oxml 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: ef18bf4918b7cecec5e065a1438d1ae31189db1cb4ff1926a1dcaad578bf513c
+  data.tar.gz: 02b8fc7882d4bc244a2dd0c126fa1e37671e9036e6617326f158036157e1f935
+SHA512:
+  metadata.gz: 12dbb1470019dcf00b6cdb1b90a4e0507daf7c20b07e11b42d21ade3d69aac454917e4eef08e9a40f7771acb1cca72019f93f2e80344088c515461bb65feb237
+  data.tar.gz: 8ea918b36f0c1bd55d940740710812f1a458e614f9fdc1ed860ac96885a5ae1bf893a705563cc8024f94a562bf860ea4c9e6a6d6aa46f2947b46f68944989f65

data/lib/oxml/parser.rb

@@ -0,0 +1,83 @@
+# frozen_string_literal: true
+
+module OXML
+  class Parser
+    EMPTY_STR = ''
+    TRUE_STR = 'true'
+    FALSE_STR = 'false'
+    DATE_TIME = /^-?\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}(?:\.\d+)?(?:Z|[+-]\d{2}:?\d{2})?$/.freeze
+    DATE = /^-?\d{4}-\d{2}-\d{2}(?:Z|[+-]\d{2}:?\d{2})?$/.freeze
+    TIME = /^\d{2}:\d{2}:\d{2}(?:\.\d+)?(?:Z|[+-]\d{2}:?\d{2})?$/.freeze
+
+    def initialize(options = {})
+      @memo = {}
+      @arr = []
+      @map = {}
+      @name = nil
+      @strip_namespaces = options.fetch(:strip_namespaces, false)
+      @delete_namespace_attributes = options.fetch(:delete_namespace_attributes, false)
+      @advanced_typecasting = options.fetch(:advanced_typecasting, false)
+    end
+
+    def to_h
+      @memo.to_h
+    end
+
+    def attr(name, str)
+      return if @delete_namespace_attributes
+
+      return if name == :version
+      return if name == :encoding
+
+      start_element("@#{name}")
+      text(str)
+      end_element(name)
+    end
+
+    def start_element(name)
+      @arr.push(@memo)
+
+      if @strip_namespaces && name.start_with?('@')
+        @name = name.to_sym
+      elsif @strip_namespaces
+        @name = @map[name] ||= Utils.snakecase(name).split(':').last.to_sym
+      else
+        @name = @map[name] ||= Utils.snakecase(name).to_sym
+      end
+
+      @memo = {}
+      text(@memo)
+    end
+
+    def end_element(_name)
+      @memo = @arr.pop
+    end
+
+    def text(value)
+      if @arr.last[@name].is_a?(Array)
+        @arr.last[@name].pop unless value == @memo
+        @arr.last[@name] << cast(value)
+      elsif @arr.last[@name] && value == @memo
+        @arr.last[@name] = [@arr.last[@name], value]
+      else
+        @arr.last[@name] = cast(value)
+      end
+    end
+
+    private
+
+    def cast(value)
+      return value unless @advanced_typecasting
+
+      case value
+      when EMPTY_STR then nil
+      when TRUE_STR then true
+      when FALSE_STR then false
+      when DATE_TIME then DateTime.parse(value)
+      when DATE then Date.parse(value)
+      when TIME then Time.parse(value)
+      else value
+      end
+    end
+  end
+end

data/lib/oxml/utils.rb

@@ -0,0 +1,15 @@
+# frozen_string_literal: true
+
+module OXML
+  module Utils
+    module_function
+
+    def snakecase(input)
+      input
+        .to_s
+        .gsub(/([A-Z]+)([A-Z][a-z])/, '\1_\2')
+        .gsub(/([a-z\d])([A-Z])/, '\1_\2')
+        .downcase
+    end
+  end
+end

data/lib/oxml/version.rb

@@ -0,0 +1,5 @@
+# frozen_string_literal: true
+
+module OXML
+  VERSION = '0.1.0'
+end

data/lib/oxml.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+require 'ox'
+
+require_relative 'oxml/version'
+require_relative 'oxml/utils'
+require_relative 'oxml/parser'
+
+module OXML
+  module_function
+
+  def parse(xml, options = {})
+    handler = Parser.new(options)
+    Ox.sax_parse(handler, xml)
+    handler.to_h
+  end
+end

metadata

@@ -0,0 +1,74 @@
+--- !ruby/object:Gem::Specification
+name: oxml
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- xkwd
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2023-11-05 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: ox
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.14'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.14'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+description: Fast XML to Hash parser built on top of the Ox gem (a native C extension)
+email:
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/oxml.rb
+- lib/oxml/parser.rb
+- lib/oxml/utils.rb
+- lib/oxml/version.rb
+homepage: https://github.com/xkwd/oxml
+licenses:
+- MIT
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '2.7'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.4.21
+signing_key:
+specification_version: 4
+summary: Fast XML to Hash parser
+test_files: []