sq 0.1.3 → 0.1.4
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- checksums.yaml.gz.sig +0 -0
- data.tar.gz.sig +1 -0
- data/bin/sq +1 -1
- data/lib/sq.rb +31 -18
- data/lib/version.rb +3 -1
- data/tests/query_tests.rb +4 -0
- data/tests/tests.rb +5 -1
- metadata +70 -33
- metadata.gz.sig +0 -0
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: aeab6710743c4e7c1d05af7282c199a21e882be6
|
4
|
+
data.tar.gz: dcc24a5f930eed920f52df8069a82e9aa0800e4a
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 10a8027f10d94fb2d4f5d37e45a85b8d528047ff57c2706c6031850054ceefc44793f8666733a9779349268ca2ee62efb305acd47f8b1d43dbeda687ef6226cd
|
7
|
+
data.tar.gz: ee670820c2729551505f23058ed9ee19884ad8b52dcda566b1a3dc416ce1027a73a84555732597dc0f75fce70ca36bf05e3378acf5f75395274c053ae8216cbc
|
checksums.yaml.gz.sig
ADDED
Binary file
|
data.tar.gz.sig
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
��Mr�`?)mC����ƒa�V$Bg�����3���E<K��'����)���� #�k�Hә�9����Z8�"��T�+�5�L?~��s�2���3�$�s���47�������ݤ�!v<�q�?���(��`��9"���=����D.h�p��K�IH���4L���x~�:��\d��c�+z]��!��S!lP}�?#~��{���p�~������vz�>%k��t,*��|77�b�ʆ�ƛ�"��;�H�m���
|
data/bin/sq
CHANGED
@@ -33,7 +33,7 @@ is '%s.pdf'.
|
|
33
33
|
%n - PDF number, starting at 0
|
34
34
|
%N - PDF number, starting at 1
|
35
35
|
%z - same as %n, but zero-padded
|
36
|
-
%Z - same as %
|
36
|
+
%Z - same as %N, but zero-padded
|
37
37
|
%c - total number of PDFs
|
38
38
|
%s - name of the PDF, extracted from its URI, without `.pdf`
|
39
39
|
%S - name of the PDF, extracted from the link text
|
data/lib/sq.rb
CHANGED
@@ -7,26 +7,37 @@ require 'fileutils'
|
|
7
7
|
require 'ruby-progressbar'
|
8
8
|
require File.expand_path(File.dirname __FILE__) + '/version'
|
9
9
|
|
10
|
+
# This module provide some tools to bulk-download a set of PDF documents, all
|
11
|
+
# linked in one HTML page.
|
10
12
|
module SQ
|
11
13
|
class << self
|
12
|
-
# return the user-agent used by SQ
|
14
|
+
# @return [String] the user-agent used by SQ
|
13
15
|
def user_agent
|
14
16
|
"SQ/#{version} +github.com/bfontaine/sq"
|
15
17
|
end
|
16
18
|
|
17
|
-
#
|
19
|
+
# Query an URI and return a list of PDFs. Each PDF is an hash with three
|
18
20
|
# keys: +:uri+ is its absolute URI, +:name+ is its name (last part of its
|
19
21
|
# URI), and +:text+ is each link text.
|
20
|
-
# @uri [String]
|
21
|
-
# @regex [Regexp]
|
22
|
+
# @param uri [String]
|
23
|
+
# @param regex [Regexp]
|
24
|
+
# @return [Array<Hash>]
|
22
25
|
def query(uri, regex=/./)
|
23
26
|
uri = 'http://' + uri unless uri =~ /^https?:\/\//
|
24
27
|
|
25
28
|
doc = Nokogiri::HTML(open(uri, 'User-Agent' => user_agent))
|
26
29
|
links = doc.css('a[href]')
|
27
30
|
|
28
|
-
uris = links.map
|
29
|
-
|
31
|
+
uris = links.map do |a|
|
32
|
+
full = begin
|
33
|
+
URI.join(uri, a.attr('href'))
|
34
|
+
rescue
|
35
|
+
nil
|
36
|
+
end
|
37
|
+
|
38
|
+
[a.text, full]
|
39
|
+
end
|
40
|
+
uris.select! { |_,u| u && u.path =~ /\.pdf$/i && u.to_s =~ regex }
|
30
41
|
|
31
42
|
uris.map do |text,u|
|
32
43
|
{
|
@@ -38,9 +49,12 @@ module SQ
|
|
38
49
|
end
|
39
50
|
|
40
51
|
# Output a formatted filename.
|
41
|
-
# @doc [Hash]
|
42
|
-
# @fmt [String]
|
43
|
-
#
|
52
|
+
# @param doc [Hash] as returned from +SQ.query+.
|
53
|
+
# @param fmt [String] format. See the project's README for more info on
|
54
|
+
# available format options
|
55
|
+
# @param opts [Hash] additional info. Supported keys include: +:number+
|
56
|
+
# (the current number), +:count+ (total files count).
|
57
|
+
# @return [String]
|
44
58
|
def format(doc, fmt='%s.pdf', opts={})
|
45
59
|
opts[:number] ||= 0
|
46
60
|
opts[:count] ||= 0
|
@@ -63,15 +77,14 @@ module SQ
|
|
63
77
|
end
|
64
78
|
end
|
65
79
|
|
66
|
-
#
|
67
|
-
#
|
68
|
-
# @
|
69
|
-
# @
|
70
|
-
#
|
71
|
-
#
|
72
|
-
#
|
73
|
-
#
|
74
|
-
#
|
80
|
+
# Query an URI and download all PDFs which match the regex.
|
81
|
+
# @param uri [String]
|
82
|
+
# @param regex [Regexp] Regex to use to match PDF URIs
|
83
|
+
# @param opts [Hash] Supported options: +:verbose+, +:directory+
|
84
|
+
# (specify the directory to use for output instead of
|
85
|
+
# the current one), and +:format+ the output format.
|
86
|
+
# See the README for details.
|
87
|
+
# @return [Integer] number of downloaded PDFs.
|
75
88
|
def process(uri, regex=/./, opts={})
|
76
89
|
uris = self.query(uri, regex)
|
77
90
|
count = uris.count
|
data/lib/version.rb
CHANGED
data/tests/query_tests.rb
CHANGED
data/tests/tests.rb
CHANGED
@@ -1,12 +1,16 @@
|
|
1
1
|
#! /usr/bin/env ruby
|
2
2
|
# -*- coding: UTF-8 -*-
|
3
3
|
|
4
|
+
require 'coveralls'
|
5
|
+
Coveralls.wear!
|
6
|
+
|
4
7
|
require 'test/unit'
|
5
8
|
require 'simplecov'
|
6
9
|
|
7
10
|
test_dir = File.expand_path( File.dirname(__FILE__) )
|
8
11
|
|
9
|
-
SimpleCov.
|
12
|
+
SimpleCov.formatter = Coveralls::SimpleCov::Formatter
|
13
|
+
SimpleCov.start { add_filter '/tests/' }
|
10
14
|
|
11
15
|
require 'sq'
|
12
16
|
|
metadata
CHANGED
@@ -1,113 +1,149 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: sq
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.4
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Baptiste Fontaine
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
|
-
cert_chain:
|
11
|
-
|
10
|
+
cert_chain:
|
11
|
+
- |
|
12
|
+
-----BEGIN CERTIFICATE-----
|
13
|
+
MIIDaDCCAlCgAwIBAgIBATANBgkqhkiG9w0BAQUFADA9MRAwDgYDVQQDDAdiYXRp
|
14
|
+
Zm9uMRUwEwYKCZImiZPyLGQBGRYFeWFob28xEjAQBgoJkiaJk/IsZAEZFgJmcjAe
|
15
|
+
Fw0xNDA4MjQxMTM5NTJaFw0xNTA4MjQxMTM5NTJaMD0xEDAOBgNVBAMMB2JhdGlm
|
16
|
+
b24xFTATBgoJkiaJk/IsZAEZFgV5YWhvbzESMBAGCgmSJomT8ixkARkWAmZyMIIB
|
17
|
+
IjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAn3uOgWl+FwXIjDdCay28i6cK
|
18
|
+
FxHWhHoS/mH9pkXzSGVctEKP2fulie6MkIvrLCP5M6TpByeaBjcJjZPadrou1FIc
|
19
|
+
Yc/O14jYjaKTqfMxpzgNfGzDdBgBo0QZ9rcHjORetdIZdUSDaZjPtI1aGS6eBMsh
|
20
|
+
W2X6GxL4UQ1kH0Lyg7iPYAH5RHnD3+G+S28iOPFfRLFzm4fwJp1k7URiiSyOHTDp
|
21
|
+
B0ZehKKrW/ibCaRMYp2VoCamcim4de1VA6CTOaYSShueqThE18n1HM6aprihziyM
|
22
|
+
04yIpo80/unO6JxlsUFdBjsb5d7oJSqPJ6/OfcFnyXa/VRm+Ed9d6PTwZvL7YwID
|
23
|
+
AQABo3MwcTAJBgNVHRMEAjAAMAsGA1UdDwQEAwIEsDAdBgNVHQ4EFgQUWyH3xMx1
|
24
|
+
8I5NLncgNEC0ZlDRKegwGwYDVR0RBBQwEoEQYmF0aWZvbkB5YWhvby5mcjAbBgNV
|
25
|
+
HRIEFDASgRBiYXRpZm9uQHlhaG9vLmZyMA0GCSqGSIb3DQEBBQUAA4IBAQAaTVya
|
26
|
+
CGgojxBwUoadCCIsFheGsXvSFhikkXYNXy0VxEYr8BaTfGwzYh9c9T5N+Y5Mu5MW
|
27
|
+
WegqwIwRhIu6Rg7huqJ7TK50pVDF0yrZcsxvWjOfd3clblBHjKGQx5Mbu7LVNGKE
|
28
|
+
+QNdTAwYVTAA8wXHpxk200cHb9xz4e9ANpb4lonGuPz8jKmb/A7Z1M5QD6zStG8l
|
29
|
+
sTlVAhA/LZiC9gL9LtW8Iq7o7xRFhxNPKWHu6JVThH9i16eli+JignOJbGna7C40
|
30
|
+
QnOQb8zHyNL+gq2m/mnZGrSehx+6AujokjOfHbmivYMfDATOQQx0eIBI18IhacZm
|
31
|
+
42WxhhIV2bwDtd77
|
32
|
+
-----END CERTIFICATE-----
|
33
|
+
date: 2014-11-10 00:00:00.000000000 Z
|
12
34
|
dependencies:
|
13
35
|
- !ruby/object:Gem::Dependency
|
14
36
|
name: nokogiri
|
15
37
|
requirement: !ruby/object:Gem::Requirement
|
16
38
|
requirements:
|
17
|
-
- - ~>
|
39
|
+
- - "~>"
|
18
40
|
- !ruby/object:Gem::Version
|
19
|
-
version: 1.6
|
41
|
+
version: '1.6'
|
20
42
|
type: :runtime
|
21
43
|
prerelease: false
|
22
44
|
version_requirements: !ruby/object:Gem::Requirement
|
23
45
|
requirements:
|
24
|
-
- - ~>
|
46
|
+
- - "~>"
|
25
47
|
- !ruby/object:Gem::Version
|
26
|
-
version: 1.6
|
48
|
+
version: '1.6'
|
27
49
|
- !ruby/object:Gem::Dependency
|
28
50
|
name: trollop
|
29
51
|
requirement: !ruby/object:Gem::Requirement
|
30
52
|
requirements:
|
31
|
-
- - ~>
|
53
|
+
- - "~>"
|
32
54
|
- !ruby/object:Gem::Version
|
33
55
|
version: '2.0'
|
34
56
|
type: :runtime
|
35
57
|
prerelease: false
|
36
58
|
version_requirements: !ruby/object:Gem::Requirement
|
37
59
|
requirements:
|
38
|
-
- - ~>
|
60
|
+
- - "~>"
|
39
61
|
- !ruby/object:Gem::Version
|
40
62
|
version: '2.0'
|
41
63
|
- !ruby/object:Gem::Dependency
|
42
64
|
name: ruby-progressbar
|
43
65
|
requirement: !ruby/object:Gem::Requirement
|
44
66
|
requirements:
|
45
|
-
- -
|
67
|
+
- - "~>"
|
46
68
|
- !ruby/object:Gem::Version
|
47
|
-
version: '
|
69
|
+
version: '1.4'
|
48
70
|
type: :runtime
|
49
71
|
prerelease: false
|
50
72
|
version_requirements: !ruby/object:Gem::Requirement
|
51
73
|
requirements:
|
52
|
-
- -
|
74
|
+
- - "~>"
|
53
75
|
- !ruby/object:Gem::Version
|
54
|
-
version: '
|
76
|
+
version: '1.4'
|
55
77
|
- !ruby/object:Gem::Dependency
|
56
78
|
name: simplecov
|
57
79
|
requirement: !ruby/object:Gem::Requirement
|
58
80
|
requirements:
|
59
|
-
- -
|
81
|
+
- - "~>"
|
60
82
|
- !ruby/object:Gem::Version
|
61
|
-
version: '0'
|
83
|
+
version: '0.8'
|
62
84
|
type: :development
|
63
85
|
prerelease: false
|
64
86
|
version_requirements: !ruby/object:Gem::Requirement
|
65
87
|
requirements:
|
66
|
-
- -
|
88
|
+
- - "~>"
|
67
89
|
- !ruby/object:Gem::Version
|
68
|
-
version: '0'
|
90
|
+
version: '0.8'
|
69
91
|
- !ruby/object:Gem::Dependency
|
70
92
|
name: rake
|
71
93
|
requirement: !ruby/object:Gem::Requirement
|
72
94
|
requirements:
|
73
|
-
- -
|
95
|
+
- - "~>"
|
74
96
|
- !ruby/object:Gem::Version
|
75
|
-
version: '
|
97
|
+
version: '10.1'
|
76
98
|
type: :development
|
77
99
|
prerelease: false
|
78
100
|
version_requirements: !ruby/object:Gem::Requirement
|
79
101
|
requirements:
|
80
|
-
- -
|
102
|
+
- - "~>"
|
81
103
|
- !ruby/object:Gem::Version
|
82
|
-
version: '
|
104
|
+
version: '10.1'
|
83
105
|
- !ruby/object:Gem::Dependency
|
84
106
|
name: test-unit
|
85
107
|
requirement: !ruby/object:Gem::Requirement
|
86
108
|
requirements:
|
87
|
-
- -
|
109
|
+
- - "~>"
|
88
110
|
- !ruby/object:Gem::Version
|
89
|
-
version: '
|
111
|
+
version: '2.5'
|
90
112
|
type: :development
|
91
113
|
prerelease: false
|
92
114
|
version_requirements: !ruby/object:Gem::Requirement
|
93
115
|
requirements:
|
94
|
-
- -
|
116
|
+
- - "~>"
|
95
117
|
- !ruby/object:Gem::Version
|
96
|
-
version: '
|
118
|
+
version: '2.5'
|
97
119
|
- !ruby/object:Gem::Dependency
|
98
120
|
name: fakeweb
|
99
121
|
requirement: !ruby/object:Gem::Requirement
|
100
122
|
requirements:
|
101
|
-
- -
|
123
|
+
- - "~>"
|
102
124
|
- !ruby/object:Gem::Version
|
103
|
-
version: '
|
125
|
+
version: '1.3'
|
104
126
|
type: :development
|
105
127
|
prerelease: false
|
106
128
|
version_requirements: !ruby/object:Gem::Requirement
|
107
129
|
requirements:
|
108
|
-
- -
|
130
|
+
- - "~>"
|
109
131
|
- !ruby/object:Gem::Version
|
110
|
-
version: '
|
132
|
+
version: '1.3'
|
133
|
+
- !ruby/object:Gem::Dependency
|
134
|
+
name: coveralls
|
135
|
+
requirement: !ruby/object:Gem::Requirement
|
136
|
+
requirements:
|
137
|
+
- - "~>"
|
138
|
+
- !ruby/object:Gem::Version
|
139
|
+
version: '0.7'
|
140
|
+
type: :development
|
141
|
+
prerelease: false
|
142
|
+
version_requirements: !ruby/object:Gem::Requirement
|
143
|
+
requirements:
|
144
|
+
- - "~>"
|
145
|
+
- !ruby/object:Gem::Version
|
146
|
+
version: '0.7'
|
111
147
|
description: Download all PDFs linked in a Web page
|
112
148
|
email: batifon@yahoo.fr
|
113
149
|
executables:
|
@@ -115,13 +151,13 @@ executables:
|
|
115
151
|
extensions: []
|
116
152
|
extra_rdoc_files: []
|
117
153
|
files:
|
154
|
+
- bin/sq
|
118
155
|
- lib/sq.rb
|
119
156
|
- lib/version.rb
|
120
157
|
- tests/format_tests.rb
|
121
158
|
- tests/process_tests.rb
|
122
159
|
- tests/query_tests.rb
|
123
160
|
- tests/tests.rb
|
124
|
-
- bin/sq
|
125
161
|
homepage: https://github.com/bfontaine/sq
|
126
162
|
licenses:
|
127
163
|
- MIT
|
@@ -132,17 +168,17 @@ require_paths:
|
|
132
168
|
- lib
|
133
169
|
required_ruby_version: !ruby/object:Gem::Requirement
|
134
170
|
requirements:
|
135
|
-
- -
|
171
|
+
- - ">="
|
136
172
|
- !ruby/object:Gem::Version
|
137
173
|
version: '0'
|
138
174
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
139
175
|
requirements:
|
140
|
-
- -
|
176
|
+
- - ">="
|
141
177
|
- !ruby/object:Gem::Version
|
142
178
|
version: '0'
|
143
179
|
requirements: []
|
144
180
|
rubyforge_project:
|
145
|
-
rubygems_version: 2.
|
181
|
+
rubygems_version: 2.2.2
|
146
182
|
signing_key:
|
147
183
|
specification_version: 4
|
148
184
|
summary: Bulk PDFs downloader
|
@@ -151,3 +187,4 @@ test_files:
|
|
151
187
|
- tests/process_tests.rb
|
152
188
|
- tests/query_tests.rb
|
153
189
|
- tests/tests.rb
|
190
|
+
has_rdoc:
|
metadata.gz.sig
ADDED
Binary file
|