Line data    Source code

       1             : // Copyright (c) 2011-2023  Made to Order Software Corp.  All Rights Reserved
       2             : //
       3             : // https://snapwebsites.org/project/snapdev
       4             : // contact@m2osw.com
       5             : //
       6             : // This program is free software: you can redistribute it and/or modify
       7             : // it under the terms of the GNU General Public License as published by
       8             : // the Free Software Foundation, either version 3 of the License, or
       9             : // (at your option) any later version.
      10             : //
      11             : // This program is distributed in the hope that it will be useful,
      12             : // but WITHOUT ANY WARRANTY; without even the implied warranty of
      13             : // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
      14             : // GNU General Public License for more details.
      15             : //
      16             : // You should have received a copy of the GNU General Public License
      17             : // along with this program.  If not, see <https://www.gnu.org/licenses/>.
      18             : #pragma once
      19             : 
      20             : /** \file
      21             :  * \brief Template used to trim a string.
      22             :  *
      23             :  * This function will trim the start, end, and inside of a string.
      24             :  *
      25             :  * The function works with any standard string. The characters to trim
      26             :  * are themselves to be defined in a string.
      27             :  *
      28             :  * The function also takes three flags to determine where the trimming
      29             :  * is to happen.
      30             :  */
      31             : 
      32             : // self
      33             : //
      34             : #include    <snapdev/reverse_cstring.h>
      35             : 
      36             : 
      37             : // C++
      38             : //
      39             : #include    <algorithm>
      40             : #include    <cwctype>
      41             : #include    <string>
      42             : 
      43             : 
      44             : 
      45             : namespace snapdev
      46             : {
      47             : 
      48             : 
      49             : /** \brief Trim a string.
      50             :  *
      51             :  * This function goes through the input string and remove
      52             :  * \p trim_characters from the \p trim_start, the \p trim_end, and the
      53             :  * \p trim_inside.
      54             :  *
      55             :  * If no \p trim_characters are specified (empty string) then the default
      56             :  * applies. The default uses the `iswspace()` on each character.
      57             :  *
      58             :  * \todo
      59             :  * Add support for UTF-8 (in which case we need to move this to our
      60             :  * libutf8 library?).
      61             :  *
      62             :  * \tparam StringT  The type of container used to output the tokens.
      63             :  * \param[in] str  The string to trim.
      64             :  * \param[in] trim_start  Trim the start of the input string.
      65             :  * \param[in] trim_end  Trim the end of the input string.
      66             :  * \param[in] trim_inside  Trim the inside of the input string.
      67             :  * \param[in] trim_characters  Trim those characters from the start/inside/end.
      68             :  *
      69             :  * \return The trimmed string.
      70             :  */
      71             : template<class StringT>
      72          48 : StringT trim_string(
      73             :           StringT const & str
      74             :         , bool const trim_start = true
      75             :         , bool const trim_end = true
      76             :         , bool const trim_inside = false
      77             :         , StringT const & trim_characters = StringT())
      78             : {
      79          48 :     typename StringT::value_type const * start(str.data());
      80          48 :     typename StringT::value_type const * end(str.data() + str.length());
      81             : 
      82          48 :     if(trim_start)
      83             :     {
      84          24 :         start = std::find_if_not(
      85             :               start
      86             :             , end
      87          96 :             , [&trim_characters](auto const c)
      88             :               {
      89          64 :                   if(trim_characters.empty())
      90             :                   {
      91          32 :                       return std::iswspace(c) != 0;
      92             :                   }
      93          32 :                   return trim_characters.find(c) != StringT::npos;
      94             :               });
      95             :     }
      96             : 
      97          48 :     if(start < end
      98          48 :     && trim_end)
      99             :     {
     100          24 :         reverse_cstring<typename StringT::value_type const> const rstr(start, end);
     101          24 :         auto const p(std::find_if_not(
     102             :               rstr.begin()
     103             :             , rstr.end()
     104          80 :             , [&trim_characters](auto const c)
     105             :               {
     106          56 :                   if(trim_characters.empty())
     107             :                   {
     108          32 :                       return std::iswspace(c) != 0;
     109             :                   }
     110          24 :                   return trim_characters.find(c) != StringT::npos;
     111             :               }));
     112          24 :         end = p.get();
     113             :     }
     114             : 
     115          48 :     if(!trim_inside)
     116             :     {
     117          24 :         return StringT(start, end);
     118             :     }
     119             : 
     120             :     // in this case, we want to copy the characters one by one
     121             :     //
     122          24 :     bool space(false);
     123          24 :     StringT result;
     124          24 :     result.reserve(end - start);
     125             : 
     126          24 :     if(!trim_start)
     127             :     {
     128          88 :         for(; start < end && !space; ++start)
     129             :         {
     130          76 :             result += *start;
     131          76 :             if(trim_characters.empty())
     132             :             {
     133          38 :                 space = std::iswspace(*start);
     134             :             }
     135             :             else
     136             :             {
     137          38 :                 space = trim_characters.find(*start) != StringT::npos;
     138             :             }
     139             :         }
     140             :     }
     141             : 
     142          24 :     if(trim_characters.empty())
     143             :     {
     144         418 :         for(; start < end; ++start)
     145             :         {
     146         406 :             if(space)
     147             :             {
     148         180 :                 space = std::iswspace(*start);
     149         180 :                 if(!space)
     150             :                 {
     151          40 :                     result += *start;
     152             :                 }
     153             :             }
     154             :             else
     155             :             {
     156         226 :                 space = std::iswspace(*start);
     157         226 :                 if(space)
     158             :                 {
     159          40 :                     result += ' ';
     160             :                 }
     161             :                 else
     162             :                 {
     163         186 :                     result += *start;
     164             :                 }
     165             :             }
     166             :         }
     167             :     }
     168             :     else
     169             :     {
     170         422 :         for(; start < end; ++start)
     171             :         {
     172         410 :             if(space)
     173             :             {
     174         150 :                 space = trim_characters.find(*start) != StringT::npos;
     175         150 :                 if(!space)
     176             :                 {
     177          68 :                     result += *start;
     178             :                 }
     179             :             }
     180             :             else
     181             :             {
     182         260 :                 space = trim_characters.find(*start) != StringT::npos;
     183         260 :                 if(space)
     184             :                 {
     185          66 :                     result += ' ';
     186             :                 }
     187             :                 else
     188             :                 {
     189         194 :                     result += *start;
     190             :                 }
     191             :             }
     192             :         }
     193             :     }
     194             : 
     195          24 :     return result;
     196          24 : }
     197             : 
     198             : } // namespace snapdev
     199             : // vim: ts=4 sw=4 et