RubyGems - grell - Versions diffs - 1.4.0 → 1.5.0 - Mend

grell 1.4.0 → 1.5.0

Files changed (7) hide show

checksums.yaml CHANGED

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 66dc6a02c6a3ff7c79e01fd7e84c21644ef5eb3a
-  data.tar.gz: 8ee3c5702299dfe64c2bd2732d2795014080e0bf
+  metadata.gz: 61bf72903f19172767e0702e86b81539d5da4b7d
+  data.tar.gz: 5fcc616f9002fc46dbecc3070b63f4ba58acae49
 SHA512:
-  metadata.gz: 6c191dd4ec5a994d2963d5c7db5f0ebb6f36d0532a05ed02e4153c6d04fb9e6022f3a63065da6e07208e0389c2f8f0e7c008ab763c2a76bd986c13407dcd2740
-  data.tar.gz: 3b05002063f76bbc0916684c352554eddf75ce88c6c294c067fc91e4973a1cb10a7f2465ea874d1aa8c162ee3a6612d9bb5289b7489edab9f0dcfc5ae79dc463
+  metadata.gz: c74181c32f2aab10655dff0023a271e47671b11aa2b81095a384ff2fb75425621ef17e3b0eb92d16d45a2575b55f1a676983c09584e6f557de2cae2a2e3f534b
+  data.tar.gz: cf699863fd4ddf88e83db631d2346ef83c6cb56d4d717eefe2852da8a7b16933b1e07884caf928e6ede6e6db097ef40c68e403edf540c44f414687c8d55ab40a

data/CHANGELOG.md CHANGED

@@ -1,3 +1,7 @@
+* Version 1.5.0
+  Grell will follow redirects.
+  Added #followed_redirects? #error? #current_url methods to the Page class
 * Version 1.4.0
   Added crawler.restart to restart browser process
   The block of code can make grell retry any given page.

data/lib/grell/page.rb CHANGED

@@ -48,10 +48,26 @@ module Grell
       unavailable_page(404, e)
     end
+    # Number of times we have retried the current page
     def retries
       [@times_visited -1, 0].max
     end
+    # The current URL, this may be different from the URL we asked for if there was some redirect
+    def current_url
+      @rawpage.current_url
+    end
+    # True if we followed a redirect to get the current contents
+    def followed_redirects?
+      current_url != @url
+    end
+    # True if there page responded with an error
+    def error?
+      !!(status.to_s =~ /[4|5]\d\d/)
+    end
     private
     def unavailable_page(status, exception)
       Grell.logger.warn "The page with the URL #{@url} was not available. Exception #{exception}"

data/lib/grell/rawpage.rb CHANGED

@@ -5,6 +5,7 @@ module Grell
     def navigate(url)
       visit(url)
+      follow_redirects!
     end
     def headers
@@ -33,5 +34,18 @@ module Grell
     def has_selector?(selector)
       page.has_selector?(selector)
     end
+    private
+    def follow_redirects!
+      # Phantom is very weird, it will follow a redirect to provide the correct body but will not fill the
+      # status and the headers, if we are in that situation, revisit the page with the correct url this time.
+      # Note that we will still fail if we have more than 5 redirects on a row
+      redirects = 0
+      while(page.status_code == nil && redirects < 5)
+        visit( CGI.unescape(page.current_url))
+        redirects = redirects + 1
+      end
+    end
   end
 end

data/lib/grell/version.rb CHANGED

@@ -1,3 +1,3 @@
 module Grell
-  VERSION = "1.4.0"
+  VERSION = "1.5.0"
 end

data/spec/lib/page_spec.rb CHANGED

@@ -86,7 +86,7 @@ RSpec.describe Grell::Page do
   end
   shared_examples_for 'an errored grell page' do
-    it 'returns empty status 404 page after navigating' do
+    it 'returns empty status 404 page after navigating' do
       expect(page.status).to eq(404)
       expect(page.links).to eq([])
       expect(page.headers).to eq(headers)
@@ -94,7 +94,7 @@ RSpec.describe Grell::Page do
       expect(page.has_selector?('html')).to eq(false)
       expect(page).to be_visited
       expect(page.timestamp).to eq(now)
-      #expect_any_instance_of(Logger).to receive(:warn) #.with(/The page with the URL #{url} was not available"/)
+      expect(page.error?).to eq(true)
     end
   end
@@ -153,6 +153,30 @@ RSpec.describe Grell::Page do
   end
+  context 'navigating to an URL with redirects, follows them transparently' do
+    let(:visited) {true}
+    let(:status) { 200}
+    let(:body) {'<html><head></head><body>nothing cool</body></html>'}
+    let(:links) {[]}
+    let(:expected_headers) {returned_headers}
+    let(:real_url) {'http://example.com/other'}
+    before do
+      proxy.stub(url).and_return(:redirect_to => real_url)
+      proxy.stub(real_url).and_return(body: body, code: status, headers: returned_headers.dup)
+      page.navigate
+    end
+    it_behaves_like 'a grell page'
+    it 'followed_redirects? is true' do
+      expect(page.followed_redirects?).to eq(true)
+    end
+    it 'current_url match the url we were redirected to' do
+      expect(page.current_url).to eq(real_url)
+    end
+  end
+  #Here also add examples that may happen for almost all pages (no errors, no redirects)
   context 'navigating to the URL we get page with no links' do
     let(:visited) {true}
     let(:status) { 200}
@@ -166,6 +190,18 @@ RSpec.describe Grell::Page do
     end
     it_behaves_like 'a grell page'
+    it 'followed_redirects is false' do
+      expect(page.followed_redirects?).to eq(false)
+    end
+    it 'current_url is url' do
+      expect(page.current_url).to eq(url)
+    end
+    it 'does not have errors' do
+      expect(page.error?).to eq(false)
+    end
   end
   context 'navigating to the URL we get page with links using a elements' do
@@ -238,7 +274,7 @@ RSpec.describe Grell::Page do
       </body></html>"
     end
     let(:links) do
-      ["http://www.example.com/trusmis.html", "http://www.example.com/help.html",
+      ["http://www.example.com/trusmis.html", "http://www.example.com/help.html",
        'http://www.example.com/more_help.html', 'http://www.example.com/help_me.html'
       ]
     end

metadata CHANGED

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: grell
 version: !ruby/object:Gem::Version
-  version: 1.4.0
+  version: 1.5.0
 platform: ruby
 authors:
 - Jordi Polo Carres
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2015-05-15 00:00:00.000000000 Z
+date: 2015-07-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: capybara