bots 1.0.7 → 1.0.9
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/browser.rb +62 -0
- metadata +3 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 63a5cef7be688d0bd777c7fde8fe8f52dd9a6789106b7691c2b6c3bfc8300144
|
4
|
+
data.tar.gz: b6e1e2df2d171bc4bae8fe50652221551d27b76b6945b76634c612fb63e37ba3
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: d0288d8e1195c903dabe6560ac1ab2482898c46ed91326ed89dbf878663c459f3fbeb043b2b03eaf2aec71dfaa387c5007fec656dc2310b93f1cd5a0eb582560
|
7
|
+
data.tar.gz: c1db09ced8386a9c8f341182f62c672110c211973fa6cd409a49de7a62d6cc53278cdccd42dc0497569fd5c5dc83ca2086f0a56e7ad43e91122ccf6af86f19b2
|
data/lib/browser.rb
ADDED
@@ -0,0 +1,62 @@
|
|
1
|
+
module BlackStack
|
2
|
+
module Bots
|
3
|
+
class Browser < Watir::Browser
|
4
|
+
LOCKFILENAME = '/tmp/blackstack.bots.browser.lock'
|
5
|
+
attr_accessor :pids, :lockfile
|
6
|
+
|
7
|
+
def initialize()
|
8
|
+
self.lockfile = File.open(LOCKFILENAME, 'w+')
|
9
|
+
|
10
|
+
n = 30 # timeout in seconds
|
11
|
+
|
12
|
+
# wait the lock file /tmp/blackstack.bots.browser.lock
|
13
|
+
self.lockfile.flock(File::LOCK_EX)
|
14
|
+
begin
|
15
|
+
# get list of PID of all opened chrome browsers, before launching this one
|
16
|
+
pids_before = `pgrep -f chrome`.split("\n")
|
17
|
+
# track # of chrome processes
|
18
|
+
#print "(#{pids_before.size})"
|
19
|
+
# setup driver
|
20
|
+
client = Selenium::WebDriver::Remote::Http::Default.new
|
21
|
+
begin
|
22
|
+
client.read_timeout = n # for newest selenium webdriver version
|
23
|
+
rescue => e
|
24
|
+
client.timeout = n # deprecated in newest selenium webdriver version
|
25
|
+
end
|
26
|
+
options = Selenium::WebDriver::Chrome::Options.new
|
27
|
+
options.add_argument('--headless')
|
28
|
+
|
29
|
+
# Add this parameter to run Chrome from a root user.
|
30
|
+
# https://stackoverflow.com/questions/50642308/webdriverexception-unknown-error-devtoolsactiveport-file-doesnt-exist-while-t
|
31
|
+
options.add_argument('--no-sandbox')
|
32
|
+
|
33
|
+
driver = Selenium::WebDriver.for :chrome, :options => options, http_client: client
|
34
|
+
# create the browser
|
35
|
+
super(driver)
|
36
|
+
# get list of PID of all opened chrome browsers, after launching this one
|
37
|
+
pids_after = `pgrep -f chrome`.split("\n")
|
38
|
+
# store list of PID regarding this launched browser
|
39
|
+
self.pids = pids_after - pids_before
|
40
|
+
rescue => e
|
41
|
+
# unlock the file
|
42
|
+
self.lockfile.flock(File::LOCK_UN)
|
43
|
+
# raise the error
|
44
|
+
raise e
|
45
|
+
ensure
|
46
|
+
# unlock the file
|
47
|
+
self.lockfile.flock(File::LOCK_UN)
|
48
|
+
end
|
49
|
+
end # initialize
|
50
|
+
|
51
|
+
def close
|
52
|
+
# kill all pids regarding this browser
|
53
|
+
self.pids.each { |pid| `kill -9 #{pid} >/dev/null 2>&1` }
|
54
|
+
end # close
|
55
|
+
|
56
|
+
def quit
|
57
|
+
# kill all pids regarding this browser
|
58
|
+
self.pids.each { |pid| `kill -9 #{pid} >/dev/null 2>&1` }
|
59
|
+
end # quit
|
60
|
+
end # Browser
|
61
|
+
end # Bots
|
62
|
+
end # BlackStack
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: bots
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.0.
|
4
|
+
version: 1.0.9
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Leandro Daniel Sardi
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2023-08-
|
11
|
+
date: 2023-08-25 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: simple_cloud_logging
|
@@ -218,6 +218,7 @@ extra_rdoc_files: []
|
|
218
218
|
files:
|
219
219
|
- lib/base.rb
|
220
220
|
- lib/bots.rb
|
221
|
+
- lib/browser.rb
|
221
222
|
- lib/google.rb
|
222
223
|
- lib/indeed.rb
|
223
224
|
- lib/scraper.rb
|