onix 0.7.6 → 0.7.7

data/CHANGELOG

@@ -1,3 +1,7 @@
+v0.7.7 (1st October 2009)
+- optimise sed usage in ONIX::Normaliser. *huge* speed improvement on
+  large files.
+
 v0.7.6 (21st September 2009)
 - provide access to the PackQuantity element
 

data/lib/onix.rb

@@ -16,7 +16,7 @@ module ONIX
   module Version #:nodoc:
     Major = 0
     Minor = 7
-    Tiny  = 6
+    Tiny  = 7
 
     String = [Major, Minor, Tiny].join('.')
   end

data/lib/onix/normaliser.rb

@@ -71,12 +71,14 @@ module ONIX
       @curfile = dest
 
       # remove entities
-      replace_named_entities(@curfile)
+      dest = next_tempfile
+      replace_named_entities(@curfile, dest)
+      @curfile = dest
 
       FileUtils.cp(@curfile, @newfile)
     end
 
-    private
+    #private
 
     # check the specified app is available on the system
     #
@@ -133,7 +135,7 @@ module ONIX
           FileUtils.cp(inpath, outpath)
         else
           FileUtils.cp(inpath, outpath)
-          `sed -i 's/<?xml.*?>/<?xml version=\"1.0\" encoding=\"UTF-8\"?>/' #{outpath}`
+          `sed -i 's/<?xml.*?>/<?xml version=\"1.0\" encoding=\"UTF-8\"?>/g' #{outpath}`
         end
       elsif src_enc
         `iconv --from-code=#{src_enc} --to-code=UTF-8 #{inpath} > #{outpath}`
@@ -154,11 +156,15 @@ module ONIX
     # replace all named entities in the specified file with
     # numeric entities.
     #
-    def replace_named_entities(path)
-      # TODO: this is horrible. 1500 sed calls.
-      entity_map.each do |named, numeric|
-        `sed -i 's/\\&#{named};/\\&#{numeric};/g' #{path}`
-      end
+    def replace_named_entities(src, dest)
+      inpath = File.expand_path(src)
+      outpath = File.expand_path(dest)
+
+      cmd = "sed " + entity_map.map do |named, numeric|
+        "-e 's/\\&#{named};/\\&#{numeric};/g'"
+      end.join(" ") + " #{inpath} > #{outpath}"
+      #raise cmd
+      `#{cmd}`
     end
 
     # return a named entity to numeric entity mapping, build by extracting

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: onix
 version: !ruby/object:Gem::Version 
-  version: 0.7.6
+  version: 0.7.7
 platform: ruby
 authors: 
 - James Healy
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-09-22 00:00:00 +10:00
+date: 2009-10-01 00:00:00 +10:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency