twirls/MexFunc/WordSplit.cpp

#include <mex.h>
#include <mat.h>
#include <iostream>
#include <algorithm>
#include <vector>
#include <string>
#include <unordered_set>
#include <ctime>
using std::cout;
using std::endl;
using namespace std;

#define STRING_BUF_SIZE 204800

/* <20><>ȡabs */
void GetAbstract(const mxArray* pMxAbs, vector<string>& vAbs) {
	int rowNum = (int)mxGetM(pMxAbs);
	int colNum = (int)mxGetN(pMxAbs);
	char *strBuf = new char[STRING_BUF_SIZE];

	vAbs.resize(rowNum * colNum);
	for (int i = 0; i < rowNum; ++i) {
		for (int j = 0; j < colNum; ++j) {
			mxArray* pCell = mxGetCell(pMxAbs, j * rowNum + i);
			if (mxGetString(pCell, strBuf, STRING_BUF_SIZE) != 0) {
				cout << "String is too large to fit in the buffer! " << i + 1 << '\t' << j + 1 << endl;
				delete[]strBuf;
				return;
			}
			vAbs[i * colNum + j] = strBuf;
		}
	}
	delete[]strBuf;
}

/*
nlhs<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ŀ(Number Left - hand side)<EFBFBD><EFBFBD><EFBFBD>Ⱥ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
plhs<EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD>(Point Left - hand side)<EFBFBD><EFBFBD><EFBFBD>Ⱥ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
nrhs<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ŀ(Number Right - hand side)<EFBFBD><EFBFBD><EFBFBD>Ⱥ<EFBFBD><EFBFBD>ұ<EFBFBD>
prhs<EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD>(Point Right - hand side)<EFBFBD><EFBFBD><EFBFBD>Ⱥ<EFBFBD><EFBFBD>ұߡ<EFBFBD>Ҫע<EFBFBD><EFBFBD>prhs<EFBFBD><EFBFBD>const<EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>飬<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ܸı<EFBFBD><EFBFBD><EFBFBD>ָ<EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ݡ<EFBFBD>
*/
void mexFunction(int nlhs, mxArray* plhs[], int nrhs, const mxArray* prhs[]) {
	//cout << "WordSplit" << endl;
	//cout << nlhs << '\t' << nrhs << endl;
	if (nrhs != 1) {
		cout << "1 arguments should be given for this function!" << endl;
		return;
	}

	/* <20><>ȡ<EFBFBD><C8A1><EFBFBD><EFBFBD><EFBFBD>е<EFBFBD>ժҪ<D5AA><D2AA>Ϣ */
	vector<string> vAbstract; // <20><>ȡabs1, Ȼ<><C8BB><EFBFBD>ָ<EFBFBD><D6B8><EFBFBD>һ<EFBFBD><D2BB>һ<EFBFBD><D2BB><EFBFBD>ĵ<EFBFBD><C4B5><EFBFBD>
	GetAbstract(prhs[0], vAbstract);

	/* <20><>ժҪ<D5AA><D2AA>Ϣ<EFBFBD>ָ<EFBFBD><D6B8><EFBFBD>һ<EFBFBD><D2BB>һ<EFBFBD><D2BB><EFBFBD>Ĵʻ<C4B4> */
	clock_t begin, finish;
	begin = clock();
	unordered_set<char> usWordChars; // <20><><EFBFBD><EFBFBD><EFBFBD>ɵ<EFBFBD><C9B5>ʵ<EFBFBD><CAB5>ַ<EFBFBD><D6B7><EFBFBD>Ҫ<EFBFBD><D2AA>Ҫ<EFBFBD><D2AA><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>֣<EFBFBD>ԭ<EFBFBD><D4AD>matlab<61><62><EFBFBD><EFBFBD>ȡ<EFBFBD><C8A1><EFBFBD><EFBFBD><EFBFBD>ֵ<EFBFBD>
	for (int i = 65; i <= 90; i++) usWordChars.insert(char(i)); // A - Z
	for (int i = 97; i <= 122; i++) usWordChars.insert(char(i)); // a - z
	for (int i = 48; i <= 57; i++) usWordChars.insert(char(i)); // 0 - 9
	usWordChars.insert('/'); usWordChars.insert('+'); usWordChars.insert('-');
	vector<vector<string> > vvWordMtx(vAbstract.size()); // <20><>ʼ<EFBFBD><CABC>СΪ<D0A1><CEAA><EFBFBD>µĸ<C2B5><C4B8><EFBFBD>
	for (int i = 0; i < vAbstract.size(); i++) {
		auto& strAbs = vAbstract[i];
		// <20><><EFBFBD><EFBFBD>ժҪ<D5AA>ַ<EFBFBD><D6B7><EFBFBD><EFBFBD><EFBFBD>ÿһ<C3BF><D2BB><EFBFBD>ַ<EFBFBD><D6B7><EFBFBD>ȡ<EFBFBD><C8A1>ÿһ<C3BF><D2BB><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
		vector<string>& vWord = vvWordMtx[i];
		if (strAbs.size() == 0) continue; // ժҪ<D5AA><D2AA>ϢΪ<CFA2>գ<EFBFBD><D5A3><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>һ<EFBFBD>㲻<EFBFBD><E3B2BB><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
		int wordStartPos = 0;
		while (wordStartPos < strAbs.size() && usWordChars.find(strAbs[wordStartPos]) == usWordChars.end())
			wordStartPos++;
		for (int curPos = wordStartPos + 1; curPos < strAbs.size(); ++curPos) {
			if (usWordChars.find(strAbs[curPos]) == usWordChars.end()) { // <20>ҵ<EFBFBD><D2B5>˷ָ<CBB7><D6B8><EFBFBD>
				vWord.push_back(strAbs.substr(wordStartPos, curPos - wordStartPos));
				wordStartPos = curPos + 1; // <20><><EFBFBD><EFBFBD>һ<EFBFBD><D2BB><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ʼλ<CABC><CEBB>
				while (wordStartPos < strAbs.size() && usWordChars.find(strAbs[wordStartPos]) == usWordChars.end())
					wordStartPos++;
				curPos = wordStartPos; // ѭ<><D1AD><EFBFBD><EFBFBD><EFBFBD>Զ<EFBFBD><D4B6><EFBFBD>1
			}
		}
	}

	/* <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>д<EFBFBD>뷵<EFBFBD>ز<EFBFBD><D8B2><EFBFBD> */
	/* <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>д<EFBFBD>뷵<EFBFBD><EBB7B5>ֵ */
	if (nlhs > 0) {
		mxArray* pCellMtx = mxCreateCellMatrix(1, vvWordMtx.size());
		for (int i = 0; i < vvWordMtx.size(); ++i) {
			mxArray* pChildCellMtx = mxCreateCellMatrix(1, vvWordMtx[i].size());
			for (int j = 0; j < vvWordMtx[i].size(); ++j) {
				mxArray* mxStr = mxCreateString(vvWordMtx[i][j].c_str());
				mxSetCell(pChildCellMtx, j, mxStr);
			}
			mxSetCell(pCellMtx, i, pChildCellMtx);
		}
		plhs[0] = pCellMtx; // <20><>ֵ<EFBFBD><D6B5><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ֵ
	}


	finish = clock();
	//cout << "split abstract time: " << (double)(finish - begin) / CLOCKS_PER_SEC << " s" << endl;

}