text_extractor 0.1.7 → 0.1.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: e956aff10058c42b11f0cab0fd26f4218e6fcaa9
-  data.tar.gz: 0f947396657eb89749366e3c84ccb08899e07d19
+  metadata.gz: 33c60b21e8e025e62025f06af668dd4859cde35c
+  data.tar.gz: 3c3b0c21fc91326431f95d323daedc84f580ed53
 SHA512:
-  metadata.gz: e9b3e596e799952397dd8ae76c2deb52c84d356f468f1788f6c173cb9df4a77683ac279780900ed45dc49a729baba422288073ddac4c67349b93070bebd7140d
-  data.tar.gz: ed8e0f2321f97843748e7d980b2e15e97fe651c2a80409f8daaa049786a426c3321eb77791e3ecb641d570c0e9dc7ae9673a2ea25086e7b513323597e4e519c4
+  metadata.gz: b6e4d910641926e734e9c0cc22d486e15ac695c549ca1eb09c1e78621caa49e6e535533becb83dcd3f2ad93b81954cfc8efa74ade21f4b618a6f56ffe74ed978
+  data.tar.gz: 2ecc4bfd0e9e123e46254dbe264d5a102593a48a1668f9a0d41f7d66b22266925b58b21cf57c138121f4ece8559be3151ca33de6eb5905f965def79cbae096c6

data/lib/text_extractor.rb

@@ -65,34 +65,10 @@ class TextExtractor
     value(id, re) { |val| IPAddr.new(val) }
   end
 
-  def strip_record(regexp, strip: nil)
-    lines = regexp.source.split("\n")
-    prefix = lines.last
-
-    if prefix =~ /\A\s*\z/
-      lines.pop if lines.first =~ /\A\s*\z/
-      lines.shift
-      strip_record_by_line(lines, prefix, strip)
-    end
-
-    Regexp.new(lines.join("\n"), regexp.options)
-  end
-
-  def strip_record_by_line(lines, prefix, strip)
-    lines.map! { |s| s.gsub(prefix.to_s, '') }
-    case strip
-    when :left  then lines.map! { |s| "\[ \t\r\f]*#{s.lstrip}" }
-    when :right then lines.map! { |s| "#{s.rstrip}\[ \t\r\f]*" }
-    when :both  then lines.map! { |s| "\[ \t\r\f]*#{s.strip}\[ \t\r\f]*" }
-    end
-  end
-
   def record(klass = Record, **kwargs, &block)
     raise "#{self.class}.record requires a block" unless block
-    @current_record_values = []
-    regexp = strip_record(instance_exec(&block), strip: kwargs.delete(:strip))
-    kwargs[:values] = @current_record_values
-    @records << klass.new(regexp, **kwargs)
+    kwargs[:values] = @current_record_values = []
+    @records << klass.new(instance_exec(&block), **kwargs)
   end
 
   def filldown(**kwargs, &block)

data/lib/text_extractor/directives.rb

@@ -39,7 +39,7 @@ class TextExtractor
     private
 
     DIRECTIVE_MAP = {
-      ' '      => { class: Comment, arguments: ->(source) { [source[1..-1]] } },
+      ' '      => { class: Comment },
       'any'    => { class: Any },
       'begin'  => { class: Begin, arguments: :parsed },
       'end'    => { class: End },
@@ -95,9 +95,9 @@ class TextExtractor
     end
 
     def parse_one_directive(source)
-      md = source.match(/^[a-z_]+/)
+      md = source.match(/^[a-z_]+/) || source.match(/^ /)
+      raise "Unknown directive(s) in #{@state.current_line}" unless md
       word = md[0]
-      raise "Unknown directive(s) #{source}" unless md
       map = DIRECTIVE_MAP.fetch(word) { raise "Unknown directive #{word}" }
       args = parse_arguments(map[:arguments], md.post_match)
       map.fetch(:class).new(@state, *args)

data/lib/text_extractor/directives/classes.rb

@@ -73,7 +73,11 @@ class TextExtractor
     # skip to end of line
     class Rest < Directive
       def call
-        state.current = [state.current, '[^\\n]*']
+        state.current = if state.newline?
+                          [state.current.chomp, '[^\\n]*\n']
+                        else
+                          [state.current, '[^\\n]*']
+                        end
       end
     end
   end

data/lib/text_extractor/record.rb

@@ -4,8 +4,9 @@ class TextExtractor
   class Record
     attr_reader :regexp, :factory, :values
 
-    def initialize(regexp, factory: nil, values: [], fill: [], directives: true)
-      @regexp = expand_regexp(regexp, directives)
+    def initialize(regexp, factory: nil, values: [], fill: [], directives: true,
+                   strip: nil)
+      @regexp = build_regexp(regexp, directives, strip)
       @factory = factory
       @constructor = FactoryAnalyzer.new(factory).to_proc
       @values = values.map { |val| [val.id, val] }.to_h
@@ -25,6 +26,33 @@ class TextExtractor
       @constructor.call(extracted)
     end
 
+    def build_regexp(regexp, directives, strip)
+      stripped = strip_regexp(regexp, strip)
+      expanded = expand_regexp(stripped, directives)
+      ignore_regexp(expanded, strip)
+    end
+
+    def strip_regexp(regexp, strip)
+      lines = regexp.source.split("\n")
+      prefix = lines.last
+      if lines.first =~ /\A\s*\z/ && prefix =~ /\A\s*\z/
+        lines.shift
+        lines = lines.map { |s| s.gsub(prefix, '') }
+        lines = lines.map(&regexp_line_stripper(strip))
+      end
+      Regexp.new(lines.join("\n"), regexp.options)
+    end
+
+    def regexp_line_stripper(strip)
+      case strip
+      when :left  then ->(s) { s.lstrip }
+      when :right then ->(s) { s.rstrip }
+      when :both  then ->(s) { s.strip }
+      when nil, false then ->(s) { s }
+      else raise "Unknown strip option: #{strip}"
+      end
+    end
+
     def expand_regexp(regexp, directives)
       if directives
         TextExtractor.expand_directives(regexp)
@@ -33,6 +61,21 @@ class TextExtractor
       end
     end
 
+    def ignore_regexp(regexp, strip)
+      return regexp unless strip
+      lines = regexp.source.split("\n").map(&regexp_line_ignorer(strip))
+      Regexp.new(lines.join("\n"), regexp.options)
+    end
+
+    def regexp_line_ignorer(strip)
+      case strip
+      when :left  then ->(s) { "\[ \\t\\r\\f]*#{s}" }
+      when :right then ->(s) { "#{s}\[ \\t\\r\\f]*" }
+      when :both  then ->(s) { "\[ \\t\\r\\f]*#{s}\[ \\t\\r\\f]*" }
+      else raise "Unknown ignore whitespace option: #{strip}"
+      end
+    end
+
     def match(string, pos = 0)
       @regexp.match(string, pos)
     end

data/lib/text_extractor/version.rb

@@ -1,5 +1,5 @@
 class TextExtractor
   def self.version
-    '0.1.7'
+    '0.1.8'
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: text_extractor
 version: !ruby/object:Gem::Version
-  version: 0.1.7
+  version: 0.1.8
 platform: ruby
 authors:
 - Ben Miller
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-03-22 00:00:00.000000000 Z
+date: 2016-03-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler