Updated scripts

Author: fabienrenaud

output/toCsv.py

@@ -0,0 +1,73 @@
+#!/usr/bin/env python3
+
+import csv
+import os
+import re
+
+
+def parseFileInfo(f):
+    words = f.replace('.txt', '').split('-')
+    return {
+        'datatype': words[0],
+        'command': words[1],
+        'number': words[2],
+        'size': words[3]
+    }
+
+
+def extractScores(f):
+    result = {}
+    with open(f) as fh:
+        lines = fh.readlines()
+    summary = False
+    for l in lines:
+        if l.startswith("# Run complete. Total time:"):
+            summary = True
+        elif summary and not l.startswith('Benchmark'):
+            values = re.split('\s+', l)
+            if (len(values) > 3):
+                name = values[0].split('.')[-1]
+                score = values[3]
+                result[name] = score
+    return result
+
+
+cwd = os.getcwd()
+sheets = {}
+for f in os.listdir(cwd):
+    if os.path.isfile(f) and f.endswith('.txt'):
+        fileInfo = parseFileInfo(f)
+        scores = extractScores(f)
+
+        sheetName = "{}-{}-{}".format(fileInfo['datatype'],
+                                      fileInfo['command'], fileInfo['number'])
+        if sheetName not in sheets:
+            sheets[sheetName] = {}
+
+        size = fileInfo['size']
+        for k, v in scores.items():
+            if k not in sheets[sheetName]:
+                sheets[sheetName][k] = {}
+            sheets[sheetName][k][size] = v
+
+
+for filename, v in sheets.items():
+    fieldnames = ['Test']
+    rows = []
+    for libname, scores in v.items():
+        row = {}
+        row['Test'] = libname
+        for size, score in scores.items():
+            fieldname = "{}kb (x{})".format(size, size)
+            if fieldname not in fieldnames:
+                fieldnames.append(fieldname)
+            row[fieldname] = float(score) * int(size)
+        rows.append(row)
+
+    if not os.path.exists('csv/'):
+        os.mkdir('csv')
+    csvfilename = "csv/{}.csv".format(filename)
+    with open(csvfilename, 'w', newline='') as csvfile:
+        writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
+        writer.writeheader()
+        writer.writerows(rows)

output/toCsv.sh

@@ -1,35 +1,43 @@
-#!/bin/bash
-set -e
+#!/usr/bin/env sh -e
+
+datatypes=$(ls *.txt | cut -d- -f1 | sort | uniq)
 
 echo $(date +%Y-%m-%d)
 echo
 
-for f in $(ls *.txt); do
-  fnoxt=${f/.txt/}
-  name=$(echo $fnoxt | cut -d- -f1)
-  n=$(echo $fnoxt | cut -d- -f2)
-  s=$(echo $fnoxt | cut -d- -f3)
 
-  title=""
-  case $name in
-    "deser") title="Deserialization" ;;
-    "ser") title="Serialization" ;;
-    *) echo "Unknow name: '$name"; exit 1 ;;
-  esac
+for datatype in $datatypes; do
 
-  title="$title - $s KB"
- 
-  echo "# JMH: $title"
-  echo 
-  tail -n30 $f | grep -A30 "Benchmark" | sed -E -e 's/± +/±/' | \
-    awk 'BEGIN { OFS=" | " } { print "",$1,$2,$3,$4,$5,$6,""; if ($1 == "Benchmark") { print "|-----------|------|-----|-------|-------|-------|"; } }' | \
-    sed -E \
-    -e 's/^.*databind\.Deserialization\.([a-zA-Z_]+)/| \1\/databind/g' \
-    -e 's/^.*stream\.Deserialization\.([a-zA-Z_]+)/| \1\/stream/g' \
-    -e 's/^.*databind\.Serialization\.([a-zA-Z_]+)/| \1\/databind/g' \
-    -e 's/^.*stream\.Serialization\.([a-zA-Z_]+)/| \1\/stream/g' \
-    -e 's/\| ± \|/±/g' \
-    -e 's/^ //g'
+  echo "# *$datatype* data type"
   echo
-done
 
+  for f in $(ls $datatype-*.txt); do
+    fnoxt=${f/.txt/}
+    cmd=$(echo $fnoxt | cut -d- -f2)
+    n=$(echo $fnoxt | cut -d- -f3)
+    s=$(echo $fnoxt | cut -d- -f4)
+  
+    title=""
+    case $cmd in
+      "deser") title="Deserialization" ;;
+      "ser") title="Serialization" ;;
+      *) echo "Unknow name: '$name"; exit 1 ;;
+    esac
+  
+    title="$title - $s KB"
+   
+    echo "## JMH: $title"
+    echo 
+    tail -n30 $f | grep -A30 "Benchmark" | sed -E -e 's/± +/±/' | \
+      awk 'BEGIN { OFS=" | " } { print "",$1,$2,$3,$4,$5,$6,""; if ($1 == "Benchmark") { print "|-----------|------|-----|-------|-------|-------|"; } }' | \
+      sed -E \
+      -e 's/^.*databind\.Deserialization\.([a-zA-Z_]+)/| \1\/databind/g' \
+      -e 's/^.*stream\.Deserialization\.([a-zA-Z_]+)/| \1\/stream/g' \
+      -e 's/^.*databind\.Serialization\.([a-zA-Z_]+)/| \1\/databind/g' \
+      -e 's/^.*stream\.Serialization\.([a-zA-Z_]+)/| \1\/stream/g' \
+      -e 's/\| ± \|/±/g' \
+      -e 's/^ //g'
+    echo
+  done
+
+done

output/toMd.sh

@@ -15,7 +15,7 @@ for datatype in ${DATA_TYPES}; do
     for n in ${NUMBERS}; do
         for size in ${SIZES}; do
             for command in ${COMMANDS}; do
-                outputFile="output/${command}-${n}-${size}-${datatype}.txt"
+                outputFile="output/${datatype}-${command}-${n}-${size}.txt"
                 echo ./run ${command} --number ${n} --size ${size} --datatype ${datatype}
                 ./run ${command} \
                     --number ${n} \