有趣的 Bug Instances

[Xiaoven]

1. 我们特有的 False Positive (如 regex的不准确而造成的、无法获取上下文信息造成的等等）
2. spotbugs 和我们都有的 False Positive
3. spotbugs 特有的 False Negative (我们没有)
4. 我们特有的 False Negative (spotbugs 没有)
5. spotbugs 特有的 FPs （如由于 representation 造成的）

[Xiaoven]

类型 1

因为没有看import，所以造成这个错误

// https://github.com/magento/magento2-phpstorm-plugin/pull/466/files#diff-028a687773810be34d4f962c42ff2634925954908884573d0f2bfacf44a6e2aaR118
import com.magento.idea.magento2plugin.magento.packages.File;
...
final String[] handlerDirectories = messageQueueClassDataName.getPath().split(
                File.separator
        );

// https://github.com/magento/magento2-phpstorm-plugin/blob/2f786d1b203c81138df234d668f48e1fbb90798f/src/com/magento/idea/magento2plugin/magento/packages/File.java
package com.magento.idea.magento2plugin.magento.packages;

public class File {
    public static String separator = "/";
}

[Xiaoven]

类型 3

Spotbugs 使用的是 bytecode representation，无法检测 method call 类型的 self computation

实际跑的时候发现的

// https://github.com/qiniu/android-sdk/pull/453/files/7356a6c3eccb942fa90edc9e537b36b349ebc89a#diff-7f67181eef83b3f2ab332944daecf2a476d7f0f78028876136b5dd05641b0755

// https://github.com/qiniu/android-sdk/blob/7356a6c3eccb942fa90edc9e537b36b349ebc89a/library/src/main/java/com/qiniu/android/http/request/HttpSingleRequest.java#L177
double second = requestMetrics.endDate.getTime() - requestMetrics.endDate.getTime();    // duration 太短了，结果等于0

endDate 是 java.util.Date 类的，用 spotbugs 试验了以下代码，没有发现 warning

double main(java.util.Date endDate){
        double second = endDate.getTime() - endDate.getTime();
        return second;
    }

[Xiaoven]

还没有找到例子来确定，先划掉

类型 5

Spotbugs 使用 bytecode representation，根据它的 description 所言，Arrays.asArray(new String[] { "a" }).toArray() will return a String []. SpotBugs attempts to detect and suppress such cases, but may miss some.

而我们直接匹配形如 (String[]) c.toArray() 的代码，如果开发者使用了 Arrays.asArray(new String[] { "a" }) 的，就不会用 (String[]) 的强制转换，也就不会被我们的工具匹配到。

pattern 描述如下：

According to the Correctness, BC: Impossible downcast of toArray() result (BC_IMPOSSIBLE_DOWNCAST_OF_TOARRAY).

This code is casting the result of calling toArray() on a collection to a type more specific than Object[], as in:

String[] getAsArray(Collection<String> c) {
    return (String[]) c.toArray();
}

This will usually fail by throwing a ClassCastException. The toArray() of almost all collections return an Object[]. They can't really do anything else, since the Collection object has no reference to the declared generic type of the collection.

The correct way to do get an array of a specific type from a collection is to use c.toArray(new String[]); or c.toArray(new String[c.size()]); (the latter is slightly more efficient).

There is one common/known exception to this. The toArray() method of lists returned by Arrays.asList(...) will return a covariantly typed array. For example, Arrays.asArray(new String[] { "a" }).toArray() will return a String [].

[Xiaoven]

类型 3

RV: Random value from 0 to 1 is coerced to the integer 0 (RV_01_TO_INT)

int main(int size, Random r){
        return (int) r.nextFloat() * size;
    }

spotbugs 检查 Math.random() 和 randomObject.nextDouble() 方法，但似乎忘了检查 randomObject.nextFloat() 方法

[Xiaoven]

类型 1

DLS_DEAD_LOCAL_INCREMENT_IN_RETURN

无法区分 field 和 local var

// https://github.com/MaximSinyak/Java/pull/6/files/3ef2bc97335cb0a460f77ff812a42223b055b539#diff-91797b67aa94c9428c50548a3fdb1dd145643d7dbe93fe4296019f7c70435e0a
@@ -0,0 +1,31 @@
 package ru.geekbrains.lesson_6;

 public class Main {
     private static int countCat = 0;
     private static int countDog = 0;

      public static void main(String[] args){
          makeCat("Barsik", 30, 40);
          makeCat("Murzik", 250, 4);
          makeDog("Tuzik", 400, 4);
          makeDog("Sharik", 600, 9);
          makeDog("Barbos", 300, 20);
 System.out.println("Создано: кошек - " + countCat +  "; собак - " + countDog);

      }

      public static int makeCat(String name, int run, int swim){
          Cat cat = new Cat(name, run, swim);
          cat.run();
          cat.swim();
          return countCat++;

[Xiaoven]

类型 3

DLS_OVERWRITTEN_INCREMENT

设计阶段发现的 FP

i = 2 + (i++);

spotbugs 只能检查 i = i++; 和 i = i--; 这种简单的 expression

[Xiaoven]

类型 3 spotbugs 特有的 False Negative (我们没有)

SA_SELF_COMPUTATION

// https://github.com/wensiqun/asmsupport/blob/master/asmsupport-sample/src/main/java/cn/wensiqun/asmsupport/sample/core/operators/LogicalOperatorGenerate.java#L37
			System.out.printf("%s\n%s: %b\n%s: %b\n%s: %b\n%s: %b\n\n",
				"Boolean logical AND (&)", "false & false", (false & false),
				"false & true", (false & true), "true & false", (true & false),
				"true & true", (true & true));

[stan6]

类型 3

DLS_OVERWRITTEN_INCREMENT

设计阶段发现的 FP

i = 2 + (i--);

spotbugs 只能检查 i = i++;

Need to file a new bug report. In the bug description of DLS_DEAD_LOCAL_INCREMENT_IN_RETURN, http://findbugs.sourceforge.net/bugDescriptions.html, they mentioned increment/decrement but they didn't actually support any decrement based on their code at
https://github.com/spotbugs/spotbugs/blob/a6f9acb2932b54f5b70ea8bc206afb552321a222/spotbugs/src/main/java/edu/umd/cs/findbugs/detect/FindDeadLocalStores.java. We can file one GItHub issue complaining about these two patterns DLS_OVERWRITTEN_INCREMENT and DLS_DEAD_LOCAL_INCREMENT_IN_RETURN because it o

[stan6]

To check: Inconsistent documentation
类型 3

DLS_DEAD_LOCAL_INCREMENT_IN_RETURN
DLS_OVERWRITTEN_INCREMENT
Both patterns do not check for decrement in their code (they used IINC to check for the increment instruction) but DLS_DEAD_LOCAL_INCREMENT_IN_RETURN mentioned "decrement" in their description and DLS_OVERWRITTEN_INCREMENT didn't mention. Need to check if both patterns support decrement.

[zhouyinn]

increment / decrement 相关的 bug patterns

DLS_DEAD_LOCAL_INCREMENT_IN_RETURN
DLS_OVERWRITTEN_INCREMENT
VO_VOLATILE_INCREMENT
QF_QUESTIONABLE_FOR_LOOP

all of them support ++ and -- detection

//DLS_DEAD_LOCAL_INCREMENT_IN_RETURN for ++
    int testDLS_DEAD_LOCAL_INCREMENT_IN_RETURN(){
        int i = 0;
        return i++;
    }

    //DLS_DEAD_LOCAL_INCREMENT_IN_RETURN for --
    int testDLS_DEAD_LOCAL_INCREMENT_IN_RETURN1(){
        int i = 0;
        return i--;
    }

    //DLS_OVERWRITTEN_INCREMENT for ++
    void testDLS_OVERWRITTEN_INCREMENT(){
        int i = 0;
        i = i++;
    }

    //DLS_OVERWRITTEN_INCREMENT for --
    void testDLS_OVERWRITTEN_INCREMENT1(){
        int i = 0;
        i = i--;
    }

    //VO_VOLATILE_INCREMENT for ++
    private volatile int x = 0;
    void testVO_VOLATILE_INCREMENT() {
        try {
            x++; // Warning -- False Positive!
        } finally {
        }
    }

    //VO_VOLATILE_INCREMENT for --
    void testVO_VOLATILE_INCREMENT1() {
        try {
            x--; // Warning -- False Positive!
        } finally {
        }
    }


    //QF_QUESTIONABLE_FOR_LOOP for ++
    void testQF_QUESTIONABLE_FOR_LOOP(){
        int i = 0;
        for(int j = 0; j < 10; i++){
            System.out.println("Hi");
            j++;
        }
    }

    //QF_QUESTIONABLE_FOR_LOOP for --
    void testQF_QUESTIONABLE_FOR_LOOP1(){
        int i = 0;
        for(int j = 0; j < 10; i--){
            System.out.println("Hi");
            j--;
        }
    }

only the description of DLS_DEAD_LOCAL_INCREMENT_IN_RETURN mentioned **increment|decrement**

[zhouyinn]

复现 SA_LOCAL_SELF_COMPUTATION
@stan6

int testSA_FIELD_SELF_COMPUTATION(){
      int flags = 1;
      return flags ^ flags; // warning SA_LOCAL_SELF_COMPUTATION
  }

  int testSA_FIELD_SELF_COMPUTATION1(){
      int flags = 1;
      return flags & flags; // no warning (False negative)
  }

  int testSA_FIELD_SELF_COMPUTATION2(){
      int flags = 1;
      return flags - flags; // no warning (False negative)
  }

  int testSA_FIELD_SELF_COMPUTATION3(){
      int flags = 1;
      return flags + flags; // no warning (False negative)
  }

  double testSA_FIELD_SELF_COMPUTATION4(java.util.Date endDate){
      // Should report nonsensical self computation
      double temp = endDate.getTime();
      double second = temp - temp; // no warning (False negative)
      return second;
  }

[zhouyinn]

In oracle/graal#3288, the author complains that the warning that rbugs reported is a false positive.

As can be seen from the figure, SpotBugs also reports the same bug.