📄 mset__tokenizer_8h-source.html
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"><html><head><meta http-equiv="Content-Type" content="text/html;charset=iso-8859-1"><title>DMTL: mset_tokenizer.h Source File</title><link href="doxygen.css" rel="stylesheet" type="text/css"><link href="tabs.css" rel="stylesheet" type="text/css"></head><body><!-- Generated by Doxygen 1.4.7 --><div class="tabs"> <ul> <li><a href="index.html"><span>Main Page</span></a></li> <li><a href="annotated.html"><span>Classes</span></a></li> <li id="current"><a href="files.html"><span>Files</span></a></li> <li><a href="pages.html"><span>Related Pages</span></a></li> </ul></div><h1>mset_tokenizer.h</h1><div class="fragment"><pre class="fragment"><a name="l00001"></a>00001 <span class="comment">/*</span><a name="l00002"></a>00002 <span class="comment"> * Copyright (C) 2005 M.J. Zaki <zaki@cs.rpi.edu> Rensselaer Polytechnic Institute</span><a name="l00003"></a>00003 <span class="comment"> * Written by parimi@cs.rpi.edu</span><a name="l00004"></a>00004 <span class="comment"> * Updated by chaojv@cs.rpi.edu, alhasan@cs.rpi.edu, salems@cs.rpi.edu</span><a name="l00005"></a>00005 <span class="comment"> * Modifications:</span><a name="l00006"></a>00006 <span class="comment"> * Added tokenizer properties -- Zaki, 5/8/06</span><a name="l00007"></a>00007 <span class="comment"> *</span><a name="l00008"></a>00008 <span class="comment"> * This program is free software; you can redistribute it and/or</span><a name="l00009"></a>00009 <span class="comment"> * modify it under the terms of the GNU General Public License</span><a name="l00010"></a>00010 <span class="comment"> * as published by the Free Software Foundation; either version 2</span><a name="l00011"></a>00011 <span class="comment"> * of the License, or (at your option) any later version.</span><a name="l00012"></a>00012 <span class="comment"> *</span><a name="l00013"></a>00013 <span class="comment"> * This program is distributed in the hope that it will be useful,</span><a name="l00014"></a>00014 <span class="comment"> * but WITHOUT ANY WARRANTY; without even the implied warranty of</span><a name="l00015"></a>00015 <span class="comment"> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the</span><a name="l00016"></a>00016 <span class="comment"> * GNU General Public License for more details.</span><a name="l00017"></a>00017 <span class="comment"> *</span><a name="l00018"></a>00018 <span class="comment"> * You should have received a copy of the GNU General Public License along</span><a name="l00019"></a>00019 <span class="comment"> * with this program; if not, write to the Free Software Foundation, Inc.,</span><a name="l00020"></a>00020 <span class="comment"> * 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA.</span><a name="l00021"></a>00021 <span class="comment"> */</span><a name="l00022"></a>00022 <span class="preprocessor">#ifndef _MSET_TOKENIZER_H_</span><a name="l00023"></a>00023 <span class="preprocessor"></span><span class="preprocessor">#define _MSET_TOKENIZER_H_</span><a name="l00024"></a>00024 <span class="preprocessor"></span><a name="l00025"></a>00025 <span class="preprocessor">#include <fstream></span><a name="l00026"></a>00026 <a name="l00027"></a>00027 <span class="preprocessor">#include "mset_can_code.h"</span><a name="l00028"></a>00028 <span class="preprocessor">#include "adj_list.h"</span><a name="l00029"></a>00029 <span class="preprocessor">#include "generic_classes.h"</span><a name="l00030"></a>00030 <span class="preprocessor">#include "tokenizer_utils.h"</span><a name="l00031"></a>00031 <span class="preprocessor">#include "typedefs.h"</span><a name="l00032"></a>00032 <a name="l00033"></a>00033 <span class="preprocessor">#include "<a class="code" href="element__parser_8h.html">element_parser.h</a>"</span><a name="l00034"></a>00034 <span class="preprocessor">#include "pattern.h"</span><a name="l00035"></a>00035 <span class="preprocessor">#include "<a class="code" href="pat__fam_8h.html">pat_fam.h</a>"</span><a name="l00036"></a>00036 <a name="l00037"></a>00037 <span class="comment">/* NOTE: the parsing scheme reads atmost the first MAXLINE chars of a line</span><a name="l00038"></a>00038 <span class="comment"> this can perhaps be improved towards a better one */</span><a name="l00039"></a>00039 <a name="l00047"></a>00047 <span class="keyword">template</span><<span class="keyword">class</span> PP, <span class="keyword">typename</span> MP, <span class="keyword">typename</span> TP, <span class="keyword">typename</span> PAT_ST, <span class="keyword">template</span><<span class="keyword">typename</span>, <span class="keyword">typename</span>, <span class="keyword">typename</span>, <span class="keyword">template</span> <<span class="keyword">typename</span>> <span class="keyword">class</span>> <span class="keyword">class </span>CC, <a name="l00048"></a>00048 template <typename> class ALLOC ><a name="l00049"></a><a class="code" href="classtokenizer_3_01MSET__PATTERN_00_01DMTL__TKNZ__PROP_00_01ALLOC_01_4.html">00049</a> class <a class="code" href="classtokenizer.html">tokenizer</a><MSET_PATTERN, DMTL_TKNZ_PROP, ALLOC ><a name="l00050"></a>00050 {<a name="l00051"></a>00051 <a name="l00052"></a>00052 <span class="keyword">public</span>: <a name="l00053"></a>00053 <span class="keyword">typedef</span> <a class="code" href="classpattern__support.html">pattern_support<V_Fkk_MINE_PROP></a> <a class="code" href="classpattern__support.html">PAT_SUP</a>;<a name="l00054"></a>00054 <span class="keyword">typedef</span> <a class="code" href="classvat.html">vat<MSET_PROP, V_Fkk_MINE_PROP, ALLOC, std::vector ></a> <a class="code" href="classvat.html">VAT</a>;<a name="l00055"></a>00055 <span class="keyword">typedef</span> <span class="keyword">typename</span> MSET_PATTERN::VERTEX_T V_T;<a name="l00056"></a>00056 <span class="keyword">typedef</span> <span class="keyword">typename</span> MSET_PATTERN::EDGE_T E_T;<a name="l00057"></a>00057 <a name="l00058"></a><a class="code" href="classtokenizer_3_01MSET__PATTERN_00_01DMTL__TKNZ__PROP_00_01ALLOC_01_4.html#6b23e5bd261d78a66610aa4b642b175b">00058</a> <span class="keyword">typedef</span> <span class="keywordtype">int</span> VAT_T; <a name="l00060"></a><a class="code" href="classtokenizer_3_01MSET__PATTERN_00_01DMTL__TKNZ__PROP_00_01ALLOC_01_4.html#34a28f77388a57a5f724a24275ba0de6">00060</a> <a class="code" href="classtokenizer.html">tokenizer</a>(<span class="keywordtype">int</span> max=LINE_SZ): MAXLINE(max) {} <a name="l00069"></a>00069 <span class="keyword">template</span><<span class="keyword">class</span> SM_T><a name="l00070"></a><a class="code" href="classtokenizer_3_01MSET__PATTERN_00_01DMTL__TKNZ__PROP_00_01ALLOC_01_4.html#c8904090f794055e42794835207bf180">00070</a> <span class="keywordtype">int</span> parse_next_trans(std::ifstream& infile, <a class="code" href="classpat__fam.html">pat_fam<MSET_PATTERN></a>& freq_pats, <a class="code" href="classstorage__manager.html">storage_manager<MSET_PATTERN, VAT, ALLOC, SM_T></a>& vat_hmap) {<a name="l00071"></a>00071 <a name="l00072"></a>00072 <span class="keywordtype">char</span>* line=<span class="keyword">new</span> <span class="keywordtype">char</span>[MAXLINE];<a name="l00073"></a>00073 <span class="keywordtype">char</span> word[MAXLINE];<a name="l00074"></a>00074 <span class="keywordtype">char</span>* startline=line;<a name="l00075"></a>00075 <a name="l00076"></a>00076 <span class="keywordtype">int</span> len;<a name="l00077"></a>00077 <span class="keywordtype">int</span> count; <span class="comment">//# of words parsed from line</span><a name="l00078"></a>00078 <span class="keywordtype">int</span> tid=-1, ts;<a name="l00079"></a>00079 <span class="keywordtype">int</span> num_items=3; <span class="comment">//# of itemsets on this transaction</span><a name="l00080"></a>00080 <a name="l00081"></a>00081 <a class="code" href="classvat.html">VAT</a>* ivat;<a name="l00082"></a>00082 <a name="l00083"></a>00083 *line=<span class="charliteral">'\0'</span>;<a name="l00084"></a>00084 infile.getline(line, MAXLINE-1);<a name="l00085"></a>00085 len=strlen(line);<a name="l00086"></a>00086 <span class="keywordflow">if</span>(!len || !line) {<a name="l00087"></a>00087 <span class="keyword">delete</span>[] startline;<a name="l00088"></a>00088 <span class="keywordflow">return</span> -1;<a name="l00089"></a>00089 }<a name="l00090"></a>00090 <a name="l00091"></a>00091 line[len++]=<span class="charliteral">'\0'</span>;<a name="l00092"></a>00092 count=0;<a name="l00093"></a>00093 <span class="keywordflow">while</span>(count<num_items+3 && line<(startline+len)) {<a name="l00094"></a>00094 <a name="l00095"></a>00095 <span class="keywordflow">if</span>(!(line=parse_word()(line, word))) {<a name="l00096"></a>00096 <span class="comment">//parse_word() failed</span><a name="l00097"></a>00097 <span class="keyword">delete</span>[] startline;<a name="l00098"></a>00098 <span class="keywordflow">return</span> -1;<a name="l00099"></a>00099 }<a name="l00100"></a>00100 count++;<a name="l00101"></a>00101 <a name="l00102"></a>00102 <span class="keywordflow">switch</span>(count) {<a name="l00103"></a>00103 <span class="keywordflow">case</span> 1:<a name="l00104"></a>00104 <span class="comment">//this is tid</span><a name="l00105"></a>00105 tid=atoi(word); <span class="keywordflow">break</span>;<a name="l00106"></a>00106 <a name="l00107"></a>00107 <span class="keywordflow">case</span> 2: <a name="l00108"></a>00108 <span class="comment">//this is timestamp</span><a name="l00109"></a>00109 ts=atoi(word); <span class="keywordflow">break</span>;<a name="l00110"></a>00110 <a name="l00111"></a>00111 <span class="keywordflow">case</span> 3:<a name="l00112"></a>00112 <span class="comment">//this is # of elements on line</span><a name="l00113"></a>00113 num_items=atoi(word);<a name="l00114"></a>00114 <span class="keywordflow">break</span>;<a name="l00115"></a>00115 <a name="l00116"></a>00116 <span class="keywordflow">default</span>:<a name="l00117"></a>00117 <span class="comment">//this is an element, insert/append to its VAT</span><a name="l00118"></a>00118 MSET_PATTERN* p = <span class="keyword">new</span> MSET_PATTERN();<a name="l00119"></a>00119 <span class="comment">// Add vertex to the empty graph.</span><a name="l00120"></a>00120 V_T v = el_prsr.parse_element(word);<a name="l00121"></a>00121 p->add_vertex(v);<a name="l00122"></a>00122 p->init_canonical_code(v);<a name="l00123"></a>00123 <a name="l00124"></a>00124 <span class="comment">//if p contains a vat in vat_hmap, append tid to the entry</span><a name="l00125"></a>00125 <span class="comment">//else create a new vat and insert it into vat_hmap,</span><a name="l00126"></a>00126 <span class="comment">//and add p to freq_pats</span><a name="l00127"></a>00127 ivat=vat_hmap.get_vat(p);<a name="l00128"></a>00128 <span class="keywordflow">if</span>(vat_hmap.find(p)) {<a name="l00129"></a>00129 <span class="comment">//vat found</span><a name="l00130"></a>00130 <a name="l00131"></a>00131 <span class="keywordflow">if</span>(tid == (*ivat)[ivat->size()-1].first) {<a name="l00132"></a>00132 (*ivat)[ivat->size()-1].second++;<a name="l00133"></a>00133 } <span class="keywordflow">else</span> {<a name="l00134"></a>00134 ivat->push_back(make_pair(tid, 1));<a name="l00135"></a>00135 }<a name="l00136"></a>00136 <span class="keyword">delete</span> p;<a name="l00137"></a>00137 }<a name="l00138"></a>00138 <span class="keywordflow">else</span> {<a name="l00139"></a>00139 <span class="comment">//create a new vat & insert it</span><a name="l00140"></a>00140 ivat=<span class="keyword">new</span> <a class="code" href="classvat.html">VAT</a>();<a name="l00141"></a>00141 ivat->push_back(make_pair(tid, 1));<a name="l00142"></a>00142 <a name="l00143"></a>00143 <span class="keywordflow">if</span>(!vat_hmap.add_vat(p, ivat)) {<a name="l00144"></a>00144 cerr<<<span class="stringliteral">"tokenizer.get_length_one: add_vat failed"</span><<endl;<a name="l00145"></a>00145 <span class="keywordflow">return</span> -1;<a name="l00146"></a>00146 }<a name="l00147"></a>00147 <a name="l00148"></a>00148 freq_pats.push_back(p);<a name="l00149"></a>00149 } <span class="comment">//end else</span><a name="l00150"></a>00150 <a name="l00151"></a>00151 } <span class="comment">//end switch</span><a name="l00152"></a>00152 <a name="l00153"></a>00153 } <span class="comment">//end while(count<..)</span><a name="l00154"></a>00154 <a name="l00155"></a>00155 <span class="keyword">delete</span>[] startline;<a name="l00156"></a>00156 <a name="l00157"></a>00157 <span class="keywordflow">return</span> tid;<a name="l00158"></a>00158 } <span class="comment">//end parse_next_trans()</span><a name="l00159"></a>00159 <a name="l00160"></a>00160 <a name="l00161"></a>00161 <a name="l00162"></a>00162 <span class="keyword">private</span>:<a name="l00163"></a>00163 <span class="keywordtype">int</span> MAXLINE; <a name="l00164"></a>00164 <a class="code" href="classelement__parser.html">element_parser<V_T></a> el_prsr; <a name="l00166"></a>00166 }; <span class="comment">//end class tokenizer<itemset></span><a name="l00167"></a>00167 <a name="l00168"></a>00168 <span class="preprocessor">#endif</span></pre></div><hr size="1"><address style="align: right;"><small>Generated on Wed Jul 26 14:01:08 2006 for DMTL by <a href="http://www.doxygen.org/index.html"><img src="doxygen.png" alt="doxygen" align="middle" border="0"></a> 1.4.7 </small></address></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -