|
|
Boost.RegexWorking With MFC/ATL String Types. |
|
The header <boost/regex/mfc.hpp> provides Boost.Regex support for MFC string types: note that this support requires Visual Studio .NET (Visual C++ 7) or later, where all of the MFC and ATL string types are based around the CSimpleStringT class template.
In the following documentation, whenever you see CSimpleStringT<charT>, then you can substitute any of the following MFC/ATL types (all of which inherit from CSimpleStringT):
CString
CStringA
CStringW
CAtlString
CAtlStringA
CAtlStringW
CStringT<charT,traits>
CFixedStringT<charT,N>
CSimpleStringT<charT>
The following typedefs are provided for the convenience of those working with TCHAR's:
typedef basic_regex<TCHAR> tregex; typedef match_results<TCHAR const*> tmatch; typedef regex_iterator<TCHAR const*> tregex_iterator; typedef regex_token_iterator<TCHAR const*> tregex_token_iterator;
If you are working with explicitly narrow or wide characters rather than TCHAR, then use the regular Boost.Regex types instead.
The following helper function is available to assist in the creation of a regular expression from an MFC/ATL string type:
template <class charT>
basic_regex<charT>
make_regex(const ATL::CSimpleStringT<charT>& s,
::boost::regex_constants::syntax_option_type f = boost::regex_constants::normal);
Effects: returns basic_regex<charT>(s.GetString(), s.GetString() + s.GetLength(), f);
For each regular expression algorithm that's overloaded for a std::basic_string argument, there is also one overloaded for the MFC/ATL string types. These algorithm signatures all look a lot more complex than they actually are, but for completeness here they are anyway:
There are two overloads, the first reports what matched in a match_results structure, the second does not.
All the usual caveats for regex_match apply, in particular the algorithm will only report a successful match if all of the input text matches the expression, if this isn't what you want then use regex_search instead.
template <class charT, class T, class A> bool regex_match( const ATL::CSimpleStringT<charT>& s, match_results<const B*, A>& what, const basic_regex<charT, T>& e, boost::regex_constants::match_flag_type f = boost::regex_constants::match_default);
Effects: returns ::boost::regex_match(s.GetString(), s.GetString() + s.GetLength(), what, e, f);
Example:
//
// Extract filename part of a path from a CString and return the result
// as another CString:
//
CString get_filename(const CString& path)
{
boost::tregex r(__T("(?:\\A|.*\\\\)([^\\\\]+)"));
boost::tmatch what;
if(boost::regex_match(path, what, r))
{
// extract $1 as a CString:
return CString(what[1].first, what.length(1));
}
else
{
throw std::runtime_error("Invalid pathname");
}
}
template <class charT, class T> bool regex_match( const ATL::CSimpleStringT<charT>& s, const basic_regex<B, T>& e, boost::regex_constants::match_flag_type f = boost::regex_constants::match_default)
Effects: returns ::boost::regex_match(s.GetString(), s.GetString() + s.GetLength(), e, f);
Example:
//
// Find out if *password* meets our password requirements,
// as defined by the regular expression *requirements*.
//
bool is_valid_password(const CString& password, const CString& requirements)
{
return boost::regex_match(password, boost::make_regex(requirements));
}
There are two additional overloads for regex_search, the first reports what matched the second does not:
template <class charT, class A, class T>
bool regex_search(const ATL::CSimpleStringT<charT>& s,
match_results<const charT*, A>& what,
const basic_regex<charT, T>& e,
boost::regex_constants::match_flag_type f = boost::regex_constants::match_default)
Effects: returns ::boost::regex_search(s.GetString(), s.GetString() + s.GetLength(), what, e, f);
Example:: Postcode extraction from an address string.
CString extract_postcode(const CString& address)
{
// searches throw address for a UK postcode and returns the result,
// the expression used is by Phil A. on www.regxlib.com:
boost::tregex r(__T("^(([A-Z]{1,2}[0-9]{1,2})|([A-Z]{1,2}[0-9][A-Z]))\\s?([0-9][A-Z]{2})$"));
boost::tmatch what;
if(boost::regex_search(address, what, r))
{
// extract $0 as a CString:
return CString(what[0].first, what.length());
}
else
{
throw std::runtime_error("No postcode found");
}
}
template <class charT, class T>
inline bool regex_search(const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT, T>& e,
boost::regex_constants::match_flag_type f = boost::regex_constants::match_default)
Effects: returns ::boost::regex_search(s.GetString(), s.GetString() + s.GetLength(), e, f);
There are two additional overloads for regex_replace, the first sends output to an output iterator, while the second creates a new string
template <class OutputIterator, class BidirectionalIterator, class traits, class
charT>
OutputIterator regex_replace(OutputIterator out,
BidirectionalIterator first,
BidirectionalIterator last,
const basic_regex<charT, traits>& e,
const ATL::CSimpleStringT<charT>& fmt,
match_flag_type flags = match_default)
Effects: returns ::boost::regex_replace(out, first, last, e, fmt.GetString(), flags);
template <class traits, charT>
ATL::CSimpleStringT<charT> regex_replace(const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT, traits>& e,
const ATL::CSimpleStringT<charT>& fmt,
match_flag_type flags = match_default)
Effects: returns a new string created using regex_replace, and the same memory manager as string s.
Example:
//
// Take a credit card number as a string of digits,
// and reformat it as a human readable string with "-"
// separating each group of four digits:
//
const boost::tregex e(__T("\\A(\\d{3,4})[- ]?(\\d{4})[- ]?(\\d{4})[- ]?(\\d{4})\\z"));
const CString human_format = __T("$1-$2-$3-$4");
CString human_readable_card_number(const CString& s)
{
return boost::regex_replace(s, e, human_format);
}
The following helper functions are provided to ease the conversion from an MFC/ATL string to a regex_iterator or regex_token_iterator:
template <class charT>
regex_iterator<charT const*>
make_regex_iterator(
const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT>& e,
::boost::regex_constants::match_flag_type f = boost::regex_constants::match_default);
Effects:returns regex_iterator(s.GetString(), s.GetString() + s.GetLength(), e, f);
Example:
void enumerate_links(const CString& html)
{
// enumerate and print all the links in some HTML text,
// the expression used is by Andew Lee on www.regxlib.com:
boost::tregex r(__T("href=[\"\']((http:\\/\\/|\\.\\/|\\/)?\\w+(\\.\\w+)*(\\/\\w+(\\.\\w+)?)*(\\/|\\?\\w*=\\w*(&\\w*=\\w*)*)?)[\"\']"));
boost::tregex_iterator i(boost::make_regex_iterator(html, r)), j;
while(i != j)
{
std::cout << (*i)[1] << std::endl;
++i;
}
}
template <class charT>
regex_token_iterator<charT const*>
make_regex_token_iterator(
const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT>& e,
int sub = 0,
::boost::regex_constants::match_flag_type f = boost::regex_constants::match_default);
Effects:returns regex_token_iterator(s.GetString(), s.GetString() + s.GetLength(), e, sub, f);
template <class charT>
regex_token_iterator<charT const*>
make_regex_token_iterator(
const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT>& e,
const std::vector<int>& subs,
::boost::regex_constants::match_flag_type f = boost::regex_constants::match_default);
Effects:returns regex_token_iterator(s.GetString(), s.GetString() + s.GetLength(), e, subs, f);
template <class charT, std::size_t N>
regex_token_iterator<charT const*>
make_regex_token_iterator(
const ATL::CSimpleStringT<charT>& s,
const basic_regex<charT>& e,
const int (& subs)[N],
::boost::regex_constants::match_flag_type f = boost::regex_constants::match_default);
Effects: returns regex_token_iterator(s.GetString(), s.GetString() + s.GetLength(), e, subs, f);
Example:
void enumerate_links2(const CString& html)
{
// enumerate and print all the links in some HTML text,
// the expression used is by Andew Lee on www.regxlib.com:
boost::tregex r(__T("href=[\"\']((http:\\/\\/|\\.\\/|\\/)?\\w+(\\.\\w+)*(\\/\\w+(\\.\\w+)?)*(\\/|\\?\\w*=\\w*(&\\w*=\\w*)*)?)[\"\']"));
boost::tregex_token_iterator i(boost::make_regex_token_iterator(html, r, 1)), j;
while(i != j)
{
std::cout << *i << std::endl;
++i;
}
}
Revised 21 Dec 2004
© Copyright John Maddock 2004
Use, modification and distribution are subject to the Boost Software License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)