diff options
Diffstat (limited to 'rep/ept/textsearch/textsearch.h.gcov.html')
| -rw-r--r-- | rep/ept/textsearch/textsearch.h.gcov.html | 286 |
1 files changed, 0 insertions, 286 deletions
diff --git a/rep/ept/textsearch/textsearch.h.gcov.html b/rep/ept/textsearch/textsearch.h.gcov.html deleted file mode 100644 index d46dca8..0000000 --- a/rep/ept/textsearch/textsearch.h.gcov.html +++ /dev/null @@ -1,286 +0,0 @@ -<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> - -<html lang="en"> - -<head> - <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"> - <title>LCOV - lcov.info - ept/textsearch/textsearch.h</title> - <link rel="stylesheet" type="text/css" href="../../gcov.css"> -</head> - -<body> - - <table width="100%" border=0 cellspacing=0 cellpadding=0> - <tr><td class="title">LTP GCOV extension - code coverage report</td></tr> - <tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr> - - <tr> - <td width="100%"> - <table cellpadding=1 border=0 width="100%"> - <tr> - <td class="headerItem" width="20%">Current view:</td> - <td class="headerValue" width="80%" colspan=4><a href="../../index.html">directory</a> - <a href="index.html">ept/textsearch</a> - textsearch.h</td> - </tr> - <tr> - <td class="headerItem" width="20%">Test:</td> - <td class="headerValue" width="80%" colspan=4>lcov.info</td> - </tr> - <tr> - <td class="headerItem" width="20%">Date:</td> - <td class="headerValue" width="20%">2008-08-14</td> - <td width="20%"></td> - <td class="headerItem" width="20%">Instrumented lines:</td> - <td class="headerValue" width="20%">15</td> - </tr> - <tr> - <td class="headerItem" width="20%">Code covered:</td> - <td class="headerValue" width="20%">86.7 %</td> - <td width="20%"></td> - <td class="headerItem" width="20%">Executed lines:</td> - <td class="headerValue" width="20%">13</td> - </tr> - </table> - </td> - </tr> - <tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr> - </table> - - <table cellpadding=0 cellspacing=0 border=0> - <tr> - <td><br></td> - </tr> - <tr> - <td><pre class="source"> -<span class="lineNum"> 1 </span> : #ifndef EPT_TEXTSEARCH_TEXTSEARCH_H -<span class="lineNum"> 2 </span> : #define EPT_TEXTSEARCH_TEXTSEARCH_H -<span class="lineNum"> 3 </span> : -<span class="lineNum"> 4 </span> : /** @file -<span class="lineNum"> 5 </span> : * @author Enrico Zini <enrico@enricozini.org> -<span class="lineNum"> 6 </span> : * Fast full-text search -<span class="lineNum"> 7 </span> : */ -<span class="lineNum"> 8 </span> : -<span class="lineNum"> 9 </span> : /* -<span class="lineNum"> 10 </span> : * Copyright (C) 2007 Enrico Zini <enrico@debian.org> -<span class="lineNum"> 11 </span> : * -<span class="lineNum"> 12 </span> : * This program is free software; you can redistribute it and/or modify -<span class="lineNum"> 13 </span> : * it under the terms of the GNU General Public License as published by -<span class="lineNum"> 14 </span> : * the Free Software Foundation; either version 2 of the License, or -<span class="lineNum"> 15 </span> : * (at your option) any later version. -<span class="lineNum"> 16 </span> : * -<span class="lineNum"> 17 </span> : * This program is distributed in the hope that it will be useful, -<span class="lineNum"> 18 </span> : * but WITHOUT ANY WARRANTY; without even the implied warranty of -<span class="lineNum"> 19 </span> : * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -<span class="lineNum"> 20 </span> : * GNU General Public License for more details. -<span class="lineNum"> 21 </span> : * -<span class="lineNum"> 22 </span> : * You should have received a copy of the GNU General Public License -<span class="lineNum"> 23 </span> : * along with this program; if not, write to the Free Software -<span class="lineNum"> 24 </span> : * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA -<span class="lineNum"> 25 </span> : */ -<span class="lineNum"> 26 </span> : -<span class="lineNum"> 27 </span> : #include <xapian.h> -<span class="lineNum"> 28 </span> : #include <vector> -<span class="lineNum"> 29 </span> : #include <string> -<span class="lineNum"> 30 </span> : -<span class="lineNum"> 31 </span> : namespace ept { -<span class="lineNum"> 32 </span> : namespace apt { -<span class="lineNum"> 33 </span> : class Apt; -<span class="lineNum"> 34 </span> : class PackageRecord; -<span class="lineNum"> 35 </span> : } -<span class="lineNum"> 36 </span> : namespace debtags { -<span class="lineNum"> 37 </span> : class Debtags; -<span class="lineNum"> 38 </span> : } -<span class="lineNum"> 39 </span> : namespace textsearch { -<span class="lineNum"> 40 </span> : -<span class="lineNum"> 41 </span> : // Allocate value indexes for known values -<span class="lineNum"> 42 </span> : const Xapian::valueno VAL_APT_INSTALLED_SIZE = 1; -<span class="lineNum"> 43 </span> : const Xapian::valueno VAL_APT_PACKAGE_SIZE = 2; -<span class="lineNum"> 44 </span> : const Xapian::valueno VAL_POPCON = 10; -<span class="lineNum"> 45 </span> : const Xapian::valueno VAL_ITERATING_RATING = 20; -<span class="lineNum"> 46 </span> : const Xapian::valueno VAL_ITERATING_FUNCTIONALITY = 21; -<span class="lineNum"> 47 </span> : const Xapian::valueno VAL_ITERATING_USABILITY = 22; -<span class="lineNum"> 48 </span> : const Xapian::valueno VAL_ITERATING_SECURITY = 23; -<span class="lineNum"> 49 </span> : const Xapian::valueno VAL_ITERATING_PERFORMANCE = 24; -<span class="lineNum"> 50 </span> : const Xapian::valueno VAL_ITERATING_QUALITY = 25; -<span class="lineNum"> 51 </span> : const Xapian::valueno VAL_ITERATING_SUPPORT = 26; -<span class="lineNum"> 52 </span> : const Xapian::valueno VAL_ITERATING_ADOPTION = 27; -<span class="lineNum"> 53 </span> : // If you need to index a value and cannot edit this file, feel free to use any -<span class="lineNum"> 54 </span> : // value starting from 1000000 -<span class="lineNum"> 55 </span> : -<span class="lineNum"> 56 </span> : -<span class="lineNum"> 57 </span> : /* -<span class="lineNum"> 58 </span> : Fallback on apt scan searches when index is not present -<span class="lineNum"> 59 </span> : -<span class="lineNum"> 60 </span> : Explicitly decide at instantiation (or at any other time) if a rebuild should -<span class="lineNum"> 61 </span> : be performed. Just adding a 'rebuildIfNeeded' method would be enough. -<span class="lineNum"> 62 </span> : -<span class="lineNum"> 63 </span> : 17:14 #xapian < enrico> Hello. I'm finally in a position of writing a library to maintain -<span class="lineNum"> 64 </span> : a xapian index with Debian package descriptions in a Debian system -<span class="lineNum"> 65 </span> : 17:14 #xapian < enrico> I have a question, though -<span class="lineNum"> 66 </span> : 17:14 #xapian < enrico> The descriptions change regularly as people run 'apt-get update' -<span class="lineNum"> 67 </span> : 17:15 #xapian < enrico> I'd need to have a way to update the description index after -<span class="lineNum"> 68 </span> : apt-get update, without rebuilding it from scratch -<span class="lineNum"> 69 </span> : 17:15 #xapian < enrico> Is there some documentation on how to do that? I can't exactly -<span class="lineNum"> 70 </span> : tell Xapian "the new description for package foo is this" because -<span class="lineNum"> 71 </span> : I'd need the xapian id -<span class="lineNum"> 72 </span> : 19:11 #xapian < omega> you can add a unique term with a boolean prefix? -<span class="lineNum"> 73 </span> : 19:11 #xapian < omega> like Qpackage-name -<span class="lineNum"> 74 </span> : 19:11 #xapian < omega> then you search for it and replace_document -<span class="lineNum"> 75 </span> : 19:24 #xapian < richardb> Or indeed, you use the "replace_document()" form which takes a -<span class="lineNum"> 76 </span> : unique_id term. -<span class="lineNum"> 77 </span> : 19:25 #xapian < richardb> Xapian::docid replace_document(const std::string & -<span class="lineNum"> 78 </span> : unique_term, -<span class="lineNum"> 79 </span> : 19:25 #xapian < richardb> const Xapian::Document & -<span class="lineNum"> 80 </span> : document); -<span class="lineNum"> 81 </span> : 19:43 #xapian < enrico> unique term -<span class="lineNum"> 82 </span> : 19:43 #xapian < enrico> nice! -<span class="lineNum"> 83 </span> : 19:44 #xapian < enrico> can I use a non-alpha prefix, like :package-name ? -<span class="lineNum"> 84 </span> : 19:45 #xapian < enrico> or pkg:package-name -<span class="lineNum"> 85 </span> : 19:45 #xapian < enrico> I suppose I can -<span class="lineNum"> 86 </span> : */ -<span class="lineNum"> 87 </span> : -<span class="lineNum"> 88 </span> : /** -<span class="lineNum"> 89 </span> : * Maintains and accesses a Xapian index of package descriptions. -<span class="lineNum"> 90 </span> : * -<span class="lineNum"> 91 </span> : * Contrarily to Debtags and Popcon, TextSearch does not attempt to create the -<span class="lineNum"> 92 </span> : * index in the home directory if no system index is found and it is not -<span class="lineNum"> 93 </span> : * running as root: this is to avoid secretly building large indexes (>50Mb) -<span class="lineNum"> 94 </span> : * in the home directory of users. -<span class="lineNum"> 95 </span> : * -<span class="lineNum"> 96 </span> : * The idea then is to have root keep the index up to date, possibly running a -<span class="lineNum"> 97 </span> : * reindexing tool once a day, or after an apt-get update. -<span class="lineNum"> 98 </span> : * -<span class="lineNum"> 99 </span> : * This works because the full text search index is useful even if it is -<span class="lineNum"> 100 </span> : * slightly out of date. -<span class="lineNum"> 101 </span> : */ -<span class="lineNum"> 102 </span> : class TextSearch -<span class="lineNum"> 103 </span><span class="lineCov"> 11 : {</span> -<span class="lineNum"> 104 </span> : protected: -<span class="lineNum"> 105 </span> : time_t m_timestamp; -<span class="lineNum"> 106 </span> : Xapian::Database m_db; -<span class="lineNum"> 107 </span> : Xapian::Stem m_stem; -<span class="lineNum"> 108 </span> : -<span class="lineNum"> 109 </span> : /// Return a lowercased copy of the string -<span class="lineNum"> 110 </span> : static std::string toLower(const std::string& str); -<span class="lineNum"> 111 </span> : -<span class="lineNum"> 112 </span> : /** -<span class="lineNum"> 113 </span> : * Add normalised tokens computed from the string to the document doc. -<span class="lineNum"> 114 </span> : * -<span class="lineNum"> 115 </span> : * pos is used as a sequence generator for entering the token position in -<span class="lineNum"> 116 </span> : * the document. -<span class="lineNum"> 117 </span> : */ -<span class="lineNum"> 118 </span> : void normalize_and_add(Xapian::Document& doc, const std::string& term, int& pos) const; -<span class="lineNum"> 119 </span> : -<span class="lineNum"> 120 </span> : public: -<span class="lineNum"> 121 </span> : struct ExtraIndexer -<span class="lineNum"> 122 </span> : { -<span class="lineNum"> 123 </span><span class="lineNoCov"> 0 : virtual ~ExtraIndexer() {}</span> -<span class="lineNum"> 124 </span> : virtual void operator()(Xapian::Document& doc, const apt::PackageRecord& rec) const = 0; -<span class="lineNum"> 125 </span> : }; -<span class="lineNum"> 126 </span> : -<span class="lineNum"> 127 </span> : TextSearch(); -<span class="lineNum"> 128 </span> : -<span class="lineNum"> 129 </span> : /// Access the Xapian database -<span class="lineNum"> 130 </span><span class="lineCov"> 3 : Xapian::Database& db() { return m_db; }</span> -<span class="lineNum"> 131 </span> : -<span class="lineNum"> 132 </span> : /// Access the Xapian database -<span class="lineNum"> 133 </span><span class="lineCov"> 4 : const Xapian::Database& db() const { return m_db; }</span> -<span class="lineNum"> 134 </span> : -<span class="lineNum"> 135 </span> : /// Timestamp of when the Xapian database was last updated -<span class="lineNum"> 136 </span><span class="lineCov"> 3 : time_t timestamp() const { return m_timestamp; }</span> -<span class="lineNum"> 137 </span> : -<span class="lineNum"> 138 </span> : /// Returns true if the index has data -<span class="lineNum"> 139 </span><span class="lineCov"> 3 : bool hasData() const { return m_timestamp > 0; }</span> -<span class="lineNum"> 140 </span> : -<span class="lineNum"> 141 </span> : /// Returns true if the index is older than the Apt database information -<span class="lineNum"> 142 </span> : bool needsRebuild(apt::Apt& apt); -<span class="lineNum"> 143 </span> : -<span class="lineNum"> 144 </span> : /** -<span class="lineNum"> 145 </span> : * Rebuild the index if needed. -<span class="lineNum"> 146 </span> : * -<span class="lineNum"> 147 </span> : * Allow to specify functors that contribute to the indexing. -<span class="lineNum"> 148 </span> : * -<span class="lineNum"> 149 </span> : * @note This requires write access to the index directory. -<span class="lineNum"> 150 </span> : * @note This is not the main way to update the index: it is provided here -<span class="lineNum"> 151 </span> : * only as a way to build a draft index for the library tests -<span class="lineNum"> 152 </span> : */ -<span class="lineNum"> 153 </span> : bool rebuildIfNeeded( -<span class="lineNum"> 154 </span> : apt::Apt& apt, -<span class="lineNum"> 155 </span> : const std::vector<const ExtraIndexer*>& extraIndexers = std::vector<const ExtraIndexer*>()); -<span class="lineNum"> 156 </span> : -<span class="lineNum"> 157 </span> : /** -<span class="lineNum"> 158 </span> : * Retrieve a Xapian docid by package name -<span class="lineNum"> 159 </span> : */ -<span class="lineNum"> 160 </span> : Xapian::docid docidByName(const std::string& pkgname) const; -<span class="lineNum"> 161 </span> : -<span class="lineNum"> 162 </span> : /** -<span class="lineNum"> 163 </span> : * Tokenize the string and build an OR query with the resulting keywords -<span class="lineNum"> 164 </span> : */ -<span class="lineNum"> 165 </span> : Xapian::Query makeORQuery(const std::string& keywords) const; -<span class="lineNum"> 166 </span> : -<span class="lineNum"> 167 </span> : /** -<span class="lineNum"> 168 </span> : * Tokenize the string and build an OR query with the resulting keywords. -<span class="lineNum"> 169 </span> : * -<span class="lineNum"> 170 </span> : * The last token in keywords is considered to be typed only partially, to -<span class="lineNum"> 171 </span> : * implement proper search-as-you-type. -<span class="lineNum"> 172 </span> : */ -<span class="lineNum"> 173 </span> : Xapian::Query makePartialORQuery(const std::string& keywords) const; -<span class="lineNum"> 174 </span> : -<span class="lineNum"> 175 </span> : /** -<span class="lineNum"> 176 </span> : * Build a query with the given keywords, specified as iterators of strings -<span class="lineNum"> 177 </span> : */ -<span class="lineNum"> 178 </span> : template<typename ITER> -<span class="lineNum"> 179 </span><span class="lineCov"> 3 : Xapian::Query makeORQuery(const ITER& begin, const ITER& end) const</span> -<span class="lineNum"> 180 </span> : { -<span class="lineNum"> 181 </span><span class="lineCov"> 3 : std::vector<std::string> terms;</span> -<span class="lineNum"> 182 </span> : // Insert both the lowercased and the stemmed lowercased query terms -<span class="lineNum"> 183 </span><span class="lineCov"> 10 : for (ITER i = begin; i != end; ++i)</span> -<span class="lineNum"> 184 </span> : { -<span class="lineNum"> 185 </span><span class="lineCov"> 7 : std::string t = toLower(*i);</span> -<span class="lineNum"> 186 </span><span class="lineCov"> 7 : std::string s = m_stem(t);</span> -<span class="lineNum"> 187 </span><span class="lineCov"> 7 : terms.push_back(t);</span> -<span class="lineNum"> 188 </span><span class="lineCov"> 7 : if (s != t)</span> -<span class="lineNum"> 189 </span><span class="lineNoCov"> 0 : terms.push_back("Z" + s);</span> -<span class="lineNum"> 190 </span> : } -<span class="lineNum"> 191 </span><span class="lineCov"> 3 : return Xapian::Query(Xapian::Query::OP_OR, terms.begin(), terms.end());</span> -<span class="lineNum"> 192 </span> : } -<span class="lineNum"> 193 </span> : -<span class="lineNum"> 194 </span> : /// Return a list of tag-based terms that can be used to expand an OR query -<span class="lineNum"> 195 </span> : std::vector<std::string> expand(Xapian::Enquire& enq) const; -<span class="lineNum"> 196 </span> : -<span class="lineNum"> 197 </span> : // std::vector<std::string> similar(const std::string& pkg); -<span class="lineNum"> 198 </span> : -<span class="lineNum"> 199 </span> : /** -<span class="lineNum"> 200 </span> : * Create a query to look for packages similar to the given one -<span class="lineNum"> 201 </span> : */ -<span class="lineNum"> 202 </span> : Xapian::Query makeRelatedQuery(const std::string& pkgname) const; -<span class="lineNum"> 203 </span> : -<span class="lineNum"> 204 </span> : /** -<span class="lineNum"> 205 </span> : * Get the integer value for -<span class="lineNum"> 206 </span> : */ -<span class="lineNum"> 207 </span> : double getDoubleValue(const std::string& pkgname, Xapian::valueno val_id) const; -<span class="lineNum"> 208 </span> : -<span class="lineNum"> 209 </span> : /** -<span class="lineNum"> 210 </span> : * Get the integer value for -<span class="lineNum"> 211 </span> : */ -<span class="lineNum"> 212 </span> : int getIntValue(const std::string& pkgname, Xapian::valueno val_id) const; -<span class="lineNum"> 213 </span> : }; -<span class="lineNum"> 214 </span> : -<span class="lineNum"> 215 </span> : } -<span class="lineNum"> 216 </span> : } -<span class="lineNum"> 217 </span> : -<span class="lineNum"> 218 </span> : // vim:set ts=4 sw=4: -<span class="lineNum"> 219 </span> : #endif -</pre> - </td> - </tr> - </table> - <br> - - <table width="100%" border=0 cellspacing=0 cellpadding=0> - <tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr> - <tr><td class="versionInfo">Generated by: <a href="http://ltp.sourceforge.net/coverage/lcov.php" target="_parent">LTP GCOV extension version 1.6</a></td></tr> - </table> - <br> - -</body> -</html> |
