inspec-core 7.2.1 → 7.3.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ed63ed9529571a4192cc3d69253bb303cddb5f6c07dd7f6c5009846e20fc2d74
-  data.tar.gz: e3f693766b0698bba07fc6ae2e823a51993484d88cd5f309dd83b6dd832a55d4
+  metadata.gz: e7a56ab7ed289f14adf2957f8407eb84cc71cb3b49bea6328dde51f544b1cad7
+  data.tar.gz: 633ee36ea973a2dd2edef07122f345d9120dd748edf30d8b1bd0847b8dd4d272
 SHA512:
-  metadata.gz: 3c0bc69773b1376970e84b073b556731af530f4706779f5c5ccf3807a6ca5ba4e55a747f08581bb001ed796468108b0080df8197f8484d7a43594d3c4410e3c2
-  data.tar.gz: ee83f8a1efd8aca287f1c9fb426083cc1245f1174252747a1a88cb8c6c488d939b2797c415498a54bb4297249ca09ae43618f0a02e55c5fe7dcfee18cf821148
+  metadata.gz: fcd8febd0b3e6b19fac35314ffcb7847dc9ddb08e5add9d9a43cd59e5544d18b3969b8273e16554ddb477b58746870d89c0ba6ac67c777c3c0cc821ff2cca4d1
+  data.tar.gz: e6ad048189427c994b6bd967c3f1aa4690013204523cf591ad9f6d274d61da922c28a7d6c3e0049b259b97ecda90186edc6e814fd24bc6b60b6f7d1b2ee52a81

data/lib/inspec/archive/zip.rb

@@ -5,7 +5,7 @@ require "pathname" unless defined?(Pathname)
 module Inspec::Archive
   class ZipArchiveGenerator
     def archive(base_dir, files, archive)
-      Zip::File.open(archive, Zip::File::CREATE) do |zipfile|
+      Zip::File.open(archive, create: true) do |zipfile|
         files.each do |input_filename|
           zipfile.add(input_filename, Pathname.new(base_dir).join(input_filename))
         end

data/lib/inspec/file_provider.rb

@@ -171,7 +171,15 @@ module Inspec
           # extracted files do not conflict.
           FileUtils.remove_entry(final_path) if File.exist?(final_path)
 
-          archive.extract(file, final_path)
+          # rubyzip 3.x no longer creates missing intermediate directories by
+          # default, so do it ourselves.
+          FileUtils.mkdir_p(File.dirname(final_path))
+
+          # Passing an absolute entry_path alongside the default
+          # destination_directory ('.') makes rubyzip 3.x join them into a
+          # mangled, CWD-relative path. Pass destination_directory explicitly
+          # instead, and let the entry's own (relative) name resolve against it.
+          archive.extract(file, destination_directory: destination_path)
         end
       end
     end

data/lib/inspec/resources/oracledb_session.rb

@@ -66,11 +66,11 @@ module Inspec::Resources
       raise Inspec::Exceptions::ResourceFailed, "#{resource_exception_message}" if resource_failed?
 
       if @sqlcl_bin && inspec.command(@sqlcl_bin).exist?
-        @bin = @sqlcl_bin
+        @bin = "#{@sqlcl_bin} -s"
         format_options = "set sqlformat csv\nSET FEEDBACK OFF"
       else
         @bin = "#{@sqlplus_bin} -S"
-        format_options = "SET PAGESIZE 32000\nSET FEEDBACK OFF\nSET UNDERLINE OFF"
+        format_options = "SET MARKUP CSV ON\nSET PAGESIZE 32000\nSET FEEDBACK OFF\nSET UNDERLINE OFF"
       end
 
       command = command_builder(format_options, sql)
@@ -165,17 +165,21 @@ module Inspec::Resources
     end
 
     def parse_csv_result(stdout)
-      output = stdout.split("oracle_query_string")[-1]
-      # comma_query_sub replaces the csv delimiter "," in the output.
-      # Handles CSV parsing of data like this (DROP,3) etc
+      output = stdout.split("oracle_query_string")[-1].to_s.sub(/\r/, "").strip
+      return [] if output.empty?
 
-      output = output.sub(/\r/, "").strip.gsub(",", "comma_query_sub")
+      # sqlplus (SET MARKUP CSV / quoted output) and sqlcl (set sqlformat csv)
+      # both emit standard RFC 4180 CSV: a header row followed by data rows, with
+      # embedded commas/quotes/newlines quoted per the spec. Parse it directly so
+      # every column is addressable and commas inside quoted fields are preserved.
       converter = ->(header) { header.downcase }
-      CSV.parse(output, headers: true, header_converters: converter).map do |row|
-        next if row.entries.flatten.empty?
+      CSV.parse(output, headers: true, header_converters: converter).filter_map do |row|
+        hash = row.to_h
+        # Drop only truly empty parsed rows (no columns); keep rows whose values
+        # are nil/empty so callers can still index by row position.
+        next if hash.empty?
 
-        revised_row = row.entries.flatten.map { |entry| entry&.gsub("comma_query_sub", ",") }
-        Hashie::Mash.new([revised_row].to_h)
+        Hashie::Mash.new(hash)
       end
     end
 

data/lib/inspec/utils/telemetry/base.rb

@@ -59,7 +59,7 @@ module Inspec
                               os: safe_platform_field(train_platform, :name) || "unknown",
                               version: safe_platform_field(train_platform, :release) || "unknown",
                               architecture: safe_platform_field(train_platform, :arch) || "unknown",
-                              id: safe_platform_field(train_platform, :uuid),
+                              id: hash_to_uuid(safe_platform_field(train_platform, :uuid)),
                             },
 
                             runtime: Inspec::VERSION,
@@ -133,6 +133,38 @@ module Inspec
         platform.send(field)
       end
 
+      # Train's platform.uuid can return a raw 64-char hex identifier (e.g. a Docker
+      # container ID from /proc/self/mountinfo) instead of a proper UUID when running
+      # inside a container. Normalize such values into a valid RFC 4122 (v5-like) UUID
+      # so downstream consumers can always rely on the UUID format. Values that are
+      # already a valid UUID, or that don't match a 64-char hex string, are returned as-is.
+      def hash_to_uuid(value)
+        return nil if value.nil?
+
+        value_str = value.to_s
+        return nil if value_str.empty?
+
+        uuid_format = /^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}$/i
+        return value_str if value_str =~ uuid_format
+
+        hex_64_format = /^[0-9a-fA-F]{64}$/i
+        return value_str unless value_str =~ hex_64_format
+
+        bytes = value_str[0, 32].scan(/.{2}/).map { |byte| byte.to_i(16) }
+
+        time_low        = bytes[0..3].inject(0)   { |a, b| (a << 8) | b }
+        time_mid        = bytes[4..5].inject(0)   { |a, b| (a << 8) | b }
+        time_hi_version = bytes[6..7].inject(0)   { |a, b| (a << 8) | b }
+        clock_seq       = bytes[8..9].inject(0)   { |a, b| (a << 8) | b }
+        node            = bytes[10..15].inject(0) { |a, b| (a << 8) | b }
+
+        # Set version (5) and variant (RFC 4122) bits per the UUID spec
+        time_hi_version = (time_hi_version & 0x0FFF) | (5 << 12)
+        clock_seq       = (clock_seq & 0x3FFF) | 0x8000
+
+        format("%08x-%04x-%04x-%04x-%012x", time_low, time_mid, time_hi_version, clock_seq, node)
+      end
+
       def note_per_run_features(opts)
         note_all_invoked_features
         note_gem_dependency_usage(opts)

data/lib/inspec/version.rb

@@ -1,3 +1,3 @@
 module Inspec
-  VERSION = "7.2.1".freeze
+  VERSION = "7.3.4".freeze
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: inspec-core
 version: !ruby/object:Gem::Version
-  version: 7.2.1
+  version: 7.3.4
 platform: ruby
 authors:
 - Chef InSpec Team
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-08-07 00:00:00.000000000 Z
+date: 2026-10-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: chef-telemetry
@@ -937,7 +937,7 @@ homepage: https://github.com/inspec/inspec
 licenses:
 - LicenseRef-Chef-EULA
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -953,7 +953,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubygems_version: 3.3.27
-signing_key: 
+signing_key:
 specification_version: 4
 summary: Infrastructure and compliance testing. Core library.
 test_files: []
+...