FilenameUtils.getName 函数源码分析

一、背景

二、源码分析

2.1 问题1：为什么需要 NonNul 检查？

2.1.1 怎么检查的？

2.1.2 为什么要做这个检查呢？

2.2 问题2：为什么不根据当前系统类型来获取分隔符？

三、Zoom Out

3.1 代码健壮性

3.2 代码严谨性

3.3 如何写注释

四、总结

一、背景

最近用到了 org.apache.commons.io.FilenameUtils#getName 这个方法，该方法可以传入文件路径，获取文件名。简单看了下源码，虽然并不复杂，但和自己设想略有区别，值得学习，本文简单分析下。

二、源码分析

org.apache.commons.io.FilenameUtils#getName

 /**
     * Gets the name minus the path from a full fileName.
     * <p>
     * This method will handle a file in either Unix or Windows format.
     * The text after the last forward or backslash is returned.
     * 
<pre>
     * a/b/c.txt --&gt; c.txt
     * a.txt     --&gt; a.txt
     * a/b/c     --&gt; c
     * a/b/c/    --&gt; ""
     * </pre>
* <p>
     * The output will be the same irrespective of the machine that the code is running on.
     *
     * @param fileName  the fileName to query, null returns null
     * @return the name of the file without the path, or an empty string if none exists.
     * Null bytes inside string will be removed
     */
    public static String getName(final String fileName) {
     // 传入 null 直接返回 null 
        if (fileName == null) {
            return null;
        }
        // NonNul 检查
        requireNonNullChars(fileName);
       //  查找最后一个分隔符
        final int index = indexOfLastSeparator(fileName);
     // 从最后一个分隔符窃到最后
        return fileName.substring(index + 1);
    }

2.1 问题1：为什么需要 NonNul 检查？ 2.1.1 怎么检查的？

org.apache.commons.io.FilenameUtils#requireNonNullChars

   /**
     * Checks the input for null bytes, a sign of unsanitized data being passed to to file level functions.
     *
     * This may be used for poison byte attacks.
     *
     * @param path the path to check
     */
    private static void requireNonNullChars(final String path) {
        if (path.indexOf(0) >= 0) {
            throw new IllegalArgumentException("Null byte present in file/path name. There are no "
                + "known legitimate use cases for such data, but several injection attacks may use it");
        }
    }

java.lang.String#indexOf(int) 源码：

 /**
     * Returns the index within this string of the first occurrence of
     * the specified character. If a character with value
     * {@code ch} occurs in the character sequence represented by
     * this {@code String} object, then the index (in Unicode
     * code units) of the first such occurrence is returned. For
     * values of {@code ch} in the range from 0 to 0xFFFF
     * (inclusive), this is the smallest value <i>k</i> such that:
     * <blockquote><pre>
     * this.charAt(<i>k</i>) == ch
     * </pre></blockquote>
     * is true. For other values of {@code ch}, it is the
     * smallest value <i>k</i> such that:
     * <blockquote><pre>
     * this.codePointAt(<i>k</i>) == ch
     * </pre></blockquote>
     * is true. In either case, if no such character occurs in this
     * string, then {@code -1} is returned.
     *
     * @param   ch   a character (Unicode code point).
     * @return  the index of the first occurrence of the character in the
     *          character sequence represented by this object, or
     *          {@code -1} if the character does not occur.
     */
    public int indexOf(int ch) {
        return indexOf(ch, 0);
    }

可知，indexOf(0) 目的是查找 ASCII 码为 0 的字符的位置，如果找到则抛出 IllegalArgumentException异常。搜索 ASCII 对照表，得知 ASCII 值为 0 代表控制字符 NUT，并不是常规的文件名所应该包含的字符。

2.1.2 为什么要做这个检查呢？

null 字节是一个值为 0 的字节，如十六进制中的 0x00。存在与 null 字节有关的安全漏洞。因为 C 语言中使用 null 字节作为字符串终结符，而其他语言（Java，PHP等）没有这个字符串终结符；例如，Java Web 项目只允许用户上传 .webp 格式的图片，但利用这个漏洞就可以上传 .jsp 文件。如用户上传 hack.jsp<NUL>.webp 文件， Java 会认为符合 .webp 格式，实际调用 C 语言系统函数写入磁盘时讲当做字符串分隔符，结果将文件保存为 hack.jsp。有些编程语言不允许在文件名中使用 ·· <NUL>，如果你使用的编程语言没有对此处理，就需要自己去处理。因此，这个检查很有必要。

代码示例：

package org.example;
import org.apache.commons.io.FilenameUtils;
public class FilenameDemo {
    public static void main(String[] args) {
        String filename= "hack.jsp\0.webp";
        System.out.println( FilenameUtils.getName(filename));
    }
}

报错信息：

Exception in thread "main" java.lang.IllegalArgumentException: Null byte present in file/path name. There are no known legitimate use cases for such data, but several injection attacks may use it
   at org.apache.commons.io.FilenameUtils.requireNonNullChars(FilenameUtils.java:998)
   at org.apache.commons.io.FilenameUtils.getName(FilenameUtils.java:984)
   at org.example.FilenameDemo.main(FilenameDemo.java:8)

如果去掉校验：

package org.example;
import org.apache.commons.io.FilenameUtils;
public class FilenameDemo {
    public static void main(String[] args) {
        String filename= "hack.jsp\0.webp";
        // 不添加校验
        String name = getName(filename);
        // 获取拓展名
        String extension = FilenameUtils.getExtension(name);
        System.out.println(extension);
    }
    public static String getName(final String fileName) {
        if (fileName == null) {
            return null;
        }
        final int index = FilenameUtils.indexOfLastSeparator(fileName);
        return fileName.substring(index + 1);
    }
}

Java 的确会将拓展名识别为 jpg

jpg

JDK 8 及其以上版本试图创建 hack.jsp\0.webp 的文件时，底层也会做类似的校验，无法创建成功。

大家感兴趣可以试试使用 C 语言写入名为 hack.jsp\0.webp 的文件，最终很可能文件名为 hack.jsp。

2.2 问题2：为什么不根据当前系统类型来获取分隔符？

查找最后一个分隔符 org.apache.commons.io.FilenameUtils#indexOfLastSeparator

 /**
     * Returns the index of the last directory separator character.
     * <p>
     * This method will handle a file in either Unix or Windows format.
     * The position of the last forward or backslash is returned.
     * <p>
     * The output will be the same irrespective of the machine that the code is running on.
     *
     * @param fileName  the fileName to find the last path separator in, null returns -1
     * @return the index of the last separator character, or -1 if there
     * is no such character
     */
    public static int indexOfLastSeparator(final String fileName) {
        if (fileName == null) {
            return NOT_FOUND;
        }
        final int lastUnixPos = fileName.lastIndexOf(UNIX_SEPARATOR);
        final int lastWindowsPos = fileName.lastIndexOf(WINDOWS_SEPARATOR);
        return Math.max(lastUnixPos, lastWindowsPos);
    }

该方法的语义是获取文件名，那么从函数的语义层面上来说，不管是啥系统的文件分隔符都必须要保证得到正确的文件名。试想一下，在 Windows 系统上调用该函数，传入一个 Unix 文件路径，得不到正确的文件名合理吗？函数设计本身就应该考虑兼容性。因此不能获取当前系统的分隔符来截取文件名。源码中分别获取 Window 和 Unix 分隔符，有哪个用哪个，显然更加合理。

三、Zoom Out 3.1 代码健壮性

我们日常编码时，要做防御性编程，对于错误的、非法的输入都要做好预防。

3.2 代码严谨性

我们写代码一定不要想当然。我们先想清楚这个函数究竟要实现怎样的功能，而且不是做一个 “CV 工程师”，无脑“拷贝”代码。同时，我们也应该写好单测，充分考虑各种异常 Case ，保证正常和异常的 Case 都覆盖到。

3.3 如何写注释

org.apache.commons.io.FilenameUtils#requireNonNullChars 函数注释部分就给出了这么设计的原因：This may be used for poison byte attacks.

注释不应该“喃喃自语”讲一些显而易见的废话。对于容易让人困惑的设计，一定要通过注释讲清楚设计原因。

此外，结合工作经验，推荐一些其他注释技巧：（1）对于稍微复杂或者重要的设计，可以通过注释给出核心的设计思路；如： java.util.concurrent.ThreadPoolExecutor#execute

    /**
     * Executes the given task sometime in the future.  The task
     * may execute in a new thread or in an existing pooled thread.
     *
     * If the task cannot be submitted for execution, either because this
     * executor has been shutdown or because its capacity has been reached,
     * the task is handled by the current {@link RejectedExecutionHandler}.
     *
     * @param command the task to execute
     * @throws RejectedExecutionException at discretion of
     *         {@code RejectedExecutionHandler}, if the task
     *         cannot be accepted for execution
     * @throws NullPointerException if {@code command} is null
     */
    public void execute(Runnable command) {
        if (command == null)
            throw new NullPointerException();
        /*
         * Proceed in 3 steps:
         *
         * 1. If fewer than corePoolSize threads are running, try to
         * start a new thread with the given command as its first
         * task.  The call to addWorker atomically checks runState and
         * workerCount, and so prevents false alarms that would add
         * threads when it shouldn't, by returning false.
         *
         * 2. If a task can be successfully queued, then we still need
         * to double-check whether we should have added a thread
         * (because existing ones died since last checking) or that
         * the pool shut down since entry into this method. So we
         * recheck state and if necessary roll back the enqueuing if
         * stopped, or start a new thread if there are none.
         *
         * 3. If we cannot queue task, then we try to add a new
         * thread.  If it fails, we know we are shut down or saturated
         * and so reject the task.
         */
        int c = ctl.get();
        if (workerCountOf(c) < corePoolSize) {
            if (addWorker(command, true))
                return;
            c = ctl.get();
        }
        if (isRunning(c) && workQueue.offer(command)) {
            int recheck = ctl.get();
            if (! isRunning(recheck) && remove(command))
                reject(command);
            else if (workerCountOf(recheck) == 0)
                addWorker(null, false);
        }
        else if (!addWorker(command, false))
            reject(command);
    }

（2)对于关联的代码，可以使用 @see 或者 {@link } 的方式，在代码中提供关联代码的快捷跳转方式。

    /**
     * Sets the core number of threads.  This overrides any value set
     * in the constructor.  If the new value is smaller than the
     * current value, excess existing threads will be terminated when
     * they next become idle.  If larger, new threads will, if needed,
     * be started to execute any queued tasks.
     *
     * @param corePoolSize the new core size
     * @throws IllegalArgumentException if {@code corePoolSize < 0}
     *         or {@code corePoolSize} is greater than the {@linkplain
     *         #getMaximumPoolSize() maximum pool size}
     * @see #getCorePoolSize
     */
    public void setCorePoolSize(int corePoolSize) {
        if (corePoolSize < 0 || maximumPoolSize < corePoolSize)
            throw new IllegalArgumentException();
        int delta = corePoolSize - this.corePoolSize;
        this.corePoolSize = corePoolSize;
        if (workerCountOf(ctl.get()) > corePoolSize)
            interruptIdleWorkers();
        else if (delta > 0) {
            // We don't really know how many new threads are "needed".
            // As a heuristic, prestart enough new workers (up to new
            // core size) to handle the current number of tasks in
            // queue, but stop if queue becomes empty while doing so.
            int k = Math.min(delta, workQueue.size());
            while (k-- > 0 && addWorker(null, true)) {
                if (workQueue.isEmpty())
                    break;
            }
        }
    }

（2）在日常业务开发中，非常推荐讲相关的文档、配置页面链接也放到注释中，极大方便后期维护。如：

    /**
     * 某某功能
     *
     * 相关文档：
     * <a href="https://blog.csdn.net/w605283073" rel="external nofollow"  rel="external nofollow" >设计文档</a>
     * <a href="https://blog.csdn.net/w605283073" rel="external nofollow"  rel="external nofollow" >三方API地址</a>
     */
    public void demo(){
        // 省略
    }

（4)对于工具类可以考虑讲给出常见的输入对应的输出。如 org.apache.commons.lang3.StringUtils#center(java.lang.String, int, char)

 /**
     * <p>Centers a String in a larger String of size {@code size}.
     * Uses a supplied character as the value to pad the String with.</p>
     *
     * <p>If the size is less than the String length, the String is returned.
     * A {@code null} String returns {@code null}.
     * A negative size is treated as zero.</p>
     *
     * <pre>
     * StringUtils.center(null, *, *)     = null
     * StringUtils.center("", 4, ' ')     = "    "
     * StringUtils.center("ab", -1, ' ')  = "ab"
     * StringUtils.center("ab", 4, ' ')   = " ab "
     * StringUtils.center("abcd", 2, ' ') = "abcd"
     * StringUtils.center("a", 4, ' ')    = " a  "
     * StringUtils.center("a", 4, 'y')    = "yayy"
     * </pre>
     *
     * @param str  the String to center, may be null
     * @param size  the int size of new String, negative treated as zero
     * @param padChar  the character to pad the new String with
     * @return centered String, {@code null} if null String input
     * @since 2.0
     */
    public static String center(String str, final int size, final char padChar) {
        if (str == null || size <= 0) {
            return str;
        }
        final int strLen = str.length();
        final int pads = size - strLen;
        if (pads <= 0) {
            return str;
        }
        str = leftPad(str, strLen + pads / 2, padChar);
        str = rightPad(str, size, padChar);
        return str;
    }

（5) 对于废弃的方法，一定要注明废弃的原因，给出替代方案。如：java.security.Signature#setParameter(java.lang.String, java.lang.Object)

    /**
     * 省略部分
     * 
     * @see #getParameter
     *
     * @deprecated Use
     * {@link #setParameter(java.security.spec.AlgorithmParameterSpec)
     * setParameter}.
     */
    @Deprecated
    public final void setParameter(String param, Object value)
            throws InvalidParameterException {
        engineSetParameter(param, value);
    }

四、总结

很多优秀的开源项目的代码设计都非常严谨，往往简单的代码中也蕴藏着缜密的思考。我们有时间可以看看一些优秀的开源项目，可以从简单的入手，可以先想想如果自己写大概该如何实现，然后和作者的实现思路对比，会有更大收获。平时看源码时，不仅要知道源码长这样，更要了解为什么这么设计。

以上就是FilenameUtils.getName 函数源码分析的详细内容，更多关于FilenameUtils.getName 函数的资料请关注易知道(ezd.cc)其它相关文章！