113 lines
3.3 KiB
C++
113 lines
3.3 KiB
C++
|
/** @file
|
||
|
|
|
||
|
|
$Id$
|
||
|
|
|
||
|
|
$Date$
|
||
|
|
$Author$
|
||
|
|
|
||
|
|
@copy © Marc Wäckerlin
|
||
|
|
@license LGPL, see file <a href="license.html">COPYING</a>
|
||
|
|
|
||
|
|
$Log$
|
||
|
|
Revision 1.1 2004/12/14 20:20:30 marc
|
||
|
|
initial version
|
||
|
|
|
||
|
|
|
||
|
|
*/
|
||
|
|
|
||
|
|
#include <regex.h>
|
||
|
|
#include <string>
|
||
|
|
#include <stdexcept>
|
||
|
|
|
||
|
|
namespace mrw {
|
||
|
|
|
||
|
|
/** @defgroup regexp Regular Expressions
|
||
|
|
|
||
|
|
A simple wrapper around the C POSIX regular expression library
|
||
|
|
with a C++ Interface.
|
||
|
|
|
||
|
|
Usage sample:
|
||
|
|
@code
|
||
|
|
std::ifstream file(filename); // read from a file
|
||
|
|
mrw::RegExp findBrackets("^\\[.*\\]$"); // look for "[some text]"
|
||
|
|
for (std::string line;
|
||
|
|
mrw::getline(file, line);) // read line by line
|
||
|
|
if (findBrackets(line)) // check for regular expression
|
||
|
|
found(line); // found, do something
|
||
|
|
@endcode
|
||
|
|
|
||
|
|
*/
|
||
|
|
|
||
|
|
/** @brief a regular expression
|
||
|
|
|
||
|
|
This class manages a simple regular expression.
|
||
|
|
|
||
|
|
Usage sample:
|
||
|
|
@code
|
||
|
|
std::ifstream file(filename); // read from a file
|
||
|
|
mrw::RegExp findBrackets("^\\[.*\\]$"); // look for "[some text]"
|
||
|
|
for (std::string line;
|
||
|
|
mrw::getline(file, line);) // read line by line
|
||
|
|
if (findBrackets(line)) // check for regular expression
|
||
|
|
found(line); // found, do something
|
||
|
|
@endcode
|
||
|
|
|
||
|
|
*/
|
||
|
|
class RegExp {
|
||
|
|
|
||
|
|
public:
|
||
|
|
|
||
|
|
/** @brief flags that influence regular expressions
|
||
|
|
|
||
|
|
Flag @c newline treats a newline in the text to be compared as
|
||
|
|
dividing the text into multiple lines, so that @c $ can match
|
||
|
|
before the newline and @c ^ can match after. Also, don't
|
||
|
|
permit @c . to match a newline, and don't permit @c [^...] to
|
||
|
|
match a newline.
|
||
|
|
|
||
|
|
Otherwise, newline acts like any other ordinary character.
|
||
|
|
|
||
|
|
Flag @c nosub is used internally, don't specify it.
|
||
|
|
|
||
|
|
*/
|
||
|
|
enum Flags {
|
||
|
|
extended = REG_EXTENDED, ///< use extended regular expressions
|
||
|
|
icase = REG_ICASE, ///< ignore case in match
|
||
|
|
nosub = REG_NOSUB, ///< @b internal (report only success/fail)
|
||
|
|
newline = REG_NEWLINE ///< treat lines individually
|
||
|
|
};
|
||
|
|
|
||
|
|
/** @brief define a new regular expression
|
||
|
|
|
||
|
|
The regular expression is compiled on instanciation and can
|
||
|
|
then be matced several times on different texts.
|
||
|
|
|
||
|
|
@param pattern the regular expression pattern, thee the @c man
|
||
|
|
page for POSIX regular expressions (on linux: @c
|
||
|
|
info 7 regex)
|
||
|
|
@param flags special flags, they default to extended|nosub and
|
||
|
|
should consist of the Flag values combined with @c |
|
||
|
|
|
||
|
|
@throw std::invalid_argument if pattern compilation fails
|
||
|
|
*/
|
||
|
|
RegExp(const std::string& pattern, int flags = extended)
|
||
|
|
throw(std::exception, std::bad_exception);
|
||
|
|
|
||
|
|
/** @brief cleans up expression from memory */
|
||
|
|
~RegExp() throw();
|
||
|
|
|
||
|
|
/** @brief apply the expression, match a text against the expression
|
||
|
|
|
||
|
|
@param text a text that is matched against the regular expression
|
||
|
|
@return
|
||
|
|
- true if @c text matches
|
||
|
|
- false otherwise */
|
||
|
|
bool operator()(const std::string& text) const throw();
|
||
|
|
|
||
|
|
private:
|
||
|
|
|
||
|
|
regex_t _regex;
|
||
|
|
|
||
|
|
};
|
||
|
|
}
|