edifact_rails 1.1.1 → 1.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1343d4c915c6df4b716c31560cf58d5b350dc18ad074c12c3c562b13b0ce3717
-  data.tar.gz: 6067a62e4bdf24f2be92306a6aa4d4e21b9aa3b69ac6eff3eba8a8ae97229627
+  metadata.gz: 3c51652b41747f2b0c07ecd834c5c79b3bebbb1537464a450f37ee58ea68150e
+  data.tar.gz: 2669313b30c7565c60f4dc568c528127577770946f7d4a8ffee7a1fc87f4c07e
 SHA512:
-  metadata.gz: 8e1ce36c70da1e16fce1596fd43b345f5a284a77b82645b901d6eb88f25ae45c45d152aae760477526bd1c0230f59890f602c40fbded3cd3be61a80a8926e1c7
-  data.tar.gz: 2feffb89e9df203d37eb0f8aab6bcb673630dd637092a5fe406de9eff0d244d074fd964b1d10e7c68a75a48b9f75f58168240d42e7c2d900f469468be0a049d5
+  metadata.gz: ded47109a99423254023e4f7316e1bdc3d1dc6602e73225a900c815baa2b2c4d9d14fe16f131aa5e1239b6b49ecab6394d7051ec9fc66923c6ffb5eb0e674ba5
+  data.tar.gz: f3c9d7ae8f793651c62eff11b3356321f6bcc531b5ce61020f23d530a08b0112d46e338edc8ff1711bc514fb53875508b5792795c3cbea8056e5ee97a60e52f1

data/CHANGELOG.md

@@ -7,3 +7,17 @@
 ## 1.1.0 (27/04/2023)
 
 * Added support for TRADACOMS input
+
+## 1.1.1 (4/05/2023)
+
+* Fixed crash caused by running the gem in a production environment.
+
+## 1.2.0 (31/05/2024)
+
+* Added support for UNA segments. Special characters different from the defaults can now be used.
+* Added `#una_special_characters` method that returns just the special characters.
+
+## 1.2.1 (4/06/2024)
+
+* `#una_special_characters` method now also returns decimal notation character, default `.`.
+* `#una_special_characters` method can now take no arguments, and will return the default special characters if so.

data/README.md

@@ -4,7 +4,7 @@ This gem parses EDIFACT or TRADACOMS input, and converts it into a ruby array st
 
 It does not handle validation itself.
 
-This gem is heavily inspired by and attempts to output similar results as [edifact_parser](https://github.com/pvdvreede/edifact_parser), credits to [pvdvreede](https://github.com/pvdvreede)
+This gem is heavily inspired by and attempts to output similar results as [edifact_parser](https://github.com/pvdvreede/edifact_parser)
 
 ## Requirements
 
@@ -15,12 +15,12 @@ This gem has been tested on the following ruby versions:
 * 3.1.2
 * 3.2.2
 
-## Getting started 
+## Getting started
 
 In your `Gemfile`:
 
 ```ruby
-gem 'edifact_rails', '~> 1.0'
+gem 'edifact_rails', '~> 1.2'
 ```
 
 Otherwise:
@@ -47,6 +47,19 @@ ruby_array = EdifactRails.parse_file("your/file/path")
 ruby_array = EdifactRails.parse("LIN+1+1+0764569104:IB'QTY+1:25'")
 ```
 
+You can pull just the special characters from the UNA segment (or the defaults if no UNA segment is present):
+```ruby
+una_special_characters = EdifactRails.una_special_characters(your_string_input)
+# una_special_characters =>
+{
+  component_data_element_seperator: ":",
+  data_element_seperator: "+",
+  decimal_notation: ".",
+  escape_character: "?",
+  segment_seperator: "'"
+}
+```
+
 ## Output
 
 ### EDIFACT

data/lib/edifact_rails/parser.rb

@@ -2,48 +2,93 @@
 
 module EdifactRails
   class Parser
-    ESCAPE_CHARACTER = "?"
-    SEGMENT_SEPARATOR = "'"
-    DATA_ELEMENT_SEPARATOR = "+"
-    COMPONENT_DATA_ELEMENT_SEPARATOR = ":"
-
     def initialize
-      # Escape the special characters for use in regex later on
-      @escape_char_rx = Regexp.quote(EdifactRails::Parser::ESCAPE_CHARACTER)
-      @segment_separator_rx = Regexp.quote(EdifactRails::Parser::SEGMENT_SEPARATOR)
-      @data_element_separator_rx = Regexp.quote(EdifactRails::Parser::DATA_ELEMENT_SEPARATOR)
-      @component_data_element_separator_rx = Regexp.quote(EdifactRails::Parser::COMPONENT_DATA_ELEMENT_SEPARATOR)
+      # Set default separators
+      # They can be overridden by the UNA segment in #detect_special_characters
+      set_special_characters
     end
 
-    # Treat the input a little, split the input string into segments, parse them
+    # Treat the input, split the input string into segments, parse those segments
     def parse(string)
+      # Trim newlines and excess spaces around those newlines
+      string = string.gsub(/\s*\n\s*/, "")
+
+      # Check for UNA segment, update special characters if so
+      detect_special_characters(string)
+
+      # Does some funky regex maniulation to handle escaped special characters
       string = treat_input(string)
 
       # Split the input string into segments
-      segments = string.split(/(?<!#{@escape_char_rx})#{@segment_separator_rx}/)
+      segments = string.split(/(?<!#{Regexp.quote(@escape_character)})#{Regexp.quote(@segment_seperator)}/)
 
       # Detect if the input is a tradacoms file
       @is_tradacoms = segments.map { |s| s[3] }.uniq == ["="]
 
-      # Drop the UNA segment, if present
+      # Drop the UNA segment, if present (we have already dealt with it in #detect_special_characters)
       segments.reject! { |s| s[0..2] == "UNA" }
 
       # Parse the segments
       segments.map { |segment| parse_segment(segment) }
     end
 
+    # Given an input string, return the special characters as defined by the UNA segment
+    # If no UNA segment is present, returns the default special characters
+    def una_special_characters(string)
+      detect_special_characters(string)
+
+      {
+        component_data_element_seperator: @component_data_element_seperator,
+        data_element_seperator: @data_element_seperator,
+        decimal_notation: @decimal_notation,
+        escape_character: @escape_character,
+        segment_seperator: @segment_seperator
+      }
+    end
+
     private
 
-    def treat_input(string)
-      # Trim newlines and excess spaces around those newlines
-      string = string.gsub(/\s*\n\s*/, "")
+    def set_special_characters(
+      component_data_element_seperator =
+        EdifactRails::DEFAULT_SPECIAL_CHARACTERS[:component_data_element_seperator],
+      data_element_seperator = EdifactRails::DEFAULT_SPECIAL_CHARACTERS[:data_element_seperator],
+      decimal_notation = EdifactRails::DEFAULT_SPECIAL_CHARACTERS[:decimal_notation],
+      escape_character = EdifactRails::DEFAULT_SPECIAL_CHARACTERS[:escape_character],
+      segment_seperator = EdifactRails::DEFAULT_SPECIAL_CHARACTERS[:segment_seperator]
+    )
+      # Set the special characters
+      @component_data_element_seperator = component_data_element_seperator
+      @data_element_seperator = data_element_seperator
+      @decimal_notation = decimal_notation
+      @escape_character = escape_character
+      @segment_seperator = segment_seperator
+    end
+
+    def detect_special_characters(string)
+      # UNA tags must be at the start of the input otherwise they are ignored
+      return unless string[0..2] == "UNA"
 
+      # UNA segments look like this:
+      #
+      # UNA:+.? '
+      #
+      # UNA followed by 6 special characters which are, in order:
+      # 1. Component data element separator
+      # 2. Data element separator
+      # 3. Decimal notation (must be . or ,)
+      # 4. Release character (aka escape character)
+      # 5. Reserved for future use, so always a space for now
+      # 6. Segment terminator
+      set_special_characters(string[3], string[4], string[5], string[6], string[8])
+    end
+
+    def treat_input(string)
       # Prepare regex
       other_specials_rx = Regexp.quote(
         [
-          EdifactRails::Parser::SEGMENT_SEPARATOR,
-          EdifactRails::Parser::DATA_ELEMENT_SEPARATOR,
-          EdifactRails::Parser::COMPONENT_DATA_ELEMENT_SEPARATOR
+          @segment_seperator,
+          @data_element_seperator,
+          @component_data_element_seperator
         ].join
       )
 
@@ -64,7 +109,10 @@ module EdifactRails
       #
       # "LIN+even????+123" => '+' is not escaped, gsub'ed => "even???? +123" => parsed => ['LIN', ['even??'], [123]]
       # "LIN+odd???+123" => '+' is escaped, not gsub'ed => "odd???+123" => parsed => ['LIN', ['odd?+123']]
-      string.gsub(/(?<!#{@escape_char_rx})((#{@escape_char_rx}{2})+)([#{other_specials_rx}])/, '\1 \3')
+      string.gsub(
+        /(?<!#{Regexp.quote(@escape_character)})((#{Regexp.quote(@escape_character)}{2})+)([#{other_specials_rx}])/,
+        '\1 \3'
+      )
     end
 
     # Split the segment into data elements, take the first as the tag, then parse the rest
@@ -72,10 +120,10 @@ module EdifactRails
       # If the input is a tradacoms file, the segment tag will be proceeded by '=' instead of '+'
       # 'QTY=1+A:B' instead of 'QTY+1+A:B'
       # Fortunately, this is easily handled by simply changing these "="s into "+"s before the split
-      segment[3] = EdifactRails::Parser::DATA_ELEMENT_SEPARATOR if @is_tradacoms && segment.length >= 4
+      segment[3] = @data_element_seperator if @is_tradacoms && segment.length >= 4
 
       # Segments are made up of data elements
-      data_elements = segment.split(/(?<!#{@escape_char_rx})#{@data_element_separator_rx}/)
+      data_elements = segment.split(/(?<!#{Regexp.quote(@escape_character)})#{Regexp.quote(@data_element_seperator)}/)
 
       # The first element is the tag, pop it off
       parsed_segment = []
@@ -88,7 +136,8 @@ module EdifactRails
     # Split the data elements into component data elements, and treat them
     def parse_data_element(element)
       # Split data element into components
-      components = element.split(/(?<!#{@escape_char_rx})#{@component_data_element_separator_rx}/)
+      components =
+        element.split(/(?<!#{Regexp.quote(@escape_character)})#{Regexp.quote(@component_data_element_seperator)}/)
 
       components.map { |component| treat_component(component) }
     end
@@ -99,16 +148,16 @@ module EdifactRails
       component.strip!
 
       # Prepare regex
-      all_special_characters = [
-        EdifactRails::Parser::SEGMENT_SEPARATOR,
-        EdifactRails::Parser::DATA_ELEMENT_SEPARATOR,
-        EdifactRails::Parser::COMPONENT_DATA_ELEMENT_SEPARATOR,
-        EdifactRails::Parser::ESCAPE_CHARACTER
+      all_special_characters_string = [
+        @segment_seperator,
+        @data_element_seperator,
+        @component_data_element_seperator,
+        @escape_character
       ].join
 
       # If the component has escaped characters in it, remove the escape character and return the character as is
       # "?+" -> "+", "??" -> "?"
-      component.gsub!(/#{@escape_char_rx}([#{Regexp.quote(all_special_characters)}])/, '\1')
+      component.gsub!(/#{Regexp.quote(@escape_character)}([#{Regexp.quote(all_special_characters_string)}])/, '\1')
 
       # Convert empty strings to nils
       component = nil if component.empty?

data/lib/edifact_rails/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module EdifactRails
-  VERSION = "1.1.1"
+  VERSION = "1.2.1"
 end

data/lib/edifact_rails.rb

@@ -3,12 +3,25 @@
 require "edifact_rails/parser"
 
 module EdifactRails
+  DEFAULT_SPECIAL_CHARACTERS = {
+    component_data_element_seperator: ":",
+    data_element_seperator: "+",
+    decimal_notation: ".",
+    escape_character: "?",
+    segment_seperator: "'"
+  }.freeze
+
   def self.parse(string)
     parser = EdifactRails::Parser.new
-    parser.parse string
+    parser.parse(string)
   end
 
   def self.parse_file(file_path)
     parse(File.read(file_path).split("\n").join)
   end
+
+  def self.una_special_characters(string = '')
+    parser = EdifactRails::Parser.new
+    parser.una_special_characters(string)
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: edifact_rails
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.2.1
 platform: ruby
 authors:
 - David Blackwood
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-05-04 00:00:00.000000000 Z
+date: 2024-06-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: byebug