Linux:如何比较目录中的所有文件?

时间:2022-12-16 12:06:12

For my homework, I have to check if two files in a directory have the same contents and if so, replace one with a hardlink to the other. My script looks like:

对于我的作业,我必须检查目录中的两个文件是否具有相同的内容,如果是,请用另一个硬链接替换另一个。我的脚本看起来像:

    cd $1 # $1 is the directory this script executes in
    FILES=`find . -type f`
    for line1 in $FILES
    do
       for line2 in $FILES
       do
         (check the two files with cmp)
       done
    done

My problem is that I can not figure out the conditional expression to make sure that the two files are not the same: If the directory has files a,b,c, and d it should not return true for checking a against a. How do I do this?

我的问题是我无法弄清楚条件表达式以确保两个文件不相同:如果目录有文件a,b,c和d,则不应该返回true来检查a。我该怎么做呢?

Edit: So I've got this:

编辑:所以我有这个:

cmp $line1 $line2 > /dev/null
if [ $? -eq 0 -a "$line1" != "$line2" ]

But it counts the files twice: It checks a and b, and then b and a. for some reason, using < with strings does not work.

但它会对文件进行两次计数:它检查a和b,然后检查b和a。出于某种原因,使用

Edit: I think I figured it out, the solution is to use a \ before the <

编辑:我想我弄清楚了,解决方案是在 <之前使用\< p>

2 个解决方案

#1


1  

Using test, or its alias [:

使用test或其别名[:

if [ "$line1" < "$line2" ]
then
    check the files
fi

Note that I'm using < here instead of != (which would otherwise work) so that, once you've compared a with b, you won't later compare b with a.

请注意,我正在使用 而不是!=(否则会起作用),这样,一旦你将a与b进行比较,你就不会在后来将b与a进行比较。

#2


0  

Here is an optimized way to do it that also properly handle files with embedded spaces or similar:

以下是一种优化的方法,可以正确处理带有嵌入空格或类似内容的文件:

find . -type f -exec sh -c '
compare() {
  first=$1
  shift
  for i do
    cmp -s "$first" "$i" || printf " %s and %s differ\n" "$first" "$i"
  done
}
while [ $# -gt 1 ]; do
  compare "$@"
  shift
done ' sh {} +

#1


1  

Using test, or its alias [:

使用test或其别名[:

if [ "$line1" < "$line2" ]
then
    check the files
fi

Note that I'm using < here instead of != (which would otherwise work) so that, once you've compared a with b, you won't later compare b with a.

请注意,我正在使用 而不是!=(否则会起作用),这样,一旦你将a与b进行比较,你就不会在后来将b与a进行比较。

#2


0  

Here is an optimized way to do it that also properly handle files with embedded spaces or similar:

以下是一种优化的方法,可以正确处理带有嵌入空格或类似内容的文件:

find . -type f -exec sh -c '
compare() {
  first=$1
  shift
  for i do
    cmp -s "$first" "$i" || printf " %s and %s differ\n" "$first" "$i"
  done
}
while [ $# -gt 1 ]; do
  compare "$@"
  shift
done ' sh {} +