utf8-cleaner 0.2.2 → 0.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 483d7ad030797ac6e4f59b23989e85cb02f8c871
-  data.tar.gz: ed06229b5cec24770bbfaba9c4586528a0f519a3
+  metadata.gz: 3430532196f27d692b381b983f9379c09e64daeb
+  data.tar.gz: 0245109f3fa9f3c6c36beefacf5180f5799f1f46
 SHA512:
-  metadata.gz: 80b2ec52b93bf99814b479d1059db52f8a8c507ee8df0434c29cc644ba93a65100573c1e4a3609f97a0bd94170cdad23bee0926064dc2e52d9c2d429d88c2635
-  data.tar.gz: 019d366a8de11b5c55625be6e4c19f3260d35d62e059b30111acc1cfc0c8fc5cd0c9c173e8003308a5fc6fc409305a6e3aaaa23066a172215cddb7f268238f62
+  metadata.gz: b2e43c9c6b904c0ae23eb06f12a392c544adef9712ea017aa75e2d0b200680f5e7e154acd7d9d65c3859d2e252a39f6458f4dcd6fb42e4a5ab4e8b4d8337d89e
+  data.tar.gz: 440f660f1e62a99e0387e8991a63605776ce7ad8e0182404aebc3cf54ea44830aa0a7c78b75bdbc143b57458237e6e5c5c19ff2d31b2fc83e00d5de897bd8a6d

data/.travis.yml

@@ -1,3 +1,5 @@
+before_install:
+  - gem install bundler
 language: ruby
 rvm:
   - 1.9.3

data/README.md

@@ -43,3 +43,4 @@ Original middleware author: @phoet - https://gist.github.com/phoet/1336754
 * Ruby 1.9.3 compatibility: @pithyless - https://gist.github.com/pithyless/3639014
 * Code review and cleanup: @nextmat
 * POST body sanitization: @salrepe
+* Bug fixes: @cosine

data/lib/utf8-cleaner/middleware.rb

@@ -40,10 +40,19 @@ module UTF8Cleaner
 
     def sanitize_env_rack_input(env)
       case env['CONTENT_TYPE']
-      when 'application/x-www-form-urlencoded','application/json'
+      when 'application/x-www-form-urlencoded'
+        # This data gets the full cleaning treatment
         cleaned_value = cleaned_string(env['rack.input'].read)
         env['rack.input'] = StringIO.new(cleaned_value) if cleaned_value
         env['rack.input'].rewind
+      when 'application/json'
+        # This data only gets cleaning of invalid UTF-8 (e.g. from another charset)
+        # but we do not URI-decode it.
+        rack_input = env['rack.input'].read
+        if rack_input && !rack_input.ascii_only?
+          env['rack.input'] = StringIO.new(tidy_bytes(rack_input))
+        end
+        env['rack.input'].rewind
       when 'multipart/form-data'
         # Don't process the data since it may contain binary content
       else

data/lib/utf8-cleaner/version.rb

@@ -1,3 +1,3 @@
 module UTF8Cleaner
-  VERSION = "0.2.2"
+  VERSION = "0.2.3"
 end

data/spec/middleware_spec.rb

@@ -61,9 +61,18 @@ module UTF8Cleaner
         before do
           env['CONTENT_TYPE'] = 'application/json'
         end
-        it "removes removes invalid %-encoded UTF-8 sequences" do
+
+        it "tidies invalid UTF-8 sequences" do
+          env['rack.input'] = StringIO.new(%Q({"foo": "\xFFbar\xF8"}))
           env['rack.input'].rewind
-          expect(new_env['rack.input'].read).to eq('foo=bar')
+          expect(new_env['rack.input'].read).to eq(%Q({"foo": "\u00FFbar\u00F8"}))
+        end
+
+        it "does not attempt to URI-decode data" do
+          json = %Q({"foo": "%FF"})
+          env['rack.input'] = StringIO.new(json)
+          env['rack.input'].rewind
+          expect(new_env['rack.input'].read).to eq(json)
         end
       end
     end

data/spec/uri_string_spec.rb

@@ -1,3 +1,4 @@
+# encoding: UTF-8
 require 'spec_helper'
 
 module UTF8Cleaner

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: utf8-cleaner
 version: !ruby/object:Gem::Version
-  version: 0.2.2
+  version: 0.2.3
 platform: ruby
 authors:
 - Leon Miller-Out
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-03-21 00:00:00.000000000 Z
+date: 2016-04-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport