contrib/less/cvt.c

   1 /*
   2  * Copyright (C) 1984-2014  Mark Nudelman
   3  *
   4  * You may distribute under the terms of either the GNU General Public
   5  * License or the Less License, as specified in the README file.
   6  *
   7  * For more information, see the README file.
   8  */
   9
  10 /*
  11  * Routines to convert text in various ways.  Used by search.
  12  */
  13
  14 #include "less.h"
  15 #include "charset.h"
  16
  17 extern int utf_mode;
  18
  19 /*
  20  * Get the length of a buffer needed to convert a string.
  21  */
  22         public int
  23 cvt_length(len, ops)
  24         int len;
  25         int ops;
  26 {
  27         if (utf_mode)
  28                 /*
  29                  * Just copying a string in UTF-8 mode can cause it to grow
  30                  * in length.
  31                  * Four output bytes for one input byte is the worst case.
  32                  */
  33                 len *= 4;
  34         return (len + 1);
  35 }
  36
  37 /*
  38  * Allocate a chpos array for use by cvt_text.
  39  */
  40         public int *
  41 cvt_alloc_chpos(len)
  42         int len;
  43 {
  44         int i;
  45         int *chpos = (int *) ecalloc(sizeof(int), len);
  46         /* Initialize all entries to an invalid position. */
  47         for (i = 0;  i < len;  i++)
  48                 chpos[i] = -1;
  49         return (chpos);
  50 }
  51
  52 /*
  53  * Convert text.  Perform the transformations specified by ops.
  54  * Returns converted text in odst.  The original offset of each
  55  * odst character (when it was in osrc) is returned in the chpos array.
  56  */
  57         public void
  58 cvt_text(odst, osrc, chpos, lenp, ops)
  59         char *odst;
  60         char *osrc;
  61         int *chpos;
  62         int *lenp;
  63         int ops;
  64 {
  65         char *dst;
  66         char *edst = odst;
  67         char *src;
  68         register char *src_end;
  69         LWCHAR ch;
  70
  71         if (lenp != NULL)
  72                 src_end = osrc + *lenp;
  73         else
  74                 src_end = osrc + strlen(osrc);
  75
  76         for (src = osrc, dst = odst;  src < src_end;  )
  77         {
  78                 int src_pos = (int) (src - osrc);
  79                 int dst_pos = (int) (dst - odst);
  80                 ch = step_char(&src, +1, src_end);
  81                 if ((ops & CVT_BS) && ch == '\b' && dst > odst)
  82                 {
  83                         /* Delete backspace and preceding char. */
  84                         do {
  85                                 dst--;
  86                         } while (dst > odst &&
  87                                 !IS_ASCII_OCTET(*dst) && !IS_UTF8_LEAD(*dst));
  88                 } else if ((ops & CVT_ANSI) && IS_CSI_START(ch))
  89                 {
  90                         /* Skip to end of ANSI escape sequence. */
  91                         src++;  /* skip the CSI start char */
  92                         while (src < src_end)
  93                                 if (!is_ansi_middle(*src++))
  94                                         break;
  95                 } else
  96                 {
  97                         /* Just copy the char to the destination buffer. */
  98                         if ((ops & CVT_TO_LC) && IS_UPPER(ch))
  99                                 ch = TO_LOWER(ch);
 100                         put_wchar(&dst, ch);
 101                         /* Record the original position of the char. */
 102                         if (chpos != NULL)
 103                                 chpos[dst_pos] = src_pos;
 104                 }
 105                 if (dst > edst)
 106                         edst = dst;
 107         }
 108         if ((ops & CVT_CRLF) && edst > odst && edst[-1] == '\r')
 109                 edst--;
 110         *edst = '\0';
 111         if (lenp != NULL)
 112                 *lenp = (int) (edst - odst);
 113         /* FIXME: why was this here?  if (chpos != NULL) chpos[dst - odst] = src - osrc; */
 114 }