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
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
|
# encoding: utf8
import datetime
import os
import re
import sys
import textwrap
import numpy
import pandas as pd
import license
download_keys = """
Origin
Original-Source
Source
Source-Code
X-Origin
X-Original-Package
X-Source
"""
def concat(xss):
all = []
for xs in xss:
all.extend(xs)
return all
def today():
return datetime.datetime.now().strftime('%Y-%m-%d')
def warn(*x):
print >>sys.stderr, ('warning:',) + x
class PkgData(object):
def __init__(self):
pkg_store = pd.HDFStore('pkg.h5')
self.pkgs = pkg_store['packages']
self.descs = pkg_store['descriptions']
pkg_store.close()
cp_store = pd.HDFStore('cp.h5')
self.cpf = cp_store['cp_files']
self.cps = cp_store['cp_summary']
cp_store.close()
class Template(object):
def __init__(self, name, values):
self.name = name
self.values = values
def __str__(self):
return '{{%s\n%s\n}}' % (
self.name,
'\n'.join(['|' + '%s=%s' %
(n, v.encode('utf8') if isinstance(v, unicode) else v)
for (n, v) in self.values]))
def parse_tags(s):
return s.replace('\n', '').split(', ')
def extract_languages(tags):
langs = []
for tag in tags:
(a, b) = tag.split('::')
if a == 'implemented-in':
langs.append(b)
elif a == 'devel' and b.startswith('lang:'):
langs.append(b.split(':')[1])
return list(set(langs))
def munge_description(s):
paras = s.split('\n .\n')
return '\n\n'.join(
textwrap.fill(para.lstrip().replace('\n', ''), 65) for para in paras)
def get_license_map():
map = {}
for para in file('license_map').read().split('\n\n'):
if not para:
continue
match = re.match('\[([^\]]+)\]', para)
assert match, para
canonical = match.group(1)
aliases = para[match.end():].lstrip().splitlines()
for alias in aliases:
map[alias] = canonical
return map
def srcpkg_extract_licenses(header, filess):
# XXX: generate template from header stanza
# XXX: flag CC licenses
# XXX: check all License stanzas were included
lmap = get_license_map()
for (_ix, files) in filess.iterrows():
lname = files['_license'].strip()
if '\n' not in lname:
# Looks like license text is present.
txt = files['License']
else:
# Licens information is a stub.
# XXX: look it up
txt = lname
canon = lmap.get(lname.lower(), 'Other')
# XXX: Should maybe bail if there's no copyright field.
cp = ''.join(
u'© %s\n' % line
for line in files.dropna().get('Copyright', '').splitlines())
cp = cp.encode('utf8')
txt = txt.encode('utf8')
yield Template('Project license', [
('License', canon),
('License note', (cp + '\n' + txt))])
def parse_person(s):
match = re.match('([^<]+)\s+<([^>]+)>', s)
if match:
return (match.group(1), match.group(2))
else:
return (s, '')
def extract_people(df):
# XXX: extract contributors, maintainers
df = df.dropna()
if 'Upstream-Contact' in df:
(name, email) = parse_person(df['Upstream-Contact'])
yield Template('Person', [
('Real name', name),
('Role', 'contact'),
('Email', email),
('Resource URL', '')])
def extract_resources(cp_header):
cp_header = cp_header.dropna()
for key in re.findall('\S+', download_keys):
if key in cp_header:
yield Template('Resource', [
('Resource kind', 'Download'),
('Resource URL', cp_header[key])])
def export_srcpkgs(data, name, srcpkg_names):
binpkgs = pd.concat([
data.pkgs[data.pkgs['_srcpkg'] == srcpkg]
for srcpkg in srcpkg_names])
if len(binpkgs) == 0:
warn('no binary packages found for', srcpkg_names)
return
binpkg_names = sorted(binpkgs['Package'], key=len)
homepages = list(binpkgs['Homepage'])
# XXX: maybe choose the one that appears the most?
homepage = homepages[0] if homepages else ''
tags = set(concat(
[parse_tags(t) for t in binpkgs['Tag'] if not pd.isnull(t)]))
langs = [s.title() for s in extract_languages(tags)]
if name in binpkg_names:
descpkg = name
else:
# Heuristic: choose the package with the shortest name.
# We could try to do something smarter, like look for the common
# prefix of the descriptions of all the binary packages.
descpkg = binpkg_names[0]
desc = list(data.descs[
data.descs['Package'] == descpkg]['Description-en'])[0]
(short_desc, full_desc) = desc.split('\n', 1)
full_desc = munge_description(full_desc)
yield Template('Entry', [
('Name', name.capitalize()),
('Short description', short_desc),
('Full description', full_desc),
('Homepage URL', homepage),
('User level', ''),
# XXX get this information from apt-file
('Component programs', ''),
('VCS checkout command', ''),
('Computer languages', ', '.join(langs)),
('Status', ''),
('Is GNU', 'No'),
('Submitted by', 'Debian import'),
('Submitted date', today())])
yield Template('Import', [
('Source', 'Debian'),
('Source link',
'http://packages.debian.org/sid/' + srcpkg_names[0]),
('Date', today())])
people = []
res = []
for srcpkg in srcpkg_names:
pkg_cps = data.cps[data.cps['_srcpkg'] == srcpkg].ix[0]
pkg_cpf = data.cpf[data.cpf['_srcpkg'] == srcpkg]
people.extend(list(extract_people(pkg_cps)))
res.extend(list(extract_resources(pkg_cps)))
#licenses = license.parse_licenses(list(pkg_cpf['_license']))
#licenses = [
# license.parse_licenses(row['_license'])
# for (_ix, row) in pkg_cpf.iterrows()]
#print licenses
#all = set(concat(l.flatten() for l in licenses))
for template in srcpkg_extract_licenses(pkg_cps, pkg_cpf):
# XXX: eliminate duplicates
yield template
for template in people:
# XXX: eliminate duplicates
yield template
for template in res:
# XXX: eliminate duplicates
yield template
#yield Template('Software category', [
# ('Resource kind', ''),
# ('Resource URL', '')])
def export(data, name):
pkg_cps = data.cps[data.cps['Upstream-Name'] == name]
srcpkg_names = list(pkg_cps['_srcpkg'])
for template in export_srcpkgs(data, name, srcpkg_names):
yield template
def filename(s):
s_ = re.sub('[^A-Za-z0-9_+.-]', '_', s)
assert s_, s
return s_ + '.wiki'
def output(path, xs):
with open(path, 'w') as f:
for x in xs:
f.write(str(x) + '\n')
def main():
data = PkgData()
args = sys.argv[1:]
outputdir = 'output'
if not os.path.exists(outputdir):
os.makedirs(outputdir)
if len(args) == 0:
# First, find all upstream names and the source packages corresponding
# to them.
unames = set(data.cps['Upstream-Name'].dropna())
for uname in unames:
if not uname:
continue
uname = uname.encode('utf8')
print uname
fname = os.path.join(outputdir, filename(uname))
output(fname, export(data, uname))
# For source packages with no upstream name, use the source package
# name as the upstream name.
no_uname = set(data.cps[
data.cps['Upstream-Name'].isnull()]['_srcpkg'])
for srcpkg in no_uname:
print srcpkg
fname = os.path.join(outputdir, filename(srcpkg))
output(fname, export_srcpkgs(data, srcpkg, [srcpkg]))
elif len(args) == 1:
# XXX: assumes argument is an upstream name
for template in export(data, args[0]):
print template
else:
raise RuntimeError()
if __name__ == '__main__':
main()
|