Click here to Skip to main content
15,892,697 members
Articles / Programming Languages / C++

Fast tokenizer

Rate me:
Please Sign up or sign in to vote.
4.71/5 (11 votes)
5 Sep 20016 min read 79.7K   1.4K   44  
Fast tokenizer for C++ - like 'lexx'
#ifdef _DEBUG
#include "cooLexxerTextInputStream.h"

bool	cooLexxer::fCheckValid() const
	{

	LPCTSTR	lpszCause = NULL;

	if(!m_lcContextInfo.fCheckValid())
		{
		TRACE(_T("cooLexxer::fCheckValid() - context information corrupted.\n"));
		return false;
		}

	if(m_plisData==NULL)
		{
		TRACE(_T("cooLexxer::fCheckValid() - input stream is NULL.\n"));
		return false;
		}

	if(!m_plisData->fCheckValid())
		{
		TRACE(_T("cooLexxer::fCheckValid() - input stream is not valid.\n"));
		return false;
		}

	if(m_pllReceiver==NULL)
		{
		TRACE(_T("cooLexxer::fCheckValid() - receiver is NULL.\n"));
		return false;
		}

	if(!m_pllReceiver->fCheckValid())
		{
		TRACE(_T("cooLexxer::fCheckValid() - receiver is not valid.\n"));
		return false;
		}

	if(m_plmLexxerMap==NULL)
		{
		TRACE(_T("cooLexxer::fCheckValid() - lexxer map is NULL.\n"));
		return false;
		}

	if(!m_plmLexxerMap->fCheckValid())
		{
		TRACE(_T("cooLexxer::fCheckValid() - lexxer map is not valid.\n"));
		return false;
		}

	return true;
	}

bool	cooLexxer::fRunDiagnostics()
	{
	class	cooTestListener : public cooLexxerListener
	{
	// Attributes
	public:
		int		nTokenNumber;
		bool	fValid;

	// Operations
	public:
		virtual	bool			fCheckValid() const	{ return true; };
		virtual bool			fShouldDelete() const { return false; };
		virtual	void			vRegisterToken(const std::tstring& strToken, const cooLexxerTokenRule* pptrRule)
			{
			TRACE(_T("Token #%02d: %s\n"),nTokenNumber, strToken.data());
			switch(nTokenNumber)
				{
				case 0:
					if(strToken.compare(_T("test"))!=0 ||
						pptrRule==NULL)
						fValid=false;
					break;
				case 1:
					if(strToken.compare(_T("--"))!=0)
						fValid=false;
					break;
				case 2:
					if(strToken.compare(_T("-"))!=0)
						fValid=false;
					break;
				case 3:
					if(	strToken.compare(_T("test1"))!=0 ||
						pptrRule!=NULL)
						fValid=false;
					break;
				case 4:
					if(strToken.compare(_T("--"))!=0)
						fValid=false;
					break;
				case 5:
					if(	strToken.compare(_T("1234"))!=0 || 
						pptrRule->fIsFlagSet(ooltrf_character_rule))
						fValid=false;
					break;
				case 6:
					if(strToken.compare(_T("-"))!=0)
						fValid=false;
					break;
				case 7:
					if(strToken.compare(_T("test"))!=0)
						fValid=false;
					break;
				};
			nTokenNumber++;
			};
	};

	std::tstringstream	strm(
		_T("[seperators]\n")
		_T("-\n")
		_T("--\n")
		_T("[tokens]\n")
		_T("test\n")
		_T("[rules]\n")
		_T("strings\n")
		_T("numbers\n") );

	cooLexxerMap				map;
	cooLexxerTextInputStream	tis(_T("test---test1--1234-\"test\""));
	cooTestListener				lis;
	cooLexxer					lexxer(&tis,&map,&lis);

	lis.nTokenNumber	=0;
	lis.fValid			=true;

	// Initialize map
	map.vLoadFromStream(strm);

	while(!tis.fIsEofReached())
		lexxer.vParseCharacter();

	return lis.fValid?lexxer.fCheckValid():false;
	}
#endif

By viewing downloads associated with this article you agree to the Terms of Service and the article's licence.

If a file you wish to view isn't highlighted, and is a text file (not binary), please let us know and we'll add colourisation support for it.

License

This article has no explicit license attached to it but may contain usage terms in the article text or the download files themselves. If in doubt please contact the author via the discussion board below.

A list of licenses authors might use can be found here


Written By
Web Developer
Germany Germany
This member has not yet provided a Biography. Assume it's interesting and varied, and probably something to do with programming.

Comments and Discussions