Ejemplo n.º 1
0
int RegularExpression::match(const std::string& subject, std::string::size_type offset, Match& mtch, int options) const
{
	poco_assert (offset <= subject.length());

	int ovec[OVEC_SIZE];
	int rc = pcre_exec(_pcre, _extra, subject.c_str(), int(subject.size()), int(offset), options & 0xFFFF, ovec, OVEC_SIZE);
	if (rc == PCRE_ERROR_NOMATCH)
	{
		mtch.offset = std::string::npos;
		mtch.length = 0;
		return 0;
	}
	else if (rc == PCRE_ERROR_BADOPTION)
	{
		throw RegularExpressionException("bad option");
	}
	else if (rc == 0)
	{
		throw RegularExpressionException("too many captured substrings");
	}
	else if (rc < 0)
	{
		std::ostringstream msg;
		msg << "PCRE error " << rc;
		throw RegularExpressionException(msg.str());
	}
	mtch.offset = ovec[0] < 0 ? std::string::npos : ovec[0];
	mtch.length = ovec[1] - mtch.offset;
	return rc;
}
Ejemplo n.º 2
0
RegularExpression::RegularExpression(const std::string& pattern, int options, bool study): _pcre(0), _extra(0)
{
	const char* error;
	int offs;
	_pcre = pcre_compile(pattern.c_str(), options, &error, &offs, 0);
	if (!_pcre)
	{
		std::ostringstream msg;
		msg << error << " (at offset " << offs << ")";
		throw RegularExpressionException(msg.str());
	}
	if (study)
		_extra = pcre_study(_pcre, 0, &error);
}
Ejemplo n.º 3
0
int RegularExpression::match(const std::string& subject, std::string::size_type offset, MatchVec& matches, int options) const
{
	poco_assert (offset <= subject.length());

	matches.clear();

	int ovec[OVEC_SIZE];
	int rc = pcre_exec(_pcre, _extra, subject.c_str(), int(subject.size()), int(offset), options & 0xFFFF, ovec, OVEC_SIZE);
	if (rc == PCRE_ERROR_NOMATCH)
	{
		return 0;
	}
	else if (rc == PCRE_ERROR_BADOPTION)
	{
		throw RegularExpressionException("bad option");
	}
	else if (rc == 0)
	{
		throw RegularExpressionException("too many captured substrings");
	}
	else if (rc < 0)
	{
		std::ostringstream msg;
		msg << "PCRE error " << rc;
		throw RegularExpressionException(msg.str());
	}
	matches.reserve(rc);
	for (int i = 0; i < rc; ++i)
	{
		Match m;
		m.offset = ovec[i*2] < 0 ? std::string::npos : ovec[i*2] ;
		m.length = ovec[i*2 + 1] - m.offset;
		matches.push_back(m);
	}
	return rc;
}
Ejemplo n.º 4
0
      /// <summary>Parses all the keywords/macros in a string and replaces them recursively</summary>
      /// <param name="text">Source text to parse</param>
      /// <param name="depth">Debugging output depth</param>
      /// <returns>Fully parsed text</returns>
      /// <exception cref="Logic::FileFormatException">Macro contains wrong number of parameters</exception>
      /// <exception cref="Logic::Language::RegularExpressionException">RegEx error</exception>
      wstring  DescriptionParser::Parse(wstring text, int depth) const 
      {
         UINT Position;
         wsmatch match;
         wstring r;
         
         try
         {
#ifdef PRINT_CONSOLE
            Console << Cons::Cyan << Indent(depth) << "Parsing: " << Cons::White << text << ENDL;
#endif

            // Find/Replace all macros:  {AAAA:bbb}, {AAAA:bbb,ccc}, {AAAA:bbb,ccc,ddd} ...
            for (Position = 0; regex_search(text.cbegin()+Position, text.cend(), match, MatchMacro); ) //  Manually track position for in-place replacement + avoid infinite loop
            {
               // Recursively generate replacement text
               r = onMatchMacro(match, depth);

#ifdef PRINT_CONSOLE
               Console << Indent(depth) << "Replacing text: " << Cons::Yellow << match[0].str() << Cons::White << " with " << Cons::Green << r << ENDL;
#endif

               // Advance position to beyond inserted text, and insert text
               Position = (match[0].first - text.cbegin()) + r.length();
               text.replace(match[0].first, match[0].second, r);
            }

            // Find/Replace all keywords:  {AAAAA}, AAAAA
            for (Position = 0; regex_search(text.cbegin()+Position, text.cend(), match, MatchKeyword); )  
            {
               // Recursively generate replacement text
               r = onMatchKeyword(match, depth);

#ifdef PRINT_CONSOLE
               Console << Indent(depth) << "Replacing text: " << Cons::Yellow << match[0].str() << Cons::White << " with " << Cons::Green << r << ENDL;
#endif

               // Advance position to beyond inserted text, and insert text
               Position = (match[0].first - text.cbegin()) + r.length();
               text.replace(match[0].first, match[0].second, r);
            }

            return text;
         }
         catch (regex_error& e) {
            throw RegularExpressionException(HERE, e);
         }
      }
Ejemplo n.º 5
0
std::string::size_type RegularExpression::substOne(std::string& subject, std::string::size_type offset, const std::string& replacement, int options) const
{
	if (offset >= subject.length()) return std::string::npos;

	int ovec[OVEC_SIZE];
	int rc = pcre_exec(_pcre, _extra, subject.c_str(), int(subject.size()), int(offset), options & 0xFFFF, ovec, OVEC_SIZE);
	if (rc == PCRE_ERROR_NOMATCH)
	{
		return std::string::npos;
	}
	else if (rc == PCRE_ERROR_BADOPTION)
	{
		throw RegularExpressionException("bad option");
	}
	else if (rc == 0)
	{
		throw RegularExpressionException("too many captured substrings");
	}
	else if (rc < 0)
	{
		std::ostringstream msg;
		msg << "PCRE error " << rc;
		throw RegularExpressionException(msg.str());
	}
	std::string result;
	std::string::size_type len = subject.length();
	std::string::size_type pos = 0;
	std::string::size_type rp = std::string::npos;
	while (pos < len)
	{
		if (ovec[0] == pos)
		{
			std::string::const_iterator it  = replacement.begin();
			std::string::const_iterator end = replacement.end();
			while (it != end)
			{
				if (*it == '$' && !(options & RE_NO_VARS))
				{
					++it;
					if (it != end)
					{
						char d = *it;
						if (d >= '0' && d <= '9')
						{
							int c = d - '0';
							if (c < rc)
							{
								int o = ovec[c*2];
								int l = ovec[c*2 + 1] - o;
								result.append(subject, o, l);
							}
						}
						else
						{
							result += '$';
							result += d;
						}
						++it;
					}
					else result += '$';
				}
				else result += *it++;
			}
			pos = ovec[1];
			rp = result.length();
		}
		else result += subject[pos++];
	}
	subject = result;
	return rp;
}