summaryrefslogtreecommitdiffstats
path: root/web-export/update.py
blob: 1f43676e34c426543e7afd6abfea2a1abc90983f (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
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
#!/usr/bin/env python

# Dependencies to run this:
#  - xmlto in $PATH

# FIXME:
#  - correctly handle all exceptions
#  - copy dtd files where they should be
#  - new structure for website:
#    specs.fd.o/index.html -- general index
#    specs.fd.o/desktop-entry/index.html -- index of all versions of desktop entry, with all formats
#    specs.fd.o/desktop-entry/1.0/desktop-entry-spec.xml -- docbook version of the spec 1.0
#    specs.fd.o/desktop-entry/1.0/index.html -- one-page html version of the spec 1.0
#    specs.fd.o/desktop-entry/1.0/split/ -- multiple-page html version of the spec 1.0
#    specs.fd.o/desktop-entry/latest/ -- link to directory containing latest version of the spec

import os
import sys

import errno

import StringIO
import hashlib
import shutil
import subprocess
import urllib
import urllib2
import urlparse

DEVELOPMENT = False

CVSWEB = 'http://cvs.freedesktop.org'
GITWEB = 'http://cgit.freedesktop.org'
HASH = 'md5'


def safe_mkdir(dir):
    if not dir:
        return

    try:
        os.mkdir(dir)
    except OSError, e:
        if e.errno != errno.EEXIST:
            raise e


def get_hash_from_fd(fd, algo = HASH, read_blocks = 1024):
    if algo not in [ 'md5' ]:
        raise Exception('Internal error: hash algorithm \'%s\' not planned in code.' % algo)

    hash = hashlib.new(algo)
    while True:
        data = fd.read(read_blocks)
        if not data:
            break
        hash.update(data)
    return hash.digest()


def get_hash_from_url(url, algo = HASH):
    fd = urllib2.urlopen(url, None)
    digest = get_hash_from_fd(fd, algo)
    fd.close()
    return digest


def get_hash_from_path(path, algo = HASH):
    fd = open(path, 'rb')
    digest = get_hash_from_fd(fd, algo, read_blocks = 32768)
    fd.close()
    return digest


def get_hash_from_data(data, algo = HASH):
    fd = StringIO.StringIO(data)
    digest = get_hash_from_fd(fd, algo, read_blocks = 32768)
    fd.close()
    return digest


class VcsObject:
    def __init__(self, vcs, repo, file, revision = None):
        self.vcs = vcs
        self.repo = repo
        self.file = file
        self.revision = revision
        self.data = None

    def get_url(self):
        query = {}
        if self.vcs == 'git':
            baseurl = GITWEB
            path = '/'.join((self.repo, 'plain', self.file))
            if self.revision:
                query['id'] = self.revision
        elif self.vcs == 'cvs':
            baseurl = CVSWEB
            path = self.file
            if self.revision:
                query['rev'] = self.revision
        else:
            raise Exception('Unknown VCS: %s' % self.vcs)

        (scheme, netloc, basepath) = urlparse.urlsplit(baseurl)[0:3]
        full_path = '/'.join((basepath, path))

        query_str = urllib.urlencode(query)
        return urlparse.urlunsplit((scheme, netloc, full_path, query_str, ''))

    def fetch(self):
        if self.data:
            return

        url = self.get_url()
        fd = urllib2.urlopen(url, None)
        self.data = fd.read()
        fd.close()

    def get_hash(self):
        self.fetch()
        return get_hash_from_data(self.data)


class SpecObject():
    def __init__(self, vcs, spec_dir, version):
        self.vcs = vcs
        self.spec_dir = spec_dir
        self.version = version

        basename = os.path.basename(self.vcs.file)
        (self.basename_no_ext, self.ext) = os.path.splitext(basename)

        self.filename = '%s-%s%s' % (self.basename_no_ext, self.version, self.ext)

        if self.ext not in ['.xml', '.sgml', '.txt', '.dtd']:
            raise Exception('Format \'%s\' not supported for %s' % (self.ext, self.vcs.get_url()))

        self.downloaded = False
        self.one_chunk = False
        self.multiple_chunks = False

    def download(self):
        safe_mkdir(self.spec_dir)
        path = os.path.join(self.spec_dir, self.filename)

        if os.path.exists(path):
            current_hash = get_hash_from_path(path)
            vcs_hash = self.vcs.get_hash()
            if current_hash == vcs_hash:
                return

        self.vcs.fetch()
        fd = open(path, 'wb')
        fd.write(self.vcs.data)
        fd.close()

        self.downloaded = True

    def htmlize(self, force = False):
        if not self.downloaded and not force:
            return

        path = os.path.join(self.spec_dir, self.filename)
        (path_no_ext, ext) = os.path.splitext(path)

        if self.ext == '.xml':
            # One-chunk HTML
            html_path = '%s%s' % (path_no_ext, '.html')
            if os.path.exists(html_path):
                os.unlink(html_path)

            retcode = subprocess.call(['xmlto', '-o', self.spec_dir, 'html-nochunks', path])

            if retcode != 0:
                raise Exception('Cannot convert \'%s\' to HTML.' % path)
            self.one_chunk = True

            # Multiple chunks
            html_dir = os.path.join(self.spec_dir, self.version)
            if os.path.exists(html_dir):
                shutil.rmtree(html_dir)
            safe_mkdir(html_dir)

            retcode = subprocess.call(['xmlto', '-o', html_dir, 'html', path])

            if retcode != 0:
                raise Exception('Cannot convert \'%s\' to multiple-chunks HTML.' % path)
            self.multiple_chunks = True

    def latestize(self):
        filename_latest = '%s-latest%s' % (self.basename_no_ext, self.ext)

        path_latest = os.path.join(self.spec_dir, filename_latest)
        if os.path.exists(path_latest):
            os.unlink(path_latest)
        os.symlink(self.filename, path_latest)

        if self.ext == '.xml':
            # One-chunk HTML
            html_path_latest = os.path.join(self.spec_dir, '%s%s' % (self.basename_no_ext, '.html'))
            if os.path.exists(html_path_latest):
                os.unlink(html_path_latest)

            (filename_no_ext, ext) = os.path.splitext(self.filename)
            html_filename = '%s%s' % (filename_no_ext, '.html')
            html_path = os.path.join(self.spec_dir, html_filename)
            if os.path.exists(html_path):
                os.symlink(html_filename, html_path_latest)

            # Multiple chunks
            html_dir_latest = os.path.join(self.spec_dir, 'latest')
            if os.path.exists(html_dir_latest):
                os.unlink(html_dir_latest)

            html_dir = os.path.join(self.spec_dir, self.version)
            if os.path.exists(html_dir):
                os.symlink(self.version, html_dir_latest)


SCRIPT = VcsObject('git', 'xdg/xdg-specs', 'web-export/update.py')
SPECS_INDEX = VcsObject('git', 'xdg/xdg-specs', 'web-export/specs.idx')


def is_up_to_date():
    current_hash = get_hash_from_path(__file__)
    vcs_hash = SCRIPT.get_hash()

    return current_hash == vcs_hash


if not DEVELOPMENT:
    if not is_up_to_date():
        print >>sys.stderr, 'Script is not up-to-date, please download %s' % SCRIPT.get_url()
        sys.exit(1)

    SPECS_INDEX.fetch()
    lines = SPECS_INDEX.data.split('\n')
else:
    lines = open('specs.idx').readlines()


latests = []

for line in lines:
    line = line.strip()
    if not line or line.startswith('#'):
        continue

    (data, revision, version, path) = line.split()
    if data.startswith("git:"):
        git_data = data.split(":")
        vcs = VcsObject('git', git_data[1], git_data[2], revision)
    else:
        vcs = VcsObject('cvs', None, data, revision)

    spec = SpecObject(vcs, path, version)

    spec.download()
    spec.htmlize()

    # Create latest links if it's the first time we see this spec
    if (spec.spec_dir, spec.basename_no_ext) not in latests:
        latests.append((spec.spec_dir, spec.basename_no_ext))
        spec.latestize()