columbo 0.1.3 → 0.1.4
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +1 -1
- data/lib/columbo/capture.rb +7 -4
- data/lib/columbo/inspector.rb +3 -1
- data/lib/columbo/version.rb +1 -1
- metadata +2 -2
data/README.md
CHANGED
@@ -45,7 +45,7 @@ Check the Rack configuration:
|
|
45
45
|
|
46
46
|
## Disclaimer
|
47
47
|
|
48
|
-
This is an alpha release
|
48
|
+
This is an alpha release, it is untested with Sinatra and Rails 3 only.
|
49
49
|
UI to explore sessions will be completed later (ETA: 2013'Q2).
|
50
50
|
|
51
51
|
## Author
|
data/lib/columbo/capture.rb
CHANGED
@@ -10,9 +10,12 @@ module Columbo
|
|
10
10
|
|
11
11
|
def initialize(app, opts={})
|
12
12
|
@app = app
|
13
|
-
|
14
|
-
@
|
15
|
-
@
|
13
|
+
# Options
|
14
|
+
@capture = opts[:capture]
|
15
|
+
@bench = opts[:capture] && opts[:bench]
|
16
|
+
@capture_crawlers = opts[:capture_crawlers]
|
17
|
+
@crawlers = opts[:crawlers] || "(Baidu|Gigabot|Googlebot|libwww-perl|lwp-trivial|msnbot|SiteUptime|Slurp|WordPress|ZIBB|ZyBorg|bot|crawler|spider|robot|crawling|facebook|w3c|coccoc)"
|
18
|
+
@mongo_uri = opts[:mongo_uri]
|
16
19
|
|
17
20
|
Columbo.logger = opts[:logger] if opts[:logger]
|
18
21
|
|
@@ -43,7 +46,7 @@ module Columbo
|
|
43
46
|
|
44
47
|
Thread.new do
|
45
48
|
begin
|
46
|
-
@inspector.investigate env, status, headers, response, start_processing, stop_processing
|
49
|
+
@inspector.investigate env, status, headers, response, start_processing, stop_processing, @crawlers, @capture_crawlers
|
47
50
|
rescue Exception => e
|
48
51
|
log_error env, e
|
49
52
|
end
|
data/lib/columbo/inspector.rb
CHANGED
@@ -10,11 +10,13 @@ module Columbo
|
|
10
10
|
@mongo_uri = mongo_uri
|
11
11
|
end
|
12
12
|
|
13
|
-
def investigate(env, status, headers, body, start, stop)
|
13
|
+
def investigate(env, status, headers, body, start, stop, crawlers, capture_crawlers)
|
14
14
|
# Lazy connection to MongoDB
|
15
15
|
client = Columbo::DbClient.new @mongo_uri
|
16
16
|
# Normalise request from env
|
17
17
|
request = Rack::Request.new(env)
|
18
|
+
rg = Regexp.new(crawlers, Regexp::IGNORECASE)
|
19
|
+
return if request.user_agent.match(rg) && !capture_crawlers
|
18
20
|
html = ''
|
19
21
|
body.each { |part| html += part }
|
20
22
|
# Retrieve plain text body for full text search
|
data/lib/columbo/version.rb
CHANGED
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: columbo
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.1.
|
4
|
+
version: 0.1.4
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -9,7 +9,7 @@ authors:
|
|
9
9
|
autorequire:
|
10
10
|
bindir: bin
|
11
11
|
cert_chain: []
|
12
|
-
date: 2013-
|
12
|
+
date: 2013-04-08 00:00:00.000000000 Z
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: rack
|