add a handy tool for benchmarking the parsing of individual pages

tools/page_bench

@@ -0,0 +1,21 @@
+#!/usr/bin/env ruby
+
+# text extraction is a handy benchmark of parsing/lexing performance, as 
+# the full content stream of each page is processed.
+#
+# run like so:
+#
+#     ruby -Ilib tools/page_bench foo.pdf
+
+require 'rubygems'
+require 'pdf/reader'
+
+reader = PDF::Reader.new(ARGV[0])
+
+require "benchmark"
+
+Benchmark.bm(1) do |x|
+  reader.pages.each do |page|
+    x.report(page.number) { page.text }
+  end
+end