src/utility/html.cpp

   1 /***************************************************************************
   2  *   Copyright (C) 2008-2016 by Andrzej Rybczak                            *
   3  *   electricityispower@gmail.com                                          *
   4  *                                                                         *
   5  *   This program is free software; you can redistribute it and/or modify  *
   6  *   it under the terms of the GNU General Public License as published by  *
   7  *   the Free Software Foundation; either version 2 of the License, or     *
   8  *   (at your option) any later version.                                   *
   9  *                                                                         *
  10  *   This program is distributed in the hope that it will be useful,       *
  11  *   but WITHOUT ANY WARRANTY; without even the implied warranty of        *
  12  *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the         *
  13  *   GNU General Public License for more details.                          *
  14  *                                                                         *
  15  *   You should have received a copy of the GNU General Public License     *
  16  *   along with this program; if not, write to the                         *
  17  *   Free Software Foundation, Inc.,                                       *
  18  *   51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.              *
  19  ***************************************************************************/
  20
  21 #include <algorithm>
  22 #include <boost/algorithm/string/replace.hpp>
  23 #include "utility/html.h"
  24
  25 std::string unescapeHtmlUtf8(const std::string &data)
  26 {
  27         std::string result;
  28         for (size_t i = 0, j; i < data.length(); ++i)
  29         {
  30                 if (data[i] == '&' && data[i+1] == '#' && (j = data.find(';', i)) != std::string::npos)
  31                 {
  32                         int n = atoi(&data.c_str()[i+2]);
  33                         if (n >= 0x800)
  34                         {
  35                                 result += (0xe0 | ((n >> 12) & 0x0f));
  36                                 result += (0x80 | ((n >> 6) & 0x3f));
  37                                 result += (0x80 | (n & 0x3f));
  38                         }
  39                         else if (n >= 0x80)
  40                         {
  41                                 result += (0xc0 | ((n >> 6) & 0x1f));
  42                                 result += (0x80 | (n & 0x3f));
  43                         }
  44                         else
  45                                 result += n;
  46                         i = j;
  47                 }
  48                 else
  49                         result += data[i];
  50         }
  51         return result;
  52 }
  53
  54 void unescapeHtmlEntities(std::string &s)
  55 {
  56         // well, at least some of them.
  57         boost::replace_all(s, "&amp;", "&");
  58         boost::replace_all(s, "&gt;", ">");
  59         boost::replace_all(s, "&lt;", "<");
  60         boost::replace_all(s, "&nbsp;", " ");
  61         boost::replace_all(s, "&quot;", "\"");
  62         boost::replace_all(s, "&ndash;", "–");
  63         boost::replace_all(s, "&mdash;", "—");
  64 }
  65
  66 void stripHtmlTags(std::string &s)
  67 {
  68         // Erase newlines so they don't duplicate with HTML ones.
  69         s.erase(std::remove_if(s.begin(), s.end(), [](char c) {
  70                                 return c == '\n' || c == '\r';
  71                         }), s.end());
  72
  73         bool is_newline;
  74         for (size_t i = s.find("<"); i != std::string::npos; i = s.find("<"))
  75         {
  76                 size_t j = s.find(">", i);
  77                 if (j != std::string::npos)
  78                 {
  79                         ++j;
  80                         is_newline
  81                                 =  s.compare(i, std::min<size_t>(3, j-i), "<p ") == 0
  82                                 || s.compare(i, j-i, "<p>") == 0
  83                                 || s.compare(i, j-i, "</p>") == 0
  84                                 || s.compare(i, j-i, "<br>") == 0
  85                                 || s.compare(i, j-i, "<br/>") == 0
  86                                 || s.compare(i, std::min<size_t>(4, j-i), "<br ") == 0;
  87                         if (is_newline)
  88                                 s.replace(i, j-i, "\n");
  89                         else
  90                                 s.replace(i, j-i, "");
  91                 }
  92                 else
  93                         break;
  94         }
  95         unescapeHtmlEntities(s);
  96 }