whois-parser-next 0.4.5 → 0.4.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1006a5373a7bed9d074cee1bc20d02bd78d9bd7a682d740c78c5fa6cf44aaf1f
-  data.tar.gz: 99899f0e0a3db2b665119c876d4c561b11deaca2de4172a89bbdecebbb9f4b0d
+  metadata.gz: bf39e6c3afbb1b6a0cb49c48cb3fb11cf6db7d0162aaba77c9027ca4d2fadc3f
+  data.tar.gz: 1134d08ab3b1bcef0551f8008c366234805614e86cbe02ffe90501f5203dae84
 SHA512:
-  metadata.gz: 0e32b486fb6c61548fe11307395704a4548ac68ac2187d7dd95f42dc5a9ecee9147466f83639ab9ea2637732f771f38ef2c1331a8cc3514cc386e8228ac80afc
-  data.tar.gz: 4939cb1e255090dc84e95a824df8db82cdd11eceda1625a1caf3a465f12c8dc835afdcb023dea059cc874bbf4ea938154d823128951601dedd1b7f3694a3b972
+  metadata.gz: addd71819acd60b21142b60e419625ac8ebbed5b160b33e6cfe5562bcd064727e04b98c820f53bb98c1e1649158829b10b10d3c9ae7b900b47b25f57c1023043
+  data.tar.gz: 56a6dde832f4f0a9990e7753ea7f709a2bcba56d461d4eac0aa299bdd69b5896fd3b584647b6766267e621c4f6d52e42961a1fca1521c619b086f96f2dc80fcf

data/CHANGELOG.md

@@ -2,6 +2,58 @@
 
 This project uses [Semantic Versioning 2.0.0](http://semver.org/).
 
+#### Release 0.4.7 (2026-10-19)
+
+- ADDED: Added strict parsing for the official `.gt` and `.com.gt` web WHOIS
+  response. Registration requires one result heading with a single valid domain
+  and `ACTIVO`; absence requires one `alert alert-danger` result with the same
+  domain and the registry's exact `no está registrado` marker. Duplicate,
+  contradictory, malformed, mismatched, incomplete, denied, throttled, or
+  challenge responses remain non-positive. The parser ignores scripts, styles,
+  comments, and unrelated page content; contacts and dates are not parsed.
+- EVIDENCE: The official UVG registry lookup is
+  [`whois.php?dn=...`](https://www.gt/sitio/whois.php?dn={domain}). Bounded
+  2026-10-19 observations returned registered results for `google.gt` and
+  `google.com.gt`, and exact absence results for generated eight-letter names
+  under both suffixes. The registry's
+  [policy](https://www.gt/sitio/registration_policy.php?lang=en) limits lookup
+  to individual names rather than lists of registered names. Regression
+  examples use only minimal synthetic HTML. This is a web WHOIS source; no
+  TCP/43 or RDAP service is claimed. See the
+  [dated evidence report](docs/audits/guatemala-web-whois-2026-10-19.md).
+
+#### Release 0.4.6 (2026-10-19)
+
+- ADDED: Added `.ga` parsing for ANINF's current French WHOIS response, requiring
+  one valid domain, `Statut: actif`, and a parseable creation date; absence
+  requires the exact `%% NOT FOUND` marker. Added `.jo` parsing for a single
+  valid domain with a creation date and `Domain Status: Active`; absence
+  requires the exact quoted-name `No match for domain` response and matching
+  echoed domain.
+- UPDATED: `.sx` recognizes the current CIRA EPP statuses
+  `clientDeleteProhibited` and `clientTransferProhibited`, plus the exact
+  `Not found: <name>` absence response, while retaining its legacy format.
+  Duplicate, contradictory, malformed, or incomplete evidence remains
+  unknown; explicit denials and throttling remain blocked.
+- SAFETY: `.il` no longer treats a missing status as absence. Production
+  registered and absent queries returned the same 790-byte disclaimer-only
+  reply without an absence marker. Absence now requires the exact
+  `% No data was found to match the request criteria.` marker; registration
+  requires one valid domain and one `Transfer Locked` or `Transfer Allowed`
+  status. Empty, denied, throttled, duplicate, mixed, or incomplete replies
+  remain unknown. This corrects false absence; `.il` coverage remains
+  unverified.
+- EVIDENCE: Bounded production-host and managed-proxy checks on 2026-10-19 UTC
+  observed registered and absent replies for all three hosts. IANA lists
+  `whois.nic.net.jo` for [`.jo`](https://www.iana.org/domains/root/db/jo.html)
+  and `whois.sx` for [`.sx`](https://www.iana.org/domains/root/db/sx.html).
+  IANA has no WHOIS listing for `.ga`; ANINF's current [official
+  webWHOIS](https://whois.nic.ga/) and [registration
+  policy](https://www.mon.ga/policy/registry_policies.pdf) establish its provenance.
+  The `.ga` and `.sx` source terms restrict circulation or use of WHOIS data,
+  so the repository fixtures are synthetic technical-shape controls, not
+  copies of the captured replies. See the fixture README for details.
+
 #### Release 0.4.5 (2026-10-18)
 
 - ADDED: Added current CoCCA response parsing for `.bf`, `.td`, and `.mr`,

data/lib/whois/parser/version.rb

@@ -1,6 +1,6 @@
 module Whois
   class Parser
     # The current library version.
-    VERSION = "0.4.5".freeze
+    VERSION = "0.4.7".freeze
   end
 end

data/lib/whois/parsers/whois.isoc.org.il.rb

@@ -8,6 +8,7 @@
 
 
 require_relative 'base'
+require_relative 'registry_response_safety'
 
 
 module Whois
@@ -25,19 +26,20 @@ module Whois
     # and examples.
     #
     class WhoisIsocOrgIl < Base
+      include RegistryResponseSafety
 
       property_supported :status do
-        if content_for_scanner =~ /status:\s+(.*?)\n/
-          case ::Regexp.last_match(1).downcase
-          when "transfer locked"
-            :registered
-          when "transfer allowed"
-            :registered
-          else
-            Whois::Parser.bug!(ParserError, "Unknown status `#{::Regexp.last_match(1)}'.")
-          end
-        else
+        names = content_for_scanner.scan(/^domain:[ \t]*(.+)$/i).flatten.map(&:strip)
+        states = content_for_scanner.scan(/^status:[ \t]*(.+)$/i).flatten.map { |value| value.strip.downcase }
+        missing = content_for_scanner.scan(/^% No data was found to match the request criteria\.[ \t]*$/).size
+        if missing == 1 && names.empty? && states.empty?
           :available
+        elsif missing.zero? && names.size == 1 && states.size == 1 &&
+              names.first.match?(/\A(?:[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.)+il\z/i) &&
+              ['transfer locked', 'transfer allowed'].include?(states.first)
+          :registered
+        else
+          :unknown
         end
       end
 
@@ -46,7 +48,7 @@ module Whois
       end
 
       property_supported :registered? do
-        !available?
+        status == :registered
       end
 
 

data/lib/whois/parsers/whois.nic.ga.rb

@@ -0,0 +1,64 @@
+# frozen_string_literal: true
+
+require_relative 'base'
+require_relative 'registry_response_safety'
+
+module Whois
+  class Parsers
+    # ANINF's current French-language registry response, observed 2026-10-19.
+    class WhoisNicGa < Base
+      include RegistryResponseSafety
+
+      property_supported(:domain) { header_field('Nom de domaine')&.downcase }
+      property_supported(:domain_id) { header_field('Domain ID') }
+      property_supported(:status) do
+        names = content_for_scanner.scan(/^Nom de domaine:[ \t]*(.+)$/).flatten
+        missing = content_for_scanner.scan(/^%% NOT FOUND[ \t]*$/).size
+        if missing == 1 && names.empty? && !header_field('Domain ID')
+          :available
+        elsif missing.zero? && names.size == 1 &&
+              names.first.strip.match?(/\A[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.ga\z/i) &&
+              header_field('Domain ID') && header_field('Statut') == 'actif' &&
+              valid_creation_date?
+          :registered
+        else
+          :unknown
+        end
+      end
+      property_supported(:available?) { status == :available }
+      property_supported(:registered?) { status == :registered }
+      property_supported(:created_on) { header_date('Date de création') }
+      property_supported(:updated_on) { header_date('Dernière modification') }
+      property_supported(:expires_on) { header_date("Date d'expiration") }
+      property_supported(:registrar) do
+        name = header_field('Registrar')
+        Parser::Registrar.new(name: name) if name
+      end
+      property_supported(:nameservers) do
+        content_for_scanner.scan(/^Serveur de noms:[ \t]*(\S+)[ \t]*$/).flatten.map do |name|
+          Parser::Nameserver.new(name: name.downcase)
+        end
+      end
+
+      private
+
+      def header_field(label)
+        header = content_for_scanner.split(/^\[/, 2).first.to_s
+        values = header.scan(/^#{Regexp.escape(label)}:[ \t]*(.+)$/).flatten.map(&:strip)
+        values.first if values.size == 1 && !values.first.empty?
+      end
+
+      def header_date(label)
+        value = header_field(label)
+        parse_time(value) if value
+      end
+
+      def valid_creation_date?
+        value = header_field('Date de création')
+        value && value.match?(/\A\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}(?:\.\d+)?Z\z/) && Date.iso8601(value[0, 10]) && Time.iso8601(value)
+      rescue ArgumentError
+        false
+      end
+    end
+  end
+end

data/lib/whois/parsers/whois.nic.net.jo.rb

@@ -0,0 +1,41 @@
+# frozen_string_literal: true
+
+require_relative 'base_cocca2'
+require_relative 'registry_response_safety'
+require 'whois/scanners/whois.nic.net.jo'
+
+module Whois
+  class Parsers
+    # Jordan's official registry endpoint, observed 2026-10-19.
+    class WhoisNicNetJo < BaseCocca2
+      include RegistryResponseSafety
+      self.scanner = Scanners::WhoisNicNetJo
+
+      property_supported(:status) do
+        names = content_for_scanner.scan(/^Domain Name:[ \t]*(.+)$/i).flatten.map(&:strip)
+        absent = content_for_scanner.scan(/^No match for domain "([^"\n]+)"[ \t]*$/i).flatten
+        statuses = content_for_scanner.scan(/^Domain Status:[ \t]*(.+)$/i).flatten.map(&:strip)
+        valid_name = names.size == 1 && names.first.match?(/\A(?:[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.)+jo\z/i)
+        if valid_name && absent.size == 1 && absent.first.casecmp?(names.first) && statuses.empty? &&
+            !content_for_scanner.match?(/^Creation Date:/i)
+          :available
+        elsif valid_name && absent.empty? && statuses == ['Active'] && valid_creation_date?
+          :registered
+        else
+          :unknown
+        end
+      end
+      property_supported(:available?) { status == :available }
+      property_supported(:registered?) { status == :registered }
+
+      private
+
+      def valid_creation_date?
+        values = content_for_scanner.scan(/^Creation Date:[ \t]*(.+)$/i).flatten.map(&:strip)
+        values.size == 1 && values.first.match?(/\A\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z\z/) && Date.iso8601(values.first[0, 10]) && Time.iso8601(values.first)
+      rescue ArgumentError
+        false
+      end
+    end
+  end
+end

data/lib/whois/parsers/whois.sx.rb

@@ -8,6 +8,7 @@
 
 
 require_relative 'base'
+require_relative 'registry_response_safety'
 require 'whois/scanners/whois.sx.rb'
 
 
@@ -20,6 +21,7 @@ module Whois
     #   The Example parser for the list of all available methods.
     #
     class WhoisSx < Base
+      include RegistryResponseSafety
       include Scanners::Scannable
 
       self.scanner = Scanners::WhoisSx
@@ -31,7 +33,7 @@ module Whois
 
 
       property_supported :domain do
-        node("Domain Name", &:downcase)
+        (node("Domain Name") || node("Not found"))&.downcase
       end
 
       property_supported :domain_id do
@@ -40,6 +42,9 @@ module Whois
 
 
       property_supported :status do
+        current = current_registry_status
+        return current if current
+
         case (s = node("Domain Status", &:downcase))
         when "available"
           :available
@@ -103,6 +108,34 @@ module Whois
       end
 
 
+      # Keep the historical scanner for old fixtures, while recognizing the
+      # current EPP record and exact missing reply before scanning legacy text.
+      def current_registry_status
+        names = content_for_scanner.scan(/^Domain Name:[ \t]*(.+)$/i).flatten.map(&:strip)
+        absent = content_for_scanner.scan(/^Not found:[ \t]*(\S+)[ \t]*$/i).flatten
+        return nil if absent.empty? && !content_for_scanner.match?(/^Registry Domain ID:/i)
+
+        valid = ->(name) { name.match?(/\A[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.sx\z/i) }
+        if absent.size == 1 && names.empty? && valid.call(absent.first) &&
+            !content_for_scanner.match?(/^(?:Creation Date|Registry Domain ID|Domain Status):/i)
+          return :available
+        end
+        return :unknown unless absent.empty? && names.size == 1 && valid.call(names.first)
+
+        statuses = content_for_scanner.scan(/^Domain Status:[ \t]*(\S+)(?:[^\n]*)$/i).flatten
+        dates = content_for_scanner.scan(/^Creation Date:[ \t]*(.+)$/i).flatten.map(&:strip)
+        return :unknown unless dates.size == 1 && dates.first.match?(/\A\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z\z/)
+        Date.iso8601(dates.first[0, 10])
+        Time.iso8601(dates.first)
+        if statuses.any? && statuses.all? { |value| value.match?(/\A(?:ok|active|(?:client|server)(?:Delete|Transfer|Update|Renew)Prohibited)\z/) }
+          :registered
+        else
+          :unknown
+        end
+      rescue ArgumentError
+        :unknown
+      end
+
       private
 
       def build_contact(element, type)

data/lib/whois/parsers/www.gt.rb

@@ -0,0 +1,107 @@
+# frozen_string_literal: true
+
+require 'cgi'
+require 'strscan'
+require_relative 'base'
+require_relative 'registry_response_safety'
+
+module Whois
+  class Parsers
+    # The registry's per-domain public web WHOIS, observed 2026-10-19.
+    # Inspect only its result heading/alert, never navigation or script text.
+    class WwwGt < Base
+      include RegistryResponseSafety
+
+      property_supported(:status) { web_result[:status] }
+      property_supported(:domain) { web_result[:domain] }
+      property_supported(:registered?) { status == :registered }
+      property_supported(:available?) { status == :available }
+
+      def response_unavailable?
+        super || content_for_scanner.match?(/<cap-widget\b|\bclass=["']g-recaptcha["']|cf-chl-/i)
+      end
+
+      private
+
+      def web_result
+        @web_result ||= classify_web_result
+      end
+
+      def classify_web_result
+        blocks = result_blocks
+        return { status: :unknown } unless blocks
+
+        candidates = blocks.filter_map do |kind, text|
+          text = CGI.unescapeHTML(text.gsub('&aacute;', 'á').gsub('&Aacute;', 'Á')).gsub(/[ \t\r\n]+/, ' ').strip
+          if kind == 'h3' && (match = /\A(\S+)\. ACTIVO\z/i.match(text))
+            [match[1].downcase, :registered]
+          elsif kind == 'div' && (match = /\A(\S+)\. no está registrado \.\z/i.match(text))
+            [match[1].downcase, :available]
+          end
+        end
+        return { status: :unknown } unless candidates.size == 1
+
+        name, state = candidates.first
+        return { status: :unknown } unless name.bytesize <= 253 &&
+          name.match?(/\A(?:[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\.)+gt\z/)
+
+        { domain: name, status: state }
+      end
+
+      # One forward scan, including malformed tags. Limit each tag before
+      # inspecting attributes and skip scripts, styles and comments entirely.
+      def result_blocks
+        input = StringScanner.new(content_for_scanner)
+        blocks = []
+        active = nil
+        depth = 0
+        text = +''
+        until input.eos?
+          chunk = input.scan_until(/</)
+          unless chunk
+            text << input.rest if active
+            input.terminate
+            break
+          end
+          text << chunk[0...-1] if active
+          if input.peek(3) == '!--'
+            return nil unless input.scan_until(/-->/)
+            next
+          end
+          tag = input.scan_until(/>/)
+          return nil unless tag && tag.bytesize <= 16_384
+          match = /\A[ \t]*(\/?)[ \t]*([a-z0-9]+)/i.match(tag)
+          next unless match
+          closing = match[1] == '/'
+          name = match[2].downcase
+          if !closing && %w[script style].include?(name)
+            return nil if active
+            return nil unless input.scan_until(%r{</#{name}[ \t\r\n]*>}i)
+            next
+          end
+          if active
+            depth += 1 if !closing && name == active
+            depth -= 1 if closing && name == active
+            if depth.zero?
+              blocks << [active, text]
+              active = nil
+              text = +''
+            else
+              text << ' '
+            end
+          elsif !closing && (name == 'h3' || (name == 'div' && alert_classes?(tag)))
+            active = name
+            depth = 1
+          end
+        end
+        active ? nil : blocks
+      end
+
+      def alert_classes?(tag)
+        value = tag[/\bclass[ \t]*=[ \t]*["']([^"']*)["']/i, 1]
+        classes = value.to_s.split
+        classes.include?('alert') && classes.include?('alert-danger')
+      end
+    end
+  end
+end

data/lib/whois/scanners/whois.nic.net.jo.rb

@@ -0,0 +1,15 @@
+# frozen_string_literal: true
+
+require_relative 'base_cocca2'
+
+module Whois
+  module Scanners
+    class WhoisNicNetJo < BaseCocca2
+      self.tokenizers += [:skip_jordan_absence]
+
+      tokenizer :skip_jordan_absence do
+        @input.skip(/^No match for domain "[^"\n]+"[ \t]*\n/i)
+      end
+    end
+  end
+end

data/lib/whois/scanners/whois.sx.rb

@@ -11,6 +11,7 @@ module Whois
           :scan_keyvalue,
           :skip_lastupdate,
           :scan_disclaimer,
+          :skip_comment,
       ]
 
 
@@ -25,6 +26,10 @@ module Whois
         @input.skip(/>>>(.+?)<<<\n/)
       end
 
+      tokenizer :skip_comment do
+        @input.skip(/^%[^\n]*\n/)
+      end
+
       tokenizer :scan_disclaimer do
         if @input.match?(/^% WHOIS LEGAL STATEMENT/)
           @ast["field:disclaimer"] = _scan_lines_to_array(/%(.*)\n/).map(&:strip).join("\n")

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: whois-parser-next
 version: !ruby/object:Gem::Version
-  version: 0.4.5
+  version: 0.4.7
 platform: ruby
 authors:
 - Simone Carletti
@@ -439,6 +439,7 @@ files:
 - lib/whois/parsers/whois.nic.furniture.rb
 - lib/whois/parsers/whois.nic.futbol.rb
 - lib/whois/parsers/whois.nic.fyi.rb
+- lib/whois/parsers/whois.nic.ga.rb
 - lib/whois/parsers/whois.nic.gal.rb
 - lib/whois/parsers/whois.nic.games.rb
 - lib/whois/parsers/whois.nic.gd.rb
@@ -537,6 +538,7 @@ files:
 - lib/whois/parsers/whois.nic.mx.rb
 - lib/whois/parsers/whois.nic.name.rb
 - lib/whois/parsers/whois.nic.net.bw.rb
+- lib/whois/parsers/whois.nic.net.jo.rb
 - lib/whois/parsers/whois.nic.net.ng.rb
 - lib/whois/parsers/whois.nic.net.sa.rb
 - lib/whois/parsers/whois.nic.net.sb.rb
@@ -751,6 +753,7 @@ files:
 - lib/whois/parsers/whois.za.org.rb
 - lib/whois/parsers/whois.zicta.zm.rb
 - lib/whois/parsers/whois1.nic.bi.rb
+- lib/whois/parsers/www.gt.rb
 - lib/whois/parsers/za_central_registry.rb
 - lib/whois/safe_record.rb
 - lib/whois/scanners/base.rb
@@ -778,6 +781,7 @@ files:
 - lib/whois/scanners/whois.nc.rb
 - lib/whois/scanners/whois.nic.cz.rb
 - lib/whois/scanners/whois.nic.it.rb
+- lib/whois/scanners/whois.nic.net.jo.rb
 - lib/whois/scanners/whois.pir.org.rb
 - lib/whois/scanners/whois.rnids.rs.rb
 - lib/whois/scanners/whois.smallregistry.net.rb