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
|
# encoding: utf8
import re
import sys
import textwrap
import numpy
import pandas as pd
import license
def concat(xss):
all = []
for xs in xss:
all.extend(xs)
return all
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) 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 export(pkgs, descs, cps, cpf, name):
pkg_cps = cps[cps['Upstream-Name'] == name]
srcpkg_names = list(pkg_cps['_srcpkg'])
print srcpkg_names
binpkgs = pd.concat([
pkgs[pkgs['Source'] == srcpkg]
for srcpkg in srcpkg_names])
binpkg_names = sorted(binpkgs['Package'], key=len)
print binpkg_names
print list(binpkgs['Package'])
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)]))
print tags
langs = [s.title() for s in extract_languages(tags)]
print langs
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 = binpkgs[0]
desc = list(descs[descs['Package'] == descpkg]['Description-en'])[0]
(short_desc, full_desc) = desc.split('\n', 1)
full_desc = munge_description(full_desc)
print 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')])
lmap = get_license_map()
for srcpkg in srcpkg_names:
pkg_cpf = cpf[cpf['_srcpkg'] == srcpkg]
#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 (_ix, files) in pkg_cpf.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')
cp = ''.join(
u'© %s\n' % line
for line in files['Copyright'].splitlines())
cp = cp.encode('utf8')
txt = txt.encode('utf8')
print Template('Project license', [
('License', canon),
('License note', (cp + '\n' + txt))])
# XXX: eliminate duplicates
# XXX: check all License stanzas were included
# XXX: generate template from header stanza
# XXX: flag CC licenses
print Template('Person', [
('Real name', ''),
('Role', ''),
('Email', ''),
('Resource URL', '')])
print Template('Resource', [
('Resource kind', ''),
('Resource URL', '')])
#print Template('Software category', [
# ('Resource kind', ''),
# ('Resource URL', '')])
def main():
pkg_store = pd.HDFStore('pkg.h5')
pkgs = pkg_store['packages']
descs = pkg_store['descriptions']
pkg_store.close()
cp_store = pd.HDFStore('cp.h5')
cpf = cp_store['cp_files']
cps = cp_store['cp_summary']
cp_store.close()
args = sys.argv[1:]
if len(args) == 0:
srcps = sorted(set(pkgs['Source']))
for pkgname in srcps[:100]:
export(pkgs, descs, cps, cpf, pkgname)
elif len(args) == 1:
export(pkgs, descs, cps, cpf, args[0])
else:
raise RuntimeError()
if __name__ == '__main__':
main()
|