datahen 0.14.16 → 0.14.17
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/datahen/cli/scraper_page.rb +31 -17
- data/lib/datahen/client/job_page.rb +5 -0
- data/lib/datahen/client/scraper_job_page.rb +6 -1
- data/lib/datahen/version.rb +1 -1
- metadata +3 -3
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: a3cf1bf50610ab9ba523ca8f9ae1fdb307831cb56018036076f31353b357edfa
|
4
|
+
data.tar.gz: '0199ea98a2f171675168699adc523932418b9822ff156c71786458c6362b6cdb'
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 8d66226573dbd9bd3ef795ce021eb9ee202b21cce1d0b79211093bb9abda6f1982a721bf8ae1fbc9bf84aae6e50d1878bf9d389ee67d685e4d1c440fd88cceeb
|
7
|
+
data.tar.gz: 6ea1a7748cf77ae1cb8f750da71df86120f68e671c84e535e8536ee3324ed71026034a042662d081108ff7b72ccb3f0ef3ab3f049bdb45c92fc0e86ba2e10d46
|
@@ -111,6 +111,7 @@ module Datahen
|
|
111
111
|
puts "Must specify either a --gid, --fetch-fail, --parse-fail or --status"
|
112
112
|
return
|
113
113
|
end
|
114
|
+
|
114
115
|
if options[:job]
|
115
116
|
client = Client::JobPage.new(options)
|
116
117
|
puts "#{client.refetch(options[:job])}"
|
@@ -129,26 +130,39 @@ module Datahen
|
|
129
130
|
option :status, type: :string, desc: 'Reparse only pages with a specific status.'
|
130
131
|
option :job, :aliases => :j, type: :numeric, desc: 'Set a specific job ID'
|
131
132
|
def reparse(scraper_name)
|
132
|
-
|
133
|
-
|
133
|
+
if !options.key?(:gid) && !options.key?(:parse_fail) && !options.key?(:status)
|
134
|
+
puts "Must specify either a --gid, --parse-fail or --status"
|
135
|
+
return
|
136
|
+
end
|
134
137
|
|
135
|
-
|
136
|
-
|
137
|
-
|
138
|
-
|
138
|
+
if options[:job]
|
139
|
+
client = Client::JobPage.new(options)
|
140
|
+
puts "#{client.reparse(options[:job])}"
|
141
|
+
else
|
142
|
+
client = Client::ScraperJobPage.new(options)
|
143
|
+
puts "#{client.reparse(scraper_name)}"
|
144
|
+
end
|
145
|
+
end
|
139
146
|
|
140
|
-
|
141
|
-
|
142
|
-
|
143
|
-
|
144
|
-
|
145
|
-
|
146
|
-
|
147
|
+
desc "limbo <scraper_name>", "Move pages on a scraper's current job to limbo"
|
148
|
+
long_desc <<-LONGDESC
|
149
|
+
Move pages in a scraper's current job to limbo. You need to specify either a --gid or --status.\x5
|
150
|
+
LONGDESC
|
151
|
+
option :gid, :aliases => :g, type: :string, desc: 'Move a specific GID to limbo'
|
152
|
+
option :status, type: :string, desc: 'Move pages with a specific status to limbo.'
|
153
|
+
option :job, :aliases => :j, type: :numeric, desc: 'Set a specific job ID'
|
154
|
+
def limbo(scraper_name)
|
155
|
+
if !options.key?(:gid) && !options.key?(:status)
|
156
|
+
puts "Must specify either a --gid or --status"
|
157
|
+
return
|
158
|
+
end
|
147
159
|
|
148
|
-
|
149
|
-
|
150
|
-
|
151
|
-
|
160
|
+
if options[:job]
|
161
|
+
client = Client::JobPage.new(options)
|
162
|
+
puts "#{client.limbo(options[:job])}"
|
163
|
+
else
|
164
|
+
client = Client::ScraperJobPage.new(options)
|
165
|
+
puts "#{client.limbo(scraper_name)}"
|
152
166
|
end
|
153
167
|
end
|
154
168
|
|
@@ -72,6 +72,11 @@ module Datahen
|
|
72
72
|
params = @options.merge(opts)
|
73
73
|
self.class.put("/jobs/#{job_id}/pages/refetch", params)
|
74
74
|
end
|
75
|
+
|
76
|
+
def limbo(job_id, opts={})
|
77
|
+
params = @options.merge(opts)
|
78
|
+
self.class.put("/jobs/#{job_id}/pages/limbo", params)
|
79
|
+
end
|
75
80
|
end
|
76
81
|
end
|
77
82
|
end
|
@@ -26,7 +26,7 @@ module Datahen
|
|
26
26
|
self.class.put("/scrapers/#{scraper_name}/current_job/pages/refetch", params)
|
27
27
|
end
|
28
28
|
|
29
|
-
# Deprecated, please use Datahen::Client::
|
29
|
+
# Deprecated, please use Datahen::Client::JobPage#refetch instead.
|
30
30
|
#
|
31
31
|
# @note This method will be removed at some point in the future.
|
32
32
|
def refetch_by_job(job_id, opts={})
|
@@ -39,6 +39,11 @@ module Datahen
|
|
39
39
|
self.class.put("/scrapers/#{scraper_name}/current_job/pages/reparse", params)
|
40
40
|
end
|
41
41
|
|
42
|
+
def limbo(scraper_name, opts={})
|
43
|
+
params = @options.merge(opts)
|
44
|
+
self.class.put("/scrapers/#{scraper_name}/current_job/pages/limbo", params)
|
45
|
+
end
|
46
|
+
|
42
47
|
def enqueue(scraper_name, method, url, opts={})
|
43
48
|
body = {}
|
44
49
|
body[:method] = method != "" ? method : "GET"
|
data/lib/datahen/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: datahen
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.14.
|
4
|
+
version: 0.14.17
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Parama Danoesubroto
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2020-10-
|
11
|
+
date: 2020-10-20 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: thor
|
@@ -266,7 +266,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
266
266
|
- !ruby/object:Gem::Version
|
267
267
|
version: '0'
|
268
268
|
requirements: []
|
269
|
-
rubygems_version: 3.
|
269
|
+
rubygems_version: 3.0.3
|
270
270
|
signing_key:
|
271
271
|
specification_version: 4
|
272
272
|
summary: DataHen toolbelt for developers
|