borrow_direct 1.0.1 → 1.0.2

checksums.yaml

@@ -1,15 +1,15 @@
 ---
 !binary "U0hBMQ==":
   metadata.gz: !binary |-
-    ZWJhNjQyYWUzNTZkNGQzMDc5MmIwZmFmM2Q2MWY5ODhmODk3NTFmOA==
+    ZWQyOTk1Y2ViNGM5ZTUyYTljZWRjMTEwNTlkMzcxZGEzM2ViNDJkOA==
   data.tar.gz: !binary |-
-    ZmIzMmNjM2NlYzk3OWNlZDE3YjkwZjhmOWMxYjVkMTM2M2IyYWEyYg==
+    MGMwOWQ2ODJlODI5MTk3NDNhYWZlMDY1ZDZjZTAzYjg3YWJhNjg4Ng==
 SHA512:
   metadata.gz: !binary |-
-    NmRlYWRmZWM3YmViNGMzZjAzMDhmN2IyMWI0MTI1OTViNGI4MWUxYjVkYTk2
-    ODFjNDE3OGI5MTU0MDg5YTI0Yjc5MWQ2ZmZlYjM2NzRhZjdiZGNmMmI2MzY3
-    M2Q5M2IzYzBlZjg2MjgyZTg5MDJkNTQyY2I0ZDMxOGMyMzA4NWE=
+    ZWU4N2IzYjM5M2EwMGJmZTU1ZTkxNzRmMTBiZDBmOTYyYmVlMTk0MDIwMDJm
+    ODBiMTBiNzA1OWNlOGFmZjNhNmQzMzM4YzI5ZjQzYzEwZTc3NzU3NzNhNTE2
+    YTdhMTRmODcyZjZkMzZlNDhjNmJkNTVkMzYzNjgyNzc4MTVhOGU=
   data.tar.gz: !binary |-
-    NjcwMzJhZTkzYjNmMTQ2YzhiMTg3ZTk2ZjQ0NTljNjk2NDhlNDJjNDdmYWFm
-    Y2VjYjM1MDFlZDc5ODZkYTQwYjUzYmU1OGY3M2U5NjBmZjMxNDQyM2NmNjQ4
-    YzQzOTFhNjI1MjcwZjIxOWYyNWM3MTM3MGI1ODZhZjExOTMzNDY=
+    ODg1YjdjM2M2ZTkxMzIzMDljZDZkMGEyNzUzMzBmNjA3MmNkYWU5OWFmNDNl
+    YjRhYzYwZmFhNjQ1M2YzMzMxMWFlNjhlZDQ0YWYzMjIyYjFlMmM4OWZjYWE1
+    MTY1ODE0Y2YyYTM2OTlhZjhiMzQyMWVjMWZkYzI4ZjMwODAwZTE=

data/lib/borrow_direct/generate_query.rb

@@ -99,9 +99,19 @@ module BorrowDirect
       # before the first colon OR semi-colon
       title.sub!(/[\:\;](.*)$/, '')
 
-      # remove any remaining non-alphanumeric, excepting apostrophe, replacing
-      # with space. The punctuation doesn't help our queries. 
-      title.gsub!(/[^[:alnum:][:space:]\']/, ' ')
+
+      # We want to remove anything that isn't a letter, number, or apostrophe.
+      # Other punctuation and weird chars don't help our query.
+      # We want to do it in a way that's unicode-aware.
+      #
+      # This is harder than expected, as ruby regex unicode-aware character
+      # classes don't seem to handle combining diacritics well.
+      #
+      # This crazy way does it, replace anything that matches unicode
+      # space (may include more than just ascii ' ') or punct class, unless it's an apostrophe --
+      # and replaces them with plain ascii space. (apostrophes are allowed to make it through,
+      # for possessive use)
+      title.gsub!(/[[:space:][:punct:]&&[^\']]/, ' ')
 
       # compress any remaining whitespace
       title.strip!

data/lib/borrow_direct/version.rb

@@ -1,3 +1,3 @@
 module BorrowDirect
-  VERSION = "1.0.1"
+  VERSION = "1.0.2"
 end

data/test/generate_query_test.rb

@@ -130,6 +130,17 @@ describe "GenerateQuery" do
       query = assert_bd_query_url(url)
     end
 
+    it "handles combining diacritics" do
+      # Some of the code we started with had a problem with combining diacritics. 
+      a_acute_combined = [97, 204, 129].pack("c*").force_encoding("UTF-8")
+
+      orig_title = "Vel#{a_acute_combined}squez's stuff...."
+
+      normalized_title = @generator.normalized_title(orig_title)
+
+      assert_equal "vel#{a_acute_combined}squez's stuff", normalized_title
+    end
+
 
 
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: borrow_direct
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.0.2
 platform: ruby
 authors:
 - Jonathan Rochkind
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-03-17 00:00:00.000000000 Z
+date: 2015-03-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: httpclient