Iconv::wget保存的文件名中的非法序列(Ruby 1.9.2)
我正在用wget使用普通标志将文件保存到ext2分区。检索他们的姓名有时会失败:Iconv::wget保存的文件名中的非法序列(Ruby 1.9.2),ruby,utf-8,find,wget,iconv,Ruby,Utf 8,Find,Wget,Iconv,我正在用wget使用普通标志将文件保存到ext2分区。检索他们的姓名有时会失败: s = `find "#{@dir}" -type f -printf "%T@\t::::\t%s\t::::\t%p\n" |sort` s.each_line {|l| file_name = l.chomp.split("\t::::\t")[2] #=> # ...66:in `split': invalid byte sequence in UTF-8 (ArgumentError)
s = `find "#{@dir}" -type f -printf "%T@\t::::\t%s\t::::\t%p\n" |sort`
s.each_line {|l|
file_name = l.chomp.split("\t::::\t")[2] #=>
# ...66:in `split': invalid byte sequence in UTF-8 (ArgumentError)
}
测试:
l.encoding #=> UTF-8
l.valid_encoding #=> false
l.inspect #=> "...St. Paul\xE2%80%99s Cathedral..."
Iconv.conv('utf-8', 'utf-8', l) #=>
# ...77:in `conv': "\xE2%80%99s Cathedr"... (Iconv::IllegalSequence)
如何获取文件名并删除该文件
忘了提及,在bash中,文件如下所示:
index.php?attTag=St. Paul?%80%99s Cathedral
将此字符串粘贴回ls不会返回此类文件或目录。在运行转换之前,您可以尝试
CGI.unescape
a = "...St. Paul\xE2%80%99s Cathedral..."
puts a
require 'cgi'
b = CGI.unescape a
puts b
require 'iconv'
c = Iconv.conv('UTF-8//TRANSLIT', 'UTF-8', b) # may not even be necessary
puts c
在我的ruby-1.9.2-p180上有哪些输出:
...St. Paul?%80%99s Cathedral...
...St. Paul’s Cathedral...
...St. Paul’s Cathedral...
我必须先做“a.force_编码('us-ascii'),它成功了。谢谢。对不起,我太匆忙了。这不是bash中文件名的外观。我已经在我的问题中添加了这个。原因可能是wget中的一个故意错误--。