[r2003-06-07 16:46:03 by cvscast] Benchmarks

Original author: cvscast
Date: 2003-06-07 16:46:22+00:00

benchmarks/Makefile

+split:
+	ocaml unix.cma bench.ml | tee split.results

benchmarks/bench.ml

+open Printf
+
+let gen = "split.ml"
+let xml = "split"
+let cduce = [ "split0.cd"; "split1.cd"; "split2.cd" ]
+let xslt = [ "split1.xsl"; "split2.xsl" ]
+let xduce = [ "split1.q"; "split2.q" ]
+
+let sp = sprintf
+let langs = 
+  [
+    "CDuce", cduce,
+    (fun script xml -> 
+       sp "../cduce --quiet %s --arg %s" script xml);
+
+    "XDuce", xduce,
+    (fun script xml -> 
+       sp "/home/frisch/xduce-0.4.0/xduce.opt %s %s" script xml);
+
+    "XSLT", xslt,
+    (fun script xml -> 
+       sp "xsltproc --noout %s %s" script xml);
+  ]
+  
+
+let pr = printf
+let rep = 1
+let time s =
+  let s = "/usr/bin/time --format=\"%e,%U\\n\" " ^ s ^ " 2>&1" in
+  flush stdout;
+  for i = 1 to rep do
+    let ic = Unix.open_process_in s in
+    let t = input_line ic in
+    ignore (Unix.close_process_in ic);
+    pr "%s " t;
+    flush stdout;
+  done;
+  pr "\n"
+
+let args = [ 10 ; 50; 100; (* 500; 1000; 5000 *) ]
+
+let run s = 
+  flush stdout; 
+  ignore (Sys.command s)
+
+let () =
+  List.iter
+    (fun s ->
+       let fn = sp "%s.%i.xml" xml s in
+       run (sp "ocaml %s %i > %s" gen s fn);
+       let ic = open_in fn in
+       let size = in_channel_length ic in
+       close_in ic;
+       pr "XML size = %i\n" size;
+       List.iter
+	 (fun (lang, scripts, cmd) ->
+	    List.iter
+	      (fun script ->
+		 pr "%6s[%20s] " lang script;
+		 time (cmd script fn)
+	      ) scripts
+	 )
+	 langs;
+       pr "====================================\n"
+    )
+    args

benchmarks/split.ml

+let random_string n =
+  let s = String.create n in
+  for i = 0 to n - 1 do
+    s.[i] <- Char.chr (65 + (Random.int 26))
+  done;
+  s;;
+    
+let nb = int_of_string Sys.argv.(1);;
+
+print_string "<?xml version=\"1.0\" encoding=\"UTF-8\"?><doc>";;
+
+let rec person p =
+  Printf.printf "<person gender=\"%s\"><name>%s</name><children>\n"
+    (if Random.int 2 = 0 then "M" else "F")
+    (random_string 20);
+  if p < 5 then  for i = 1 to Random.int 5 do person (p+1) done;
+  Printf.printf "</children></person>\n";;
+
+for i = 1 to nb do
+  person 0;
+done;;
+
+print_string "</doc>";;

benchmarks/split0.cd

+type Person = FPerson | MPerson
+type FPerson = <person gender = "F" >[ Name Children ]
+type MPerson = <person gender="M">[ Name Children ]
+type Children = <children>[Person*]
+type Name = <name>[ PCDATA ]
+
+type Man = <man name=String>[ Sons Daughters ]
+type Woman = <woman name=String>[ Sons Daughters ]
+type Sons = <sons>[ Man* ]
+type Daughters = <daughters>[ Woman* ]
+
+let split (MPerson -> Man ; FPerson -> Woman)
+ <person gender=g>[ <name>n <children>[(mc::MPerson | fc::FPerson)*] ] ->
+   let tag = match g with "F" -> `woman | "M" -> `man in
+   let s = map mc with x -> split x in
+   let d = map fc with x -> split x in       
+   <(tag) name=n>[ <sons>s  <daughters>d ] 
+
+match argv with
+  | [ f ] -> load_xml f
+  | _ -> raise "Invalid command line"

benchmarks/split1.cd

+type Person = FPerson | MPerson
+type FPerson = <person gender = "F" >[ Name Children ]
+type MPerson = <person gender="M">[ Name Children ]
+type Children = <children>[Person*]
+type Name = <name>[ PCDATA ]
+
+type Man = <man name=String>[ Sons Daughters ]
+type Woman = <woman name=String>[ Sons Daughters ]
+type Sons = <sons>[ Man* ]
+type Daughters = <daughters>[ Woman* ]
+
+let split (MPerson -> Man ; FPerson -> Woman)
+ <person gender=g>[ <name>n <children>[(mc::MPerson | fc::FPerson)*] ] ->
+   let tag = match g with "F" -> `woman | "M" -> `man in
+   let s = map mc with x -> split x in
+   let d = map fc with x -> split x in       
+   <(tag) name=n>[ <sons>s  <daughters>d ] 
+
+match argv with
+  | [ f ] -> 
+   (match load_xml f with
+    |  <doc>(l & [Person*]) -> <doc>(map l with x -> split x)
+    | _ -> raise "Invalid document") 
+  | _ -> raise "Invalid command line"

benchmarks/split1.q

+import "xml.q"
+
+type Person = FPerson | MPerson
+type FPerson = person[ @gender["F"], Name, Children ]
+type MPerson = person[ @gender["M"], Name, Children ]
+type Children = children[Person*]
+type Name = name[ String ]
+
+type Man = man[  @name[String], Sons,Daughters ]
+type Woman = woman[ @name[String], Sons,Daughters ]
+type Sons = sons[ Man* ]
+type Daughters = daughters[ Woman* ]
+
+fun split_children (val c as Person* ) : (Sons,Daughters) =
+ let val s = filter c {( val m as MPerson { split_m(m) } | FPerson { () })*} in
+ let val d = filter c {( MPerson { () } | val f as FPerson { split_f(f) })*} in
+ sons[s], daughters[d]
+
+fun split_m (val p as MPerson) : Man =
+ match p with
+  person[ @gender[String], name[val n], children[val c] ] ->
+   man[ @name[n], split_children(c) ] 
+
+fun split_f (val p as FPerson) : Woman =
+ match p with
+  person[ @gender[String], name[val n], children[val c] ] ->
+   woman[ @name[n], split_children(c) ]
+
+fun split_seq (val p as Person* ) : (Man|Woman)* =
+ filter p { 
+   ( val f as FPerson { split_f(f) } 
+   | val m as MPerson { split_m(m) })* 
+ }
+
+let val _ = 
+ match argv() with
+  val fn as String ->
+   ( match load_xml(fn) with
+       doc[val p as Person*] -> doc[ split_seq(p) ]
+     | Any -> raise("Invalid document") )
+ | Any -> raise("Invalid command line")

benchmarks/split1.xsl

+<?xml version="1.0" encoding="ISO-8859-1"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+<xsl:template match="person[@gender = 'M']">
+  <man name="{name/text()}">
+    <sons>
+      <xsl:apply-templates select="children/person[@gender='M']"/>
+    </sons>
+    <daughters>
+      <xsl:apply-templates select="children/person[@gender='F']"/>
+    </daughters>
+  </man>
+</xsl:template>
+
+<xsl:template match="person[@gender = 'F']">
+  <woman name="{name/text()}">
+    <sons>
+      <xsl:apply-templates select="children/person[@gender='M']"/>
+    </sons>
+    <daughters>
+      <xsl:apply-templates select="children/person[@gender='F']"/>
+    </daughters>
+  </woman>
+</xsl:template>
+
+</xsl:stylesheet>

benchmarks/split2.cd

+type Person = FPerson | MPerson
+type FPerson = <person gender = "F" >[ Name Children ]
+type MPerson = <person gender="M">[ Name Children ]
+type Children = <children>[Person*]
+type Name = <name>[ PCDATA ]
+
+type Man = <man name=String>[ Sons Daughters ]
+type Woman = <woman name=String>[ Sons Daughters ]
+type Sons = <sons>[ Man* ]
+type Daughters = <daughters>[ Woman* ]
+
+let split (MPerson -> Man ; FPerson -> Woman)
+ <_ gender=g>[ <_>n <_>[(mc::<_ gender="M">_ | fc::_)*] ] ->
+   let tag = match g with "F" -> `woman | "M" -> `man in
+   let s = map mc with x -> split x in
+   let d = map fc with x -> split x in       
+   <(tag) name=n>[ <sons>s  <daughters>d ] 
+
+match argv with
+  | [ f ] -> 
+   (match load_xml f with
+    |  <doc>(l & [Person*]) -> <doc>(map l with x -> split x)
+    | _ -> raise "Invalid document") 
+  | _ -> raise "Invalid command line"

benchmarks/split2.q

+import "xml.q"
+
+type Person = FPerson | MPerson
+type MPerson = person[ @gender["M"], Name, Children ]
+type FPerson = person[ @gender["F"], Name, Children ]
+type Children = children[Person*]
+type Name = name[ String ]
+
+type Man = man[  @name[String], Sons,Daughters ]
+type Woman = woman[ @name[String], Sons,Daughters ]
+type Sons = sons[ Man* ]
+type Daughters = daughters[ Woman* ]
+
+fun split_children (val c as Person* ) : (Sons,Daughters) =
+ let val s = filter c {( val m as ~[ @gender["M"], ~[Any]* ] { split_m(m) } | ~[Any] { () })*} in
+ let val d = filter c {( val f as ~[ @gender["F"], ~[Any]* ] { split_f(f) } | ~[Any] { () })*} in
+ sons[s], daughters[d]
+
+fun split_m (val p as MPerson) : Man =
+ match p with
+  ~[ @gender[String], ~[val n], ~[val c] ] ->
+   man[ @name[n], split_children(c) ] 
+
+fun split_f (val p as FPerson) : Woman =
+ match p with
+  ~[ @gender[String], ~[val n], ~[val c] ] ->
+   woman[ @name[n], split_children(c) ]
+
+fun split_seq (val p as Person* ) : (Man|Woman)* =
+ filter p { 
+   ( val f as ~[ @gender["F"], ~[Any], ~[Any] ] { split_f(f) } 
+   | val m as ~[ @gender["M"], ~[Any], ~[Any] ] { split_m(m) })* 
+ }
+
+let val _ = 
+ match argv() with
+  val fn as String ->
+   ( match load_xml(fn) with
+       doc[val p as Person*] -> doc[ split_seq(p) ]
+     | Any -> raise("Invalid document") )
+ | Any -> raise("Invalid command line")
+

benchmarks/split2.xsl

+<?xml version="1.0" encoding="ISO-8859-1"?>
+<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
+
+<xsl:template match="person">
+  <xsl:variable name="gender">
+    <xsl:choose>
+      <xsl:when test="@gender = 'M'">man</xsl:when>
+      <xsl:otherwise>women</xsl:otherwise>
+    </xsl:choose>
+  </xsl:variable>
+  <xsl:element name="{$gender}">
+    <xsl:attribute name="name">
+      <xsl:value-of select="name/text()"/>
+    </xsl:attribute>
+    <sons>
+      <xsl:apply-templates select="children/person[@gender='M']"/>
+    </sons>
+    <daughters>
+      <xsl:apply-templates select="children/person[@gender='F']"/>
+    </daughters>
+  </xsl:element>
+</xsl:template>
+
+</xsl:stylesheet>