mwalker-smail 0.0.3

data/lib/smail/header.rb

@@ -0,0 +1,95 @@
+class SMail
+  class Header #:nodoc: all
+    attr_reader :crlf, :head
+
+    def initialize(headers, smail)
+      @crlf = smail.crlf || "\r\n"
+
+      @order = Array.new
+      @head = Hash.new
+      headers.each do |header|
+        @order << header.first
+        if @head[header.first].nil?
+          @head[header.first] = Array.new
+        end
+        @head[header.first] << header.last
+      end
+
+      @header_names = @head.collect {|header,value| [header.downcase, header] }
+    end
+
+    def to_s #:nodoc:
+      header_pairs.collect {|pair| fold(pair.join(": "))}.join(@crlf) + @crlf
+    end
+
+    def header_names
+      @header_names.collect {|pair| pair.last }
+    end
+
+    def header_pairs
+      headers = []
+      seen = {}
+      seen.default = -1
+      @order.each do |header|
+        headers << [header, @head[header][seen[header] += 1]]
+      end
+        
+      headers
+    end
+
+    def header(field)
+      return unless names = @header_names.assoc(field.downcase)
+      headers(field).first
+    end
+
+    def headers(field)
+      return unless names = @header_names.assoc(field.downcase)
+      
+      @head[names.last]
+    end
+
+    def header_set(field, lines)
+      unless field =~ /^[\x21-\x39\x3b-\x7e]+$/
+        raise(ArgumentError, "Field name contains illegal characters")
+      end
+      unless field =~ /^[\w-]+$/
+        raise(ArgumentError, "Field name is not limited to hyphens and alphanumerics")
+      end
+
+      old_length = 0
+      if @header_names.assoc(field.downcase)
+        old_length = @head[@header_names.assoc(field.downcase).last].length
+      else
+        @header_names << [field.downcase, field]
+      #  @order << field # FIXME: add new fields onto the end
+      end
+
+      @head[@header_names.assoc(field.downcase).last] = lines
+      if lines.length < old_length
+        # Remove the last x instances of field from @order
+        headers_to_remove = old_length - lines.length
+        headers_seen = 0
+        @order.collect! {|field_name|
+          if field_name.downcase == field.downcase
+            headers_seen += 1
+            next if headers_seen > headers_to_remove
+          end
+          field_name
+        }.compact!
+      elsif lines.length > old_length
+        # Add x instances of field to the end of @order
+        @order = @order + ([field] * (lines.length - old_length))
+      end
+
+      headers(field)
+    end
+
+private
+
+    def fold(line)
+      # FIXME: this needs to actually fold the line
+      line.gsub( /^(.{1,78})(\s+|$)/ ){ $2.empty? ? $1 : "#{$1}#{@crlf}\t" }
+    end
+
+  end
+end

data/lib/smail/parser.rb

@@ -0,0 +1,55 @@
+class SMail
+
+private
+
+  # We are liberal in what we accept.
+  PATTERN_CRLF			= "\n\r|\r\n|\n|\r" #:nodoc:
+
+  RE_CRLF           = Regexp.new(PATTERN_CRLF,      Regexp::MULTILINE) #:nodoc:
+
+  def split_head_from_body(text)
+    # The body is a sequence of characters after the header separated by an empty line
+    if text =~ /(.*?(#{PATTERN_CRLF}))\2(.*)/m
+      return $1, $3 || '', $2
+    else  # The body is, of course, optional.
+      return text, "", "\r\n"
+    end
+  end
+
+  def read_header(head)
+    headers = headers_to_list(head)
+
+    SMail::Header.new(headers, self)
+  end
+
+  # Header fields are lines composed of a field name, followed by a colon (":"),
+  # followed by a field body, and terminated by CRLF.  A field name MUST be
+  # composed of printable US-ASCII characters (i.e., characters that have values
+  # between 33 and 126, inclusive), except colon.  A field body may be composed
+  # of any US-ASCII characters, except for CR and LF.
+
+  # However, a field body may contain CRLF when used in header "folding" and
+  # "unfolding" as described in section 2.2.3.
+
+  def headers_to_list(head)
+    headers = Array.new
+
+    head.split(RE_CRLF).each do |line|
+      if line.gsub!(/^\s+/, '') or line !~ /([^:]+):\s*(.*)/
+        # This is a continuation line, we fold it on to the end of the previous header
+        next if headers.empty? # Most likely an mbox From line, skip it
+
+        #headers.last.last << (headers.last.empty? ? line : " #{line}")
+        if headers.last.empty?
+          headers.last.last << line
+        else
+          headers.last.last << " #{line}"
+        end
+      else
+        headers << [$1, $2]
+      end
+    end
+
+    headers
+  end
+end

data/lib/smail/smail.rb

@@ -0,0 +1,61 @@
+# SMail
+#
+# A very simple library for parsing email messages.
+#
+# Based on Email::Simple from CPAN
+#
+# No decoding of any fields or the body is attempted, see SMail::MIME.
+
+class SMail
+  # The line ending found in this email.
+  attr_reader :crlf
+  # The body text of the email.
+  attr_accessor :body
+
+  def initialize(text = '')
+    head, @body, @crlf = split_head_from_body(text)
+
+    @head = read_header(head)
+  end
+
+  # Returns the first value for the named header
+  def header(field)
+    @head.header(field)
+  end
+
+  # Returns an array containing every value for the named header, for the first instance
+  # see header
+  def headers(field)
+    @head.headers(field)
+  end
+
+  # Sets the header to contain the given data, if there is more than one existing header
+  # the extra headers are removed.
+  def header_set(field, line)
+    headers_set(field, line).first
+  end
+
+  # Sets the headers to contain the given data, passing in multiple lines results in
+  # multiple headers and order is retained.
+  def headers_set(field, *lines)
+    @head.header_set(field, lines)
+  end
+
+  def header=(header) #:nodoc:
+    # FIXME: takes a string and splits it??
+  end
+
+  # Returns the list of header names currently in the message. The order is not significant.
+  def header_names
+    @head.header_names
+  end
+
+  # Returns a list of pairs describing the contents of the header.
+  def header_pairs
+    @head.header_pairs
+  end
+
+  def to_s #:nodoc:
+    @head.to_s + @crlf + @body
+  end
+end

data/lib/smail/version.rb

@@ -0,0 +1,9 @@
+class SMail
+  module VERSION #:nodoc:
+    MAJOR = 0
+    MINOR = 0
+    TINY  = 3
+
+    STRING = [MAJOR, MINOR, TINY].join('.')
+  end
+end

data/lib/smail.rb

@@ -0,0 +1,2 @@
+# $Id: email_controller.rb 856 2006-10-26 03:09:39Z pete $
+Dir[File.join(File.dirname(__FILE__), 'smail/**/*.rb')].sort.each { |lib| require lib }

metadata

@@ -0,0 +1,57 @@
+--- !ruby/object:Gem::Specification 
+name: mwalker-smail
+version: !ruby/object:Gem::Version 
+  version: 0.0.3
+platform: ruby
+authors: 
+- Matthew Walker
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2009-05-16 00:00:00 -07:00
+default_executable: 
+dependencies: []
+
+description: 
+email: matthew@walker.wattle.id.au
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- lib/smail.rb
+- lib/smail/header.rb
+- lib/smail/parser.rb
+- lib/smail/smail.rb
+- lib/smail/version.rb
+has_rdoc: false
+homepage: http://github.com/mwalker/smail
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.2.0
+signing_key: 
+specification_version: 2
+summary: A simple RFC2822 email parser
+test_files: []
+