video_grabber 1.5.0 → 1.6.0
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/README.md +1 -1
- data/lib/video_grabber/config.rb +2 -1
- data/lib/video_grabber/scraper.rb +4 -4
- data/lib/video_grabber/version.rb +1 -1
- 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: 42b16033bdf111d37375a5253c02a89d04d5a93e
|
|
4
|
+
data.tar.gz: fb3cf3dbd626dd8c51e9391981b56598445d5f34
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 020ea57cb51f5b61abd8b64fb0a7ca7e6e290f0a50c33a801d3641be0b4953e03b7eede781ea87498f01afba6d3857600a49d5cdf52cf83521303f4a3401291e
|
|
7
|
+
data.tar.gz: 4e2677496204205b512e59d9209a50d39ca5e0ec92f49df5ceaae2959b2654e167632b5a5eb4db3881b4ab75b926b202f5d154ca885b39ac66fba0079c9bfe03
|
data/README.md
CHANGED
|
@@ -69,7 +69,7 @@ Or you can directly pass the param `keep_browser_open` during initialization.
|
|
|
69
69
|
- **timeout:** *(default: 60)* The timeout for the scraper. Will trigger a `VideoGrabber::Timeout` if the delay is met.
|
|
70
70
|
- **keep_browser_open** *(default: false)* If activated, will keep the scraper's browser open as long as you do not stop it (using the `stop` public method.).
|
|
71
71
|
- **headless_enabled** *(default: true)* If disabled, will open your Firefox browser to crawl your links.
|
|
72
|
-
- **
|
|
72
|
+
- **html_attributes** *(default: {controls: true})* This option enables you to pass html attributes that will be passed to your crawled links elements.
|
|
73
73
|
- **firefox_extension_path** If passed, your Scraper instance will run using the given extension (`.xpi` file). Useful if you want to benefit from an Adblocker for instance
|
|
74
74
|
|
|
75
75
|
## Versioning
|
data/lib/video_grabber/config.rb
CHANGED
|
@@ -23,12 +23,13 @@ module VideoGrabber
|
|
|
23
23
|
class Configuration
|
|
24
24
|
|
|
25
25
|
attr_accessor :url, :keep_browser_open, :timeout, :headless_enabled,
|
|
26
|
-
:firefox_extension_path, :
|
|
26
|
+
:firefox_extension_path, :html_attributes
|
|
27
27
|
|
|
28
28
|
def initialize
|
|
29
29
|
@keep_browser_open = false
|
|
30
30
|
@timeout = 60
|
|
31
31
|
@headless_enabled = true
|
|
32
|
+
@html_attributes = { controls: true }
|
|
32
33
|
end
|
|
33
34
|
end
|
|
34
35
|
end
|
|
@@ -2,7 +2,7 @@ module VideoGrabber
|
|
|
2
2
|
class Scraper
|
|
3
3
|
|
|
4
4
|
attr_reader :url, :browser, :timeout, :keep_browser_open, :headless_enabled,
|
|
5
|
-
:firefox_extension_path, :profile, :
|
|
5
|
+
:firefox_extension_path, :profile, :html_attributes
|
|
6
6
|
|
|
7
7
|
def initialize(config)
|
|
8
8
|
@keep_browser_open = config.keep_browser_open
|
|
@@ -10,7 +10,7 @@ module VideoGrabber
|
|
|
10
10
|
@timeout = config.timeout
|
|
11
11
|
@headless_enabled = config.headless_enabled
|
|
12
12
|
@firefox_extension_path = config.firefox_extension_path
|
|
13
|
-
@
|
|
13
|
+
@html_attributes = config.html_attributes
|
|
14
14
|
end
|
|
15
15
|
|
|
16
16
|
def start
|
|
@@ -74,13 +74,13 @@ module VideoGrabber
|
|
|
74
74
|
end
|
|
75
75
|
|
|
76
76
|
def add_attributes(list)
|
|
77
|
-
return unless
|
|
77
|
+
return unless html_attributes
|
|
78
78
|
|
|
79
79
|
list.map do |element|
|
|
80
80
|
|
|
81
81
|
parsed_element = Nokogiri::XML(element)
|
|
82
82
|
|
|
83
|
-
|
|
83
|
+
html_attributes.each do |key, value|
|
|
84
84
|
parsed_element.xpath('//video').first.set_attribute(key, value)
|
|
85
85
|
end
|
|
86
86
|
|