lib/sh/casemod.c

   1 /* casemod.c -- functions to change case of strings */
   2
   3 /* Copyright (C) 2008,2009 Free Software Foundation, Inc.
   4
   5    This file is part of GNU Bash, the Bourne Again SHell.
   6
   7    Bash is free software: you can redistribute it and/or modify
   8    it under the terms of the GNU General Public License as published by
   9    the Free Software Foundation, either version 3 of the License, or
  10    (at your option) any later version.
  11
  12    Bash is distributed in the hope that it will be useful,
  13    but WITHOUT ANY WARRANTY; without even the implied warranty of
  14    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15    GNU General Public License for more details.
  16
  17    You should have received a copy of the GNU General Public License
  18    along with Bash.  If not, see <http://www.gnu.org/licenses/>.
  19 */
  20
  21 #if defined (HAVE_CONFIG_H)
  22 #  include <config.h>
  23 #endif
  24
  25 #if defined (HAVE_UNISTD_H)
  26 #  include <unistd.h>
  27 #endif /* HAVE_UNISTD_H */
  28
  29 #include <stdc.h>
  30
  31 #include <bashansi.h>
  32 #include <bashintl.h>
  33 #include <bashtypes.h>
  34
  35 #include <stdio.h>
  36 #include <ctype.h>
  37 #include <xmalloc.h>
  38
  39 #include <shmbutil.h>
  40 #include <chartypes.h>
  41
  42 #include <glob/strmatch.h>
  43
  44 #define _to_wupper(wc)  (iswlower (wc) ? towupper (wc) : (wc))
  45 #define _to_wlower(wc)  (iswupper (wc) ? towlower (wc) : (wc))
  46
  47 #if !defined (HANDLE_MULTIBYTE)
  48 #  define cval(s, i)    ((s)[(i)])
  49 #  define iswalnum(c)   (isalnum(c))
  50 #  define TOGGLE(x)     (ISUPPER (x) ? tolower (x) : (TOUPPER (x)))
  51 #else
  52 #  define TOGGLE(x)     (iswupper (x) ? towlower (x) : (_to_wupper(x)))
  53 #endif
  54
  55 /* These must agree with the defines in externs.h */
  56 #define CASE_NOOP       0x0000
  57 #define CASE_LOWER      0x0001
  58 #define CASE_UPPER      0x0002
  59 #define CASE_CAPITALIZE 0x0004
  60 #define CASE_UNCAP      0x0008
  61 #define CASE_TOGGLE     0x0010
  62 #define CASE_TOGGLEALL  0x0020
  63 #define CASE_UPFIRST    0x0040
  64 #define CASE_LOWFIRST   0x0080
  65
  66 #define CASE_USEWORDS   0x1000          /* modify behavior to act on words in passed string */
  67
  68 extern char *substring __P((char *, int, int));
  69
  70 #if defined (HANDLE_MULTIBYTE)
  71 static wchar_t
  72 cval (s, i)
  73      char *s;
  74      int i;
  75 {
  76   size_t tmp;
  77   wchar_t wc;
  78   int l;
  79   mbstate_t mps;
  80
  81   if (MB_CUR_MAX == 1)
  82     return ((wchar_t)s[i]);
  83   l = strlen (s);
  84   if (i >= (l - 1))
  85     return ((wchar_t)s[i]);
  86   memset (&mps, 0, sizeof (mbstate_t));
  87   tmp = mbrtowc (&wc, s + i, l - i, &mps);
  88   if (MB_INVALIDCH (tmp) || MB_NULLWCH (tmp))
  89     return ((wchar_t)s[i]);
  90   return wc;
  91 }
  92 #endif
  93
  94 /* Modify the case of characters in STRING matching PAT based on the value of
  95    FLAGS.  If PAT is null, modify the case of each character */
  96 char *
  97 sh_modcase (string, pat, flags)
  98      const char *string;
  99      char *pat;
 100      int flags;
 101 {
 102   int start, next, end;
 103   int inword, c, nc, nop, match, usewords;
 104   char *ret, *s;
 105   wchar_t wc;
 106 #if defined (HANDLE_MULTIBYTE)
 107   wchar_t nwc;
 108   char mb[MB_LEN_MAX+1];
 109   int mlen;
 110   size_t m;
 111   mbstate_t state;
 112 #endif
 113
 114   if (string == 0 || *string == 0)
 115     {
 116       ret = (char *)xmalloc (1);
 117       ret[0] = '\0';
 118       return ret;
 119     }
 120
 121 #if defined (HANDLE_MULTIBYTE)
 122   memset (&state, 0, sizeof (mbstate_t));
 123 #endif
 124
 125   start = 0;
 126   end = strlen (string);
 127
 128   ret = (char *)xmalloc (end + 1);
 129   strcpy (ret, string);
 130
 131   /* See if we are supposed to split on alphanumerics and operate on each word */
 132   usewords = (flags & CASE_USEWORDS);
 133   flags &= ~CASE_USEWORDS;
 134
 135   inword = 0;
 136   while (start < end)
 137     {
 138       wc = cval (ret, start);
 139
 140       if (iswalnum (wc) == 0)
 141         {
 142           inword = 0;
 143           ADVANCE_CHAR (ret, end, start);
 144           continue;
 145         }
 146
 147       if (pat)
 148         {
 149           next = start;
 150           ADVANCE_CHAR (ret, end, next);
 151           s = substring (ret, start, next);
 152           match = strmatch (pat, s, FNM_EXTMATCH) != FNM_NOMATCH;
 153           free (s);
 154           if (match == 0)
 155             {
 156               start = next;
 157               inword = 1;
 158               continue;
 159             }
 160         }
 161
 162       /* XXX - for now, the toggling operators work on the individual
 163          words in the string, breaking on alphanumerics.  Should I
 164          leave the capitalization operators to do that also? */
 165       if (flags == CASE_CAPITALIZE)
 166         {
 167           if (usewords)
 168             nop = inword ? CASE_LOWER : CASE_UPPER;
 169           else
 170             nop = (start > 0) ? CASE_LOWER : CASE_UPPER;
 171           inword = 1;
 172         }
 173       else if (flags == CASE_UNCAP)
 174         {
 175           if (usewords)
 176             nop = inword ? CASE_UPPER : CASE_LOWER;
 177           else
 178             nop = (start > 0) ? CASE_UPPER : CASE_LOWER;
 179           inword = 1;
 180         }
 181       else if (flags == CASE_UPFIRST)
 182         {
 183           if (usewords)
 184             nop = inword ? CASE_NOOP : CASE_UPPER;
 185           else
 186             nop = (start > 0) ? CASE_NOOP : CASE_UPPER;
 187           inword = 1;
 188         }
 189       else if (flags == CASE_LOWFIRST)
 190         {
 191           if (usewords)
 192             nop = inword ? CASE_NOOP : CASE_LOWER;
 193           else
 194             nop = (start > 0) ? CASE_NOOP : CASE_LOWER;
 195           inword = 1;
 196         }
 197       else if (flags == CASE_TOGGLE)
 198         {
 199           nop = inword ? CASE_NOOP : CASE_TOGGLE;
 200           inword = 1;
 201         }
 202       else
 203         nop = flags;
 204
 205       if (MB_CUR_MAX == 1 || isascii (wc))
 206         {
 207           switch (nop)
 208           {
 209           default:
 210           case CASE_NOOP:  nc = wc; break;
 211           case CASE_UPPER:  nc = TOUPPER (wc); break;
 212           case CASE_LOWER:  nc = TOLOWER (wc); break;
 213           case CASE_TOGGLEALL:
 214           case CASE_TOGGLE: nc = TOGGLE (wc); break;
 215           }
 216           ret[start] = nc;
 217         }
 218 #if defined (HANDLE_MULTIBYTE)
 219       else
 220         {
 221           m = mbrtowc (&wc, string + start, end - start, &state);
 222           if (MB_INVALIDCH (m))
 223             wc = (wchar_t)string[start];
 224           else if (MB_NULLWCH (m))
 225             wc = L'\0';
 226           switch (nop)
 227           {
 228           default:
 229           case CASE_NOOP:  nwc = wc; break;
 230           case CASE_UPPER:  nwc = TOUPPER (wc); break;
 231           case CASE_LOWER:  nwc = TOLOWER (wc); break;
 232           case CASE_TOGGLEALL:
 233           case CASE_TOGGLE: nwc = TOGGLE (wc); break;
 234           }
 235           if  (nwc != wc)       /*  just skip unchanged characters */
 236             {
 237               mlen = wcrtomb (mb, nwc, &state);
 238               if (mlen > 0)
 239                 mb[mlen] = '\0';
 240               /* Assume the same width */
 241               strncpy (ret + start, mb, mlen);
 242             }
 243         }
 244 #endif
 245
 246       /*  This assumes that the upper and lower case versions are the same width. */
 247       ADVANCE_CHAR (ret, end, start);
 248     }
 249
 250   return ret;
 251 }