Revamped the profiling setup.

This removes the need for dozens of standalone gnuplot scripts, adds extra
profiling data and makes the actual profiling easier.

profile/gnuplot/memory_time.gpi

@@ -0,0 +1,21 @@
+# The `sample_file` variable should contain the path to the variable to plot.
+
+set title sample_file
+
+set xlabel "Time"
+set ylabel "Memory (MB)"
+
+set xdata time
+set timefmt "%Y-%m-%dT%H:%M:%S"
+
+set yrange [0:*]
+
+set term qt persist size 900,600
+set grid
+
+set style line 1 lc rgb "#0060ad" lt 1 lw 2 pt 7 ps 1
+
+plot sample_file \
+    using 1:($2 / 1024 ** 2) \
+    with linespoints ls 1 \
+    title "Memory"

profile/lexer/big_xml.rb

@@ -1,8 +1,7 @@
 require_relative '../profile_helper'
 
-xml   = read_big_xml
+xml = read_big_xml
-lexer = Oga::XML::Lexer.new(xml)
 
-profile_memory('lexer/big_xml')
+profile_memory('lexer/big_xml') do
-
+  Oga::XML::Lexer.new(xml).advance { }
-lexer.advance { |tok| }
+end

profile/parser/big_xml.rb

@@ -1,8 +1,7 @@
 require_relative '../profile_helper'
 
-xml    = read_big_xml
+xml = read_big_xml
-parser = Oga::XML::Parser.new(xml)
 
-profile_memory('parser/big_xml')
+profile_memory('parser/big_xml') do
-
+  Oga::XML::Parser.new(xml).parse
-parser.parse
+end

profile/plot.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+sample_file="$1"
+plot_script="$2"
+
+if [[ ! -f "$sample_file" ]]
+then
+    echo "The sample file ${sample_file} does not exist"
+    exit 1
+fi
+
+if [[ ! -f "$plot_script" ]]
+then
+    echo "The gnuplot script ${plot_script} does not exist"
+    exit 1
+fi
+
+gnuplot -e "sample_file='${sample_file}'" "${plot_script}"

profile/plot/lexer/big_xml.gnuplot

@@ -1,20 +0,0 @@
-#!/usr/bin/env gnuplot
-
-set title "Lexing 10MB of XML"
-
-set xlabel "Sample"
-set ylabel "Memory (MB)"
-
-set yrange [0:*]
-
-set term qt persist size 900,600
-set grid
-
-set style line 1 lc rgb "#0060ad" lt 1 lw 2 pt 7 ps 1
-
-plot "profile/samples/lexer/big_xml.txt" \
-    using 0:($1 / 1024 ** 2) \
-    with linespoints ls 1 \
-    title "Memory"
-
-# vim: set ft=gnuplot:

profile/plot/parser/big_xml.gnuplot

@@ -1,20 +0,0 @@
-#!/usr/bin/env gnuplot
-
-set title "DOM parsing 10MB of XML"
-
-set xlabel "Sample"
-set ylabel "Memory (MB)"
-
-set yrange [0:*]
-
-set term qt persist size 900,600
-set grid
-
-set style line 1 lc rgb "#0060ad" lt 1 lw 2 pt 7 ps 1
-
-plot "profile/samples/parser/big_xml.txt" \
-    using 0:($1 / 1024 ** 2) \
-    with linespoints ls 1 \
-    title "Memory"
-
-# vim: set ft=gnuplot:

profile/plot/pull_parser/big_xml.gnuplot

@@ -1,20 +0,0 @@
-#!/usr/bin/env gnuplot
-
-set title "Pull parsing 10MB of XML"
-
-set xlabel "Sample"
-set ylabel "Memory (MB)"
-
-set yrange [0:*]
-
-set term qt persist size 900,600
-set grid
-
-set style line 1 lc rgb "#0060ad" lt 1 lw 2 pt 7 ps 1
-
-plot "profile/samples/pull_parser/big_xml.txt" \
-    using 0:($1 / 1024 ** 2) \
-    with linespoints ls 1 \
-    title "Memory"
-
-# vim: set ft=gnuplot:

profile/profile_helper.rb

@@ -1,5 +1,10 @@
+require 'timeout'
+require 'time'
+
 require_relative '../lib/oga'
 
+Thread.abort_on_exception = true
+
 ##
 # Returns memory usage in bytes. This relies on the /proc filesystem, it won't
 # work without it.
@@ -20,22 +25,52 @@ def read_big_xml
 end
 
 ##
-# Writes memory samples to a file until the thread is killed.
+# Runs the specified block for at least N seconds while profiling memory usage
+# at semi random intervals.
 #
 # @param [String] name The name of the samples file.
-# @param [Fixnum] interval The sample interval. The default is 200 ms.
+# @param [String] duration The amount of seconds to run.
-# @return [Thread]
 #
-def profile_memory(name, interval = 0.2)
+def profile_memory(name, duration = 30)
-  return Thread.new do
+  monitor = true
+  threads = []
+
+  threads << Thread.new do
+    puts 'Starting sampler...'
+
     path        = File.expand_path("../samples/#{name}.txt", __FILE__)
     handle      = File.open(path, 'w')
     handle.sync = true
 
-    loop do
+    while monitor
-      handle.write("#{memory_usage}\n")
+      usage    = memory_usage
+      usage_mb = (usage / 1024 / 1024).round(2)
+      time     = Time.now.strftime('%Y-%m-%dT%H:%M:%S')
 
-      sleep(interval)
+      handle.write("#{time} #{usage}\n")
+
+      puts "#{time}: #{usage_mb} MB"
+
+      sleep(rand)
+    end
+
+    puts 'Stopping sampler...'
+
+    handle.close
+  end
+
+  threads << Thread.new do
+    start = Time.now
+
+    begin
+      Timeout.timeout(duration) { loop { yield} }
+    rescue Timeout::Error
+      diff    = Time.now - start
+      monitor = false
+
+      puts "Finished running after #{diff.round(3)} seconds"
     end
   end
+
+  threads.each(&:join)
 end

profile/pull_parser/big_xml.rb

@@ -1,8 +1,8 @@
 require_relative '../profile_helper'
 
-xml    = read_big_xml
+xml = read_big_xml
-parser = Oga::XML::PullParser.new(xml)
 
-profile_memory('pull_parser/big_xml')
+profile_memory('pull_parser/big_xml') do
+  Oga::XML::PullParser.new(xml).parse { }
+end
 
-parser.parse { |node| }