Home | History | Annotate | Download | only in regression 
      1 /*
      2  * Copyright (C) 2011 Google Inc.
      3  *
      4  * Licensed under the Apache License, Version 2.0 (the "License");
      5  * you may not use this file except in compliance with the License.
      6  * You may obtain a copy of the License at
      7  *
      8  * http://www.apache.org/licenses/LICENSE-2.0
      9  *
     10  * Unless required by applicable law or agreed to in writing, software
     11  * distributed under the License is distributed on an "AS IS" BASIS,
     12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
     13  * See the License for the specific language governing permissions and
     14  * limitations under the License.
     15  */
     16 
     17 package benchmarks.regression;
     18 
     19 import com.google.caliper.Param;
     20 import java.util.Locale;
     21 
     22 public class StringCaseMappingBenchmark {
     23     enum Inputs {
     24         EMPTY(""),
     25 
     26         // TODO: include hairy inputs like turkish and greek.
     27         // TODO: locale makes a difference too.
     28 
     29         LOWER2(lower(2)),
     30         UPPER2(upper(2)),
     31         MIXED2(mixed(2)),
     32 
     33         LOWER8(lower(8)),
     34         UPPER8(upper(8)),
     35         MIXED8(mixed(8)),
     36 
     37         LOWER32(lower(32)),
     38         UPPER32(upper(32)),
     39         MIXED32(mixed(32)),
     40 
     41         LOWER512(lower(512)),
     42         UPPER512(upper(512)),
     43         MIXED512(mixed(512)),
     44 
     45         LOWER2048(lower(2048)),
     46         UPPER2048(upper(2048)),
     47         MIXED2048(mixed(2048)),
     48 
     49         LOWER_1M(lower(1024*1024)),
     50         UPPER_1M(upper(1024*1024)),
     51         MIXED_1M(mixed(1024*1024));
     52 
     53         final String value;
     54         private Inputs(String value) { this.value = value; }
     55         private static String lower(int length) {
     56             return makeString(length, "a0b1c2d3e4f5g6h7i8j9klmnopqrstuvwxyz");
     57         }
     58         private static String upper(int length) {
     59             return makeString(length, "A0B1C2D3E4F5G6H7I8J9KLMNOPQRSTUVWXYZ");
     60         }
     61         private static String mixed(int length) {
     62             return makeString(length, "Aa0Bb1Cc2Dd3Ee4Ff5Gg6Hh7Ii8Jj9KkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz");
     63         }
     64         private static String makeString(int length, String alphabet) {
     65             StringBuilder sb = new StringBuilder(length);
     66             for (int i = 0; i < length; ++i) {
     67                 sb.append(alphabet.charAt(i % alphabet.length()));
     68             }
     69             return sb.toString();
     70         }
     71     }
     72     @Param private Inputs s;
     73 
     74     public void timeToUpperCase_US(int reps) {
     75         for (int i = 0; i < reps; ++i) {
     76             s.value.toUpperCase(Locale.US);
     77         }
     78     }
     79 
     80     // toUpperCase for Greek is an extra-hard case that uses icu4c's Transliterator.
     81     public void timeToUpperCase_el_GR(int reps) {
     82         Locale el_GR = new Locale("el", "GR");
     83         for (int i = 0; i < reps; ++i) {
     84             s.value.toUpperCase(el_GR);
     85         }
     86     }
     87 
     88     public void timeToLowerCase_US(int reps) {
     89         for (int i = 0; i < reps; ++i) {
     90             s.value.toLowerCase(Locale.US);
     91         }
     92     }
     93 
     94     public void timeToUpperCase_Ascii(int reps) {
     95         for (int i = 0; i < reps; ++i) {
     96             toUpperCaseAscii(s.value);
     97         }
     98     }
     99 
    100     public void timeToLowerCase_Ascii(int reps) {
    101         for (int i = 0; i < reps; ++i) {
    102             toLowerCaseAscii(s.value);
    103         }
    104     }
    105 
    106     public void timeToUpperCase_ICU(int reps) {
    107         for (int i = 0; i < reps; ++i) {
    108             libcore.icu.ICU.toUpperCase(s.value, Locale.US);
    109         }
    110     }
    111 
    112     public void timeToLowerCase_ICU(int reps) {
    113         for (int i = 0; i < reps; ++i) {
    114             libcore.icu.ICU.toLowerCase(s.value, Locale.US);
    115         }
    116     }
    117 
    118     public static String toUpperCaseAscii(String s) {
    119         for (int i = 0, length = s.length(); i < length; i++) {
    120             char c = s.charAt(i);
    121             if (c < 'a' || c > 'z') {
    122                 continue; // fast path avoids allocation
    123             }
    124 
    125             // slow path: s contains lower case chars
    126             char[] result = s.toCharArray();
    127             for (; i < length; i++) {
    128                 c = result[i];
    129                 if (c >= 'a' && c <= 'z') {
    130                     result[i] -= ('a' - 'A');
    131                 }
    132             }
    133             return new String(result);
    134         }
    135         return s;
    136     }
    137 
    138     public static String toLowerCaseAscii(String s) {
    139         for (int i = 0, length = s.length(); i < length; i++) {
    140             char c = s.charAt(i);
    141             if (c < 'A' || c > 'Z') {
    142                 continue; // fast path avoids allocation
    143             }
    144 
    145             // slow path: s contains upper case chars
    146             char[] result = s.toCharArray();
    147             for (; i < length; i++) {
    148                 c = result[i];
    149                 if (c >= 'A' && c <= 'Z') {
    150                     result[i] += ('a' - 'A');
    151                 }
    152             }
    153             return new String(result);
    154         }
    155         return s;
    156     }
    157 }
    158