I have used "[\x0-\x7f]" instead of "/p{IsBasicLatin}" to construct the regular expression (expression=boost::make_u32regex("[\\x0-\\x7f]" )). The regular expression has been constructed correctly but it cannot accecpt instance string either "a" or "	" (boost:u32match("a",expression)==false).I am wondering whether it has something to do with unicode? I have tried expression=boost::regex("[\\x0-\\x7f]" )); then I can pass the string "a" but not string "	"(boost:match("a",expression)==true), which I think is reaonable for boost:regex since it does not support the unicode. So my point is: why the boost:u32match doesn't work well?

在2008-08-08 16:48:38，"John Maddock" <john@johnmaddock.co.uk> 写道： >John Maddock wrote: >>> gj_uestc wrote: >>>>> Hi,all >>>>> Nowadays I am using boost:u32regex to do some regular expression >>>>> processing.But it seems that "/p{IsBasicLatin}"is not a accessable >>>>> expression by boost::make_u32regex(tmp).Does boost:regex not suppor >>>>> the named unicode blocks or I have to pass some other flags to the >>>>> library? Now I was using the defult flag wich indicate using perl >>>>> syntactic. >>> >>> The named properties/character classes supported are here: >>> http://www.boost.org/doc/libs/1_35_0/libs/regex/doc/html/boost_regex/syntax/character_classes/optional_char_class_names.html >>> >>> As you can see I haven't added support for language-specific blocks >>> yet :-( > >I forgot to mention that \p{IsBasicLatin} is the same as: [\x0-\x7f], >likewise the other continuous blocks can be expressed in the same way. > >HTH, John. >