grep为什么还包括.和..在我的文件列表中,即使它与我给它的正则表达式不匹配？

Question

我想要做的是获取所有UTF16 Unicode图表的集合.我从http://unicode.org/charts/PDF/下载了所有pdf文件,并决定使用perl来删除所有特殊或UTF32图表,其中包含以下脚本:

#!/usr/bin/perl

opendir(my $dir, ".");
my @files = grep {!/^U[0-9,A-F]{4}\.pdf/ && !/utf16only.pl/} readdir($dir);
for $f (@files)
{
    print "deleting $f...\n";
    #unlink $f;
}
closedir($dir);

当我运行脚本时,我得到以下输出:

C:\Users\Evan\Downloads\Unicode 6.1 Charts>utf16only.pl
deleting ....
deleting .....
deleting 10FF80.pdf...
deleting ErrorLink.pdf...
deleting U10000.pdf...
deleting U100000.pdf...
deleting U10080.pdf...
deleting U10100.pdf...
deleting U10140.pdf...
deleting U10190.pdf...
deleting U101D0.pdf...
deleting U10280.pdf...
deleting U102A0.pdf...
deleting U10300.pdf...
deleting U10330.pdf...
deleting U10380.pdf...
deleting U103A0.pdf...
deleting U10400.pdf...
deleting U10450.pdf...
deleting U10480.pdf...
deleting U10800.pdf...
deleting U10840.pdf...
deleting U10900.pdf...
deleting U10920.pdf...
deleting U10980.pdf...
deleting U109A0.pdf...
deleting U10A00.pdf...
deleting U10A60.pdf...
deleting U10B00.pdf...
deleting U10B40.pdf...
deleting U10B60.pdf...
deleting U10C00.pdf...
deleting U10E60.pdf...
deleting U10FF80.pdf...
deleting U11000.pdf...
deleting U11080.pdf...
deleting U110D0.pdf...
deleting U11100.pdf...
deleting U11180.pdf...
deleting U11680.pdf...
deleting U12000.pdf...
deleting U12400.pdf...
deleting U13000.pdf...
deleting U16800.pdf...
deleting U16F00.pdf...
deleting U1B000.pdf...
deleting U1D000.pdf...
deleting U1D100.pdf...
deleting U1D200.pdf...
deleting U1D300.pdf...
deleting U1D360.pdf...
deleting U1D400.pdf...
deleting U1EE00.pdf...
deleting U1F000.pdf...
deleting U1F030.pdf...
deleting U1F0A0.pdf...
deleting U1F100.pdf...
deleting U1F200.pdf...
deleting U1F300.pdf...
deleting U1F600.pdf...
deleting U1F680.pdf...
deleting U1F700.pdf...
deleting U1FF80.pdf...
deleting U20000.pdf...
deleting U2A700.pdf...
deleting U2B740.pdf...
deleting U2F800.pdf...
deleting U2FF80.pdf...
deleting U3FF80.pdf...
deleting U4FF80.pdf...
deleting U5FF80.pdf...
deleting U6FF80.pdf...
deleting U7FF80.pdf...
deleting U8FF80.pdf...
deleting U9FF80.pdf...
deleting UAFF80.pdf...
deleting UBFF80.pdf...
deleting UBOOP.pdf...
deleting UCFF80.pdf...
deleting UDFF80.pdf...
deleting UE0000.pdf...
deleting UE0100.pdf...
deleting UEFF80.pdf...
deleting UF0000.pdf...
deleting UFFF80.pdf...

第2行仍然.和..和,我想是因为我想取消链接.,它删除大量我并不想删除的文件.我不知道的问题是与我regexs, ,grep,readdir或unlink,但它删除方式的文件超过它应该.

Answer 1

这一行:

grep {!/^U[0-9,A-F]{4}\.pdf/ && !/utf16only.pl/}

仅包含与正则表达式不匹配的文件.那包括.和...要排除这两个,你必须扩展它:

grep {!/^U[0-9,A-F]{4}\.pdf/ && !/utf16only.pl/ && !/^\.{1,2}$/}

Answer 2

这是你的正则表达式:

grep {!/^U[0-9,A-F]{4}\.pdf/ && !/utf16only.pl/} readdir($dir);

这说"匹配所有不匹配的 文件U[0-9,A-F]{4}.pdf(注意 - 你真的想要那个逗号吗？),而且也不是utf16only.pl."

由于.和..不匹配U [0-9A-F] {4} .pdf并且不匹配utf16only.pl,它们也会被删除.

!/^\./在您的添加grep中也可以从删除列表中排除这些文件:

grep {!/^U[0-9A-F]{4}\.pdf/ && !/^\./ && !/utf16only.pl/} readdir($dir);

注-我改变了你[0-9,A-F]来[0-9A-F],因为我不认为你的文件名会在他们的逗号.