcef79815b3b4903d32ae9da463764929e6aa (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
#
#  Copyright (C) 2017 Mathieu Bridon
#
#  This program is free software; you can redistribute it and/or
#  modify it under the terms of the GNU Lesser General Public
#  License as published by the Free Software Foundation; either
#  version 2 of the License, or (at your option) any later version.
#
#  This library is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
#  Lesser General Public License for more details.
#
#  You should have received a copy of the GNU Lesser General Public
#  License along with this library. If not, see <http://www.gnu.org/licenses/>.
#
#  Authors:
#        Mathieu Bridon <bochecha@daitauha.fr>

"""
zip - stage files from zip archives
===================================

**Usage:**

.. code:: yaml

   # Specify the zip source kind
   kind: zip

   # Specify the zip url. Using an alias defined in your project
   # configuration is encouraged. 'bst source track' will update the
   # sha256sum in 'ref' to the downloaded file's sha256sum.
   url: upstream:foo.zip

   # Specify the ref. It's a sha256sum of the file you download.
   ref: 6c9f6f68a131ec6381da82f2bff978083ed7f4f7991d931bfa767b7965ebc94b

   # Specify a glob pattern to indicate the base directory to extract
   # from the archive. The first matching directory will be used.
   #
   # Note that this is '*' by default since most standard release
   # archives contain a self named subdirectory at the root which
   # contains the files one normally wants to extract to build.
   #
   # To extract the root of the archive directly, this can be set
   # to an empty string.
   base-dir: '*'

See :ref:`built-in functionality doumentation <core_source_builtins>` for
details on common configuration options for sources.

.. attention::

   File permissions are not preserved. All extracted directories have
   permissions 0755 and all extracted files have permissions 0644.
"""

import os
import zipfile
import stat

from buildstream import SourceError
from buildstream import utils

from ._downloadablefilesource import DownloadableFileSource


class ZipSource(DownloadableFileSource):
    # pylint: disable=attribute-defined-outside-init

    def configure(self, node):
        super().configure(node)

        self.base_dir = self.node_get_member(node, str, 'base-dir', '*') or None

        self.node_validate(node, DownloadableFileSource.COMMON_CONFIG_KEYS + ['base-dir'])

    def get_unique_key(self):
        return super().get_unique_key() + [self.base_dir]

    def stage(self, directory):
        exec_rights = (stat.S_IRWXU | stat.S_IRWXG | stat.S_IRWXO) & ~(stat.S_IWGRP | stat.S_IWOTH)
        noexec_rights = exec_rights & ~(stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH)

        try:
            with zipfile.ZipFile(self._get_mirror_file()) as archive:
                base_dir = None
                if self.base_dir:
                    base_dir = self._find_base_dir(archive, self.base_dir)

                if base_dir:
                    members = self._extract_members(archive, base_dir)
                else:
                    members = archive.namelist()

                for member in members:
                    written = archive.extract(member, path=directory)

                    # zipfile.extract might create missing directories
                    rel = os.path.relpath(written, start=directory)
                    assert not os.path.isabs(rel)
                    rel = os.path.dirname(rel)
                    while rel:
                        os.chmod(os.path.join(directory, rel), exec_rights)
                        rel = os.path.dirname(rel)

                    if os.path.islink(written):
                        pass
                    elif os.path.isdir(written):
                        os.chmod(written, exec_rights)
                    else:
                        os.chmod(written, noexec_rights)

        except (zipfile.BadZipFile, zipfile.LargeZipFile, OSError) as e:
            raise SourceError("{}: Error staging source: {}".format(self, e)) from e

    # Override and translate which filenames to extract
    def _extract_members(self, archive, base_dir):
        if not base_dir.endswith(os.sep):
            base_dir = base_dir + os.sep

        L = len(base_dir)
        for member in archive.infolist():
            if member.filename == base_dir:
                continue

            if member.filename.startswith(base_dir):
                member.filename = member.filename[L:]
                yield member

    # We want to iterate over all paths of an archive, but namelist()
    # is not enough because some archives simply do not contain the leading
    # directory paths for the archived files.
    def _list_archive_paths(self, archive):

        visited = {}
        for member in archive.infolist():

            # ZipInfo.is_dir() is only available in python >= 3.6, but all
            # it does is check for a trailing '/' in the name
            #
            if not member.filename.endswith('/'):

                # Loop over the components of a path, for a path of a/b/c/d
                # we will first visit 'a', then 'a/b' and then 'a/b/c', excluding
                # the final component
                components = member.filename.split('/')
                for i in range(len(components) - 1):
                    dir_component = '/'.join([components[j] for j in range(i + 1)])
                    if dir_component not in visited:
                        visited[dir_component] = True
                        try:
                            # Dont yield directory members which actually do
                            # exist in the archive
                            _ = archive.getinfo(dir_component)
                        except KeyError:
                            if dir_component != '.':
                                yield dir_component

                continue

            # Avoid considering the '.' directory, if any is included in the archive
            # this is to avoid the default 'base-dir: *' value behaving differently
            # depending on whether the archive was encoded with a leading '.' or not
            elif member.filename == '.' or member.filename == './':
                continue

            yield member.filename

    def _find_base_dir(self, archive, pattern):
        paths = self._list_archive_paths(archive)
        matches = sorted(list(utils.glob(paths, pattern)))
        if not matches:
            raise SourceError("{}: Could not find base directory matching pattern: {}".format(self, pattern))

        return matches[0]


def setup():
    return ZipSource