wayback_machine_downloader 0.3.0 → 0.4.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/lib/wayback_machine_downloader.rb +16 -8
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: f581f0aa09dbfecb7080ae0788073c38bef547b1
|
|
4
|
+
data.tar.gz: 8013f6f7b57cf41b8674e37eb0beb9dada4d1b2b
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 02ce907c84817031c068a10c4b78eeaf08ff4b2c4805159d01bb1a0fb40eb4fa0d212a0a101f4c61cd1309dc6aafc0f95969759e0930dacddcc566f500051cfc
|
|
7
|
+
data.tar.gz: 7131a398e753f271a8db19c7ad1677de9241b3cddf951efa9bf561e97d10b8a81e8f41ae0a676d7eccd8ed0d76770c3b7fb582989f8cab2659af94fa07681d0c
|
|
@@ -8,13 +8,14 @@ require_relative 'wayback_machine_downloader/to_regex'
|
|
|
8
8
|
|
|
9
9
|
class WaybackMachineDownloader
|
|
10
10
|
|
|
11
|
-
VERSION = "0.
|
|
11
|
+
VERSION = "0.4.0"
|
|
12
12
|
|
|
13
|
-
attr_accessor :base_url, :
|
|
13
|
+
attr_accessor :base_url, :from_timestamp, :to_timestamp, :only_filter, :exclude_filter
|
|
14
14
|
|
|
15
15
|
def initialize params
|
|
16
16
|
@base_url = params[:base_url]
|
|
17
|
-
@
|
|
17
|
+
@from_timestamp = params[:from_timestamp].to_i
|
|
18
|
+
@to_timestamp = params[:to_timestamp].to_i
|
|
18
19
|
@only_filter = params[:only_filter]
|
|
19
20
|
@exclude_filter = params[:exclude_filter]
|
|
20
21
|
end
|
|
@@ -54,20 +55,27 @@ class WaybackMachineDownloader
|
|
|
54
55
|
end
|
|
55
56
|
|
|
56
57
|
def get_file_list_curated
|
|
57
|
-
|
|
58
|
-
|
|
58
|
+
parameters_for_wayback_machine_api = "&fl=timestamp,original&fastLatest=true&filter=statuscode:200&collapse=original"
|
|
59
|
+
if @from_timestamp and @from_timestamp != 0
|
|
60
|
+
parameters_for_wayback_machine_api += "&from=" + @from_timestamp.to_s
|
|
61
|
+
end
|
|
62
|
+
if @to_timestamp and @to_timestamp != 0
|
|
63
|
+
parameters_for_wayback_machine_api += "&to=" + @to_timestamp.to_s
|
|
64
|
+
end
|
|
65
|
+
index_file_list_raw = open ("http://web.archive.org/cdx/search/xd?url=#{@base_url}" + parameters_for_wayback_machine_api)
|
|
66
|
+
all_file_list_raw = open ("http://web.archive.org/cdx/search/xd?url=#{@base_url}/*" + parameters_for_wayback_machine_api)
|
|
59
67
|
file_list_curated = Hash.new
|
|
60
68
|
[index_file_list_raw, all_file_list_raw].each do |file|
|
|
61
69
|
file.each_line do |line|
|
|
62
70
|
line = line.split(' ')
|
|
63
|
-
file_timestamp = line[
|
|
64
|
-
file_url = line[
|
|
71
|
+
file_timestamp = line[0].to_i
|
|
72
|
+
file_url = line[1]
|
|
65
73
|
file_id = file_url.split('/')[3..-1].join('/')
|
|
66
74
|
file_id = CGI::unescape file_id
|
|
67
75
|
file_id = file_id.tidy_bytes unless file_id == ""
|
|
68
76
|
if file_id.nil?
|
|
69
77
|
puts "Malformed file url, ignoring: #{file_url}"
|
|
70
|
-
|
|
78
|
+
else
|
|
71
79
|
if match_exclude_filter(file_url)
|
|
72
80
|
puts "File url matches exclude filter, ignoring: #{file_url}"
|
|
73
81
|
elsif not match_only_filter(file_url)
|