mail-sanitizer 0.3.0 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 658fd411823dcad053eede5f1c678558c216e60a3719c6e027051594734031f4
-  data.tar.gz: ed9a14a70b8d9453f5c0ba23959d0c85926d369f1daf67e0db8025074bcfd164
+  metadata.gz: 68bc7ac18fa8f7c6ab281621af51b38ef15182c34c994dcc6eb40ad3b5e96395
+  data.tar.gz: e218cbd8dd5ba270b61195ed1f8295bc18468ad1659129070da9bcde8c0c2067
 SHA512:
-  metadata.gz: acfa6e0931da027d2c016735f650f809f1e90c89596c41cc03be9c1052d2b58818482e75450b4cd2f81a84453e6e0fd629bd33ef49576853aede72b699ff5f51
-  data.tar.gz: 6254de9db9f514f511fa9b2df7947a819409faea88a1ac2228a308cb8407479f4d5c4d438604a24e2554c7058b15abf2d75c7db6f84898c822a45e432d758f3a
+  metadata.gz: 7b436a57c57a46d4d745565fdfd382ee48b7d739653146426fc44f30ec76e92f2a9459706c801288b16142ce047f5c1cd73fe0970184247a45592f9321b9093a
+  data.tar.gz: f31f2e76788bafa8c4c518402331126757acd57e10ca469c03d1c4e6bd347d679a5674d0efda9028528a560daffac10870034290b979eda6600ef05fa835610c

data/.github/workflows/ci.yml

@@ -0,0 +1,24 @@
+name: CI
+
+on: [push, pull_request]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+
+    strategy:
+      fail-fast: false
+      matrix:
+        ruby: ['2.6', '2.7', '3.0', '3.1', '3.2']
+
+    name: ruby ${{ matrix.ruby }}
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby }}
+          bundler-cache: true
+      - name: Run test
+        run: |
+          bundle exec rspec

data/lib/mail/sanitizer/constant.rb

@@ -2,7 +2,7 @@ module Mail
   module Sanitizer
     class Constant
       QUOT_SYMBOL_PATTERN = /^>/
-      QUOT_PATTERN = /wrote:|-originalmessage-|-forwardedmessage-|forwardedby/
+      QUOT_PATTERN = /wrote:|-originalmessage-|-forwardedmessage-|forwardedby|beginforwardedmessage/
       QUOT_DATETIME_PATTERN = /^(on|at)/
       QUOT_KEYWORD_SET = [
         ['from:', 'sent:', 'to:', 'subject:'],

data/lib/mail/sanitizer/sanitizer.rb

@@ -7,6 +7,7 @@ module Mail
 
       def initialize(body)
         @src = body.dup
+        @src = @src.encode('utf-8', invalid: :replace, undef: :replace) unless @src.nil?
         @sanitized_body = nil
         @quot = nil
         @sign = nil
@@ -48,14 +49,18 @@ module Mail
         numrow = lines.size
 
         line_types = Array.new(numrow, :normal)
-        sign = false
+        sign_index = nil
         look_sign_symbol = false
         lines.each_with_index.reverse_each do |line, i|
           if line =~ Mail::Sanitizer::Constant::QUOT_SYMBOL_PATTERN
             line_types[i] = :quot
           elsif line =~ Mail::Sanitizer::Constant::SIGN_PATTERN
-            line_types[i] = :sign
-            sign = !sign
+            if sign_index.nil?
+              sign_index = i
+            else
+              line_types[i, sign_index - i + 1] = Array.new(sign_index - i + 1, :sign)
+              sign_index = nil
+            end
           elsif line =~ Mail::Sanitizer::Constant::SIGN_SYMBOL_PATTERN
             next if look_sign_symbol
             if line_types[i..-1].include?(:quot)
@@ -67,8 +72,6 @@ module Mail
             look_sign_symbol = true
           elsif line =~ /^[[:space:]]*$/
             line_types[i] = nil
-          elsif sign
-            line_types[i] = :sign
           end
         end
 
@@ -93,10 +96,9 @@ module Mail
             break if keywords.values.all?
           end
 
-          downcased_line = Mail::Sanitizer::String.downcase(line)
-          if keywords.values.all? || downcased_line =~ Mail::Sanitizer::Constant::QUOT_PATTERN ||
-             (downcased_line =~ Mail::Sanitizer::Constant::QUOT_DATETIME_PATTERN && Mail::Sanitizer::String.include_datetime?(line)) ||
-             (Mail::Sanitizer::String.include_datetime?(line) && Mail::Sanitizer::String.include_email_address?(line))
+          if keywords.values.all? ||
+             Mail::Sanitizer::String.quot_pattern?(line) ||
+             Mail::Sanitizer::String.include_datetime_and_email_address?(line)
             line_types[i, numrow - i] = Array.new(numrow - i, :quot) unless line_types[i] == :quot
             break
           end

data/lib/mail/sanitizer/string.rb

@@ -1,7 +1,7 @@
 module Mail
   module Sanitizer
     class String
-      ADDRESS_REGEXP = /\A([^@\s]+)@((?:[-a-z0-9]+\.)+[a-z]{2,})\z/i
+      ADDRESS_REGEXP = /([a-zA-Z0-9_!#$%&`'"*+\-{|}~^\/=?\.]+@[a-zA-Z0-9][a-zA-Z0-9\.\-]+)/
       SP    = "[[:space:]]"
       DIGIT = "[0-9０-９]"
       YEAR  = "(#{DIGIT}{4})#{SP}*年"
@@ -17,13 +17,24 @@ module Mail
 
       class << self
         def split_line(str)
-          str.split(/[\r\n]/)
+          str.split(/\r\n|\r|\n/)
         end
 
         def downcase(str)
           str.downcase.gsub(/[[:space:]]/, '')
         end
 
+        def quot_pattern?(str)
+          s = downcase(str)
+          (s =~ Mail::Sanitizer::Constant::QUOT_PATTERN) ||
+          (s =~ Mail::Sanitizer::Constant::QUOT_DATETIME_PATTERN && Mail::Sanitizer::String.include_datetime?(str))
+        end
+
+        def include_datetime_and_email_address?(str)
+          s = Mail::Sanitizer::String.remove_email_address(str)
+          Mail::Sanitizer::String.include_email_address?(str) && Mail::Sanitizer::String.include_datetime?(s)
+        end
+
         def include_datetime?(str)
           str = replace_jp_datetime(str)
           DateTime.parse(str)
@@ -33,13 +44,11 @@ module Mail
         end
 
         def include_email_address?(str)
-          texts = str.split(/#{SP}/)
-          texts.each do |text|
-            text.strip!
-            text.gsub!(/^\(|^<|[^[:alpha:]]*$/, '')
-            return true if ADDRESS_REGEXP === text
-          end
-          return false
+          ADDRESS_REGEXP.match?(str)
+        end
+
+        def remove_email_address(str)
+          str.gsub(ADDRESS_REGEXP, '')
         end
 
         def replace_jp_datetime(str)

data/lib/mail/sanitizer/version.rb

@@ -1,5 +1,5 @@
 module Mail
   module Sanitizer
-    VERSION = "0.3.0"
+    VERSION = "0.4.1"
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: mail-sanitizer
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.1
 platform: ruby
 authors:
 - MasatoMiyoshi
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2019-04-05 00:00:00.000000000 Z
+date: 2023-10-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -73,9 +73,9 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
+- ".github/workflows/ci.yml"
 - ".gitignore"
 - ".rspec"
-- ".travis.yml"
 - CODE_OF_CONDUCT.md
 - Gemfile
 - LICENSE.txt
@@ -108,7 +108,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.3
+rubygems_version: 3.3.3
 signing_key: 
 specification_version: 4
 summary: A simple sanitizer for mail bodies

data/.travis.yml

@@ -1,7 +0,0 @@
----
-sudo: false
-language: ruby
-cache: bundler
-rvm:
-  - 2.6.1
-before_install: gem install bundler -v 1.17.2