summaryrefslogtreecommitdiff
path: root/compose/utils.py
blob: a1e5e6435d84e27e388559da7856bea9a96b52e9 (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
182
183
184
185
186
187
from __future__ import absolute_import
from __future__ import unicode_literals

import codecs
import hashlib
import json.decoder
import logging
import ntpath
import random

import six
from docker.errors import DockerException
from docker.utils import parse_bytes as sdk_parse_bytes

from .errors import StreamParseError
from .timeparse import MULTIPLIERS
from .timeparse import timeparse


json_decoder = json.JSONDecoder()
log = logging.getLogger(__name__)


def get_output_stream(stream):
    if six.PY3:
        return stream
    return codecs.getwriter('utf-8')(stream)


def stream_as_text(stream):
    """Given a stream of bytes or text, if any of the items in the stream
    are bytes convert them to text.

    This function can be removed once docker-py returns text streams instead
    of byte streams.
    """
    for data in stream:
        if not isinstance(data, six.text_type):
            data = data.decode('utf-8', 'replace')
        yield data


def line_splitter(buffer, separator=u'\n'):
    index = buffer.find(six.text_type(separator))
    if index == -1:
        return None
    return buffer[:index + 1], buffer[index + 1:]


def split_buffer(stream, splitter=None, decoder=lambda a: a):
    """Given a generator which yields strings and a splitter function,
    joins all input, splits on the separator and yields each chunk.

    Unlike string.split(), each chunk includes the trailing
    separator, except for the last one if none was found on the end
    of the input.
    """
    splitter = splitter or line_splitter
    buffered = six.text_type('')

    for data in stream_as_text(stream):
        buffered += data
        while True:
            buffer_split = splitter(buffered)
            if buffer_split is None:
                break

            item, buffered = buffer_split
            yield item

    if buffered:
        try:
            yield decoder(buffered)
        except Exception as e:
            log.error(
                'Compose tried decoding the following data chunk, but failed:'
                '\n%s' % repr(buffered)
            )
            raise StreamParseError(e)


def json_splitter(buffer):
    """Attempt to parse a json object from a buffer. If there is at least one
    object, return it and the rest of the buffer, otherwise return None.
    """
    buffer = buffer.strip()
    try:
        obj, index = json_decoder.raw_decode(buffer)
        rest = buffer[json.decoder.WHITESPACE.match(buffer, index).end():]
        return obj, rest
    except ValueError:
        return None


def json_stream(stream):
    """Given a stream of text, return a stream of json objects.
    This handles streams which are inconsistently buffered (some entries may
    be newline delimited, and others are not).
    """
    return split_buffer(stream, json_splitter, json_decoder.decode)


def json_hash(obj):
    dump = json.dumps(obj, sort_keys=True, separators=(',', ':'), default=lambda x: x.repr())
    h = hashlib.sha256()
    h.update(dump.encode('utf8'))
    return h.hexdigest()


def microseconds_from_time_nano(time_nano):
    return int(time_nano % 1000000000 / 1000)


def nanoseconds_from_time_seconds(time_seconds):
    return int(time_seconds / MULTIPLIERS['nano'])


def parse_seconds_float(value):
    return timeparse(value or '')


def parse_nanoseconds_int(value):
    parsed = timeparse(value or '')
    if parsed is None:
        return None
    return nanoseconds_from_time_seconds(parsed)


def build_string_dict(source_dict):
    return dict((k, str(v if v is not None else '')) for k, v in source_dict.items())


def splitdrive(path):
    if len(path) == 0:
        return ('', '')
    if path[0] in ['.', '\\', '/', '~']:
        return ('', path)
    return ntpath.splitdrive(path)


def parse_bytes(n):
    try:
        return sdk_parse_bytes(n)
    except DockerException:
        return None


def unquote_path(s):
    if not s:
        return s
    if s[0] == '"' and s[-1] == '"':
        return s[1:-1]
    return s


def generate_random_id():
    while True:
        val = hex(random.getrandbits(32 * 8))[2:-1]
        try:
            int(truncate_id(val))
            continue
        except ValueError:
            return val


def truncate_id(value):
    if ':' in value:
        value = value[value.index(':') + 1:]
    if len(value) > 12:
        return value[:12]
    return value


def unique_everseen(iterable, key=lambda x: x):
    "List unique elements, preserving order. Remember all elements ever seen."
    seen = set()
    for element in iterable:
        unique_key = key(element)
        if unique_key not in seen:
            seen.add(unique_key)
            yield element


def truncate_string(s, max_chars=35):
    if len(s) > max_chars:
        return s[:max_chars - 2] + '...'
    return s