text_extractor 0.0.2

data/lib/text_extractor/extraction.rb

@@ -0,0 +1,36 @@
+class TextExtractor
+  # represents a single execution of a TextExtractor
+  class Extraction
+    attr_reader :input, :extractor, :re, :pos, :matches, :values
+
+    def initialize(input, extractor)
+      @input = input
+      @extractor = extractor
+      @pos = 0
+      @matches = []
+      @last_match = nil
+    end
+
+    def extraction_matches
+      @fill = {}
+      matches.flat_map do |match|
+        extraction_match(match)
+      end
+    end
+
+    def extraction_match(match)
+      extractor.find_record_for(match).extraction(match, @fill)
+    end
+
+    def scan
+      re = extractor.to_re
+      loop do
+        match = input.match(re, pos)
+        break unless match
+        @pos = match.end(0)
+        @matches << match
+      end
+      self
+    end
+  end # class Extraction
+end # class TextExtractor

data/lib/text_extractor/filldown.rb

@@ -0,0 +1,10 @@
+require_relative "record"
+
+class TextExtractor
+  class Filldown < Record
+    def extraction(match, fill)
+      fill.merge!(extract_values(match))
+      []
+    end
+  end # class Filldown < Record
+end # class TextExtractor

data/lib/text_extractor/record.rb

@@ -0,0 +1,40 @@
+class TextExtractor
+  class Record
+    attr_reader :regexp, :factory, :values
+
+    def initialize(regexp, factory: nil, values: [], fill: [])
+      @regexp = regexp
+      @factory = factory
+      @values = values.map { |val| [val.id, val] }.to_h
+      @default_values = values.map { |val| [val.id, nil] }.to_h
+      @fill = Array(fill)
+    end
+
+    def extraction(match, fill)
+      extracted = {}.merge!(@default_values)
+                  .merge!(extract_fills fill)
+                  .merge!(extract_values match)
+      factory ? factory.new(*extracted.values) : extracted
+    end
+
+    def match(string, pos = 0)
+      @regexp.match(string, pos)
+    end
+
+    def source
+      @regexp.source
+    end
+
+    def options
+      @regexp.options
+    end
+
+    def extract_fills(fill)
+      @fill.zip(fill.values_at(*@fill)).to_h
+    end
+
+    def extract_values(match)
+      values.keys.map { |id| [id, values[id].convert(match[id])] }.to_h
+    end
+  end # class Record
+end # class TextExtractor

data/lib/text_extractor/value.rb

@@ -0,0 +1,15 @@
+class TextExtractor
+  class Value
+    attr_reader :id, :re
+
+    def initialize(id, re, &block)
+      @id = id
+      @re = re
+      @block = block if block_given?
+    end
+
+    def convert(value)
+      @block ? @block.call(*value) : value
+    end
+  end
+end

data/lib/text_extractor/version.rb

@@ -0,0 +1,5 @@
+class TextExtractor
+  def self.version
+    "0.0.2"
+  end
+end

data/lib/text_extractor.rb

@@ -0,0 +1,105 @@
+require_relative "text_extractor/extraction"
+require_relative "text_extractor/filldown"
+require_relative "text_extractor/record"
+require_relative "text_extractor/value"
+
+# represents an extractor definition
+class TextExtractor
+  attr_reader :records, :values
+
+  def initialize(&block)
+    fail "#{self.class}.new requires a block" unless block
+    @values = {}
+    @fill = {}
+    @values = {}
+    @records = []
+    @filldowns = []
+    @current_record_values = []
+    instance_exec(&block)
+  end
+
+  module Patterns
+    INTEGER = /\d+/
+    FLOAT = /\d+\.?|\d*\.\d+/
+    RATIONAL = %r(\d+/\d+)
+    IPV4 = /[0-9.]{7,15}/
+    IPV6 = /[:a-fA-F0-9\.]{2,45}/
+    IPADDR = Regexp.union(IPV4, IPV6)
+    IPV4_NET = /#{IPV4}\/\d{1,2}/
+    IPV6_NET = /#{IPV6}\/\d{1,3}/
+    IPNETADDR = Regexp.union(IPV4_NET, IPV6_NET)
+    TRUE = /y|yes|t|true|on/i
+    FALSE = /n|no|f|false|off/i
+    BOOLEAN = Regexp.union(TRUE, FALSE)
+  end
+
+  def value(id, re, &block)
+    val = @values[id] = Value.new(id, re, &block)
+    define_singleton_method(id) do
+      @current_record_values << val
+      "(?<#{id}>#{re.source})"
+    end
+  end
+
+  def boolean(id, re = Patterns::BOOLEAN)
+    value(id, re) { |val| !val.match(Patterns::FALSE) }
+  end
+
+  def integer(id, re = Patterns::INTEGER)
+    value(id, re) { |val| Integer(val) }
+  end
+
+  def float(id, re = Patterns::FLOAT)
+    value(id, re) { |val| Float(val) }
+  end
+
+  def rational(id, re = Patterns::RATIONAL)
+    value(id, re) { |val| Rational(val) }
+  end
+
+  def ipaddr(id, re = Patterns::IPADDR)
+    value(id, re) { |val| IPAddr.new(val) }
+  end
+
+  def ipnetaddr(id, re = Patterns::IPNETADDR)
+    value(id, re) { |val| IPAddr.new(val) }
+  end
+
+  def strip_record(regexp)
+    lines = regexp.source.lines
+    prefix = lines.last
+    lines.map! { |s| s.gsub("#{prefix}", "") } if prefix =~ /\A\s*\z/
+    Regexp.new(lines.join.strip, regexp.options)
+  end
+
+  def record(klass = Record, **kwargs, &block)
+    fail "#{self.class}.record requires a block" unless block
+    @current_record_values = []
+    regexp = strip_record(instance_exec(&block))
+    kwargs[:values] = @current_record_values
+    @records << klass.new(regexp, **kwargs)
+  end
+
+  def filldown(**kwargs, &block)
+    fail "#{self.class}.filldown requires a block" unless block
+    record(Filldown, **kwargs, &block)
+  end
+
+  def find_record_for(match)
+    records[records.length.times.find_index { |i| match["__#{i}"] }]
+  end
+
+  def scan(input)
+    Extraction.new(input, self).scan.extraction_matches
+  end
+
+  def regexps
+    @records.map.with_index do |record, i|
+      Regexp.new("(?<__#{i}>#{record.source})", record.options)
+    end
+  end
+
+  def to_re
+    Regexp.union(*regexps)
+  end
+end # class TextExtractor

metadata

@@ -0,0 +1,51 @@
+--- !ruby/object:Gem::Specification
+name: text_extractor
+version: !ruby/object:Gem::Version
+  version: 0.0.2
+  prerelease: 
+platform: ruby
+authors:
+- Ben Miller
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2015-10-20 00:00:00.000000000 Z
+dependencies: []
+description: 
+email: bjmllr@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/text_extractor.rb
+- lib/text_extractor/version.rb
+- lib/text_extractor/extraction.rb
+- lib/text_extractor/value.rb
+- lib/text_extractor/record.rb
+- lib/text_extractor/filldown.rb
+homepage: https://github.com/bjmllr/text_extractor
+licenses:
+- MIT
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: 2.0.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.23
+signing_key: 
+specification_version: 3
+summary: Easily extract data from text
+test_files: []