NeoMutt  2021-02-05-666-ge300cd
Teaching an old dog new tricks
DOXYGEN
mbyte.c File Reference

Multi-byte String manipulation functions. More...

#include "config.h"
#include <ctype.h>
#include <limits.h>
#include <stdbool.h>
#include <stdlib.h>
#include <string.h>
#include <wchar.h>
#include <wctype.h>
#include "mbyte.h"
#include "buffer.h"
#include "charset.h"
#include "memory.h"
#include "string2.h"
+ Include dependency graph for mbyte.c:

Go to the source code of this file.

Functions

int mutt_mb_charlen (const char *s, int *width)
 Count the bytes in a (multibyte) character. More...
 
bool mutt_mb_get_initials (const char *name, char *buf, size_t buflen)
 Turn a name into initials. More...
 
int mutt_mb_width (const char *str, int col, bool display)
 Measure a string's display width (in screen columns) More...
 
int mutt_mb_wcwidth (wchar_t wc)
 Measure the screen width of a character. More...
 
int mutt_mb_wcswidth (const wchar_t *s, size_t n)
 Measure the screen width of a string. More...
 
size_t mutt_mb_width_ceiling (const wchar_t *s, size_t n, int w1)
 Keep the end of the string on-screen. More...
 
void mutt_mb_wcstombs (char *dest, size_t dlen, const wchar_t *src, size_t slen)
 Convert a string from wide to multibyte characters. More...
 
size_t mutt_mb_mbstowcs (wchar_t **pwbuf, size_t *pwbuflen, size_t i, const char *buf)
 Convert a string from multibyte to wide characters. More...
 
bool mutt_mb_is_shell_char (wchar_t ch)
 Is character not typically part of a pathname. More...
 
bool mutt_mb_is_lower (const char *s)
 Does a multi-byte string contain only lowercase characters? More...
 
bool mutt_mb_is_display_corrupting_utf8 (wchar_t wc)
 Will this character corrupt the display? More...
 
int mutt_mb_filter_unprintable (char **s)
 Replace unprintable characters. More...
 

Variables

bool OptLocales
 (pseudo) set if user has valid locale definition More...
 

Detailed Description

Multi-byte String manipulation functions.

Authors
  • Richard Russon

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

Definition in file mbyte.c.

Function Documentation

◆ mutt_mb_charlen()

int mutt_mb_charlen ( const char *  s,
int *  width 
)

Count the bytes in a (multibyte) character.

Parameters
[in]sString to be examined
[out]widthNumber of screen columns the character would use
Return values
numBytes in the first (multibyte) character of input consumes
<0Conversion error
=0End of input
>0Length (bytes)

Definition at line 54 of file mbyte.c.

55 {
56  if (!s || (*s == '\0'))
57  return 0;
58 
59  wchar_t wc;
60  mbstate_t mbstate;
61  size_t k, n;
62 
63  n = mutt_str_len(s);
64  memset(&mbstate, 0, sizeof(mbstate));
65  k = mbrtowc(&wc, s, n, &mbstate);
66  if (width)
67  *width = wcwidth(wc);
68  return ((k == (size_t) (-1)) || (k == (size_t) (-2))) ? -1 : k;
69 }
size_t mutt_str_len(const char *a)
Calculate the length of a string, safely.
Definition: string.c:664
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ mutt_mb_get_initials()

bool mutt_mb_get_initials ( const char *  name,
char *  buf,
size_t  buflen 
)

Turn a name into initials.

Parameters
nameString to be converted
bufBuffer for the result
buflenSize of the buffer
Return values
1Success
0Failure

Take a name, e.g. "John F. Kennedy" and reduce it to initials "JFK". The function saves the first character from each word. Words are delimited by whitespace, or hyphens (so "Jean-Pierre" becomes "JP").

Definition at line 83 of file mbyte.c.

84 {
85  if (!name || !buf)
86  return false;
87 
88  while (*name)
89  {
90  /* Char's length in bytes */
91  int clen = mutt_mb_charlen(name, NULL);
92  if (clen < 1)
93  return false;
94 
95  /* Ignore punctuation at the beginning of a word */
96  if ((clen == 1) && ispunct(*name))
97  {
98  name++;
99  continue;
100  }
101 
102  if (clen >= buflen)
103  return false;
104 
105  /* Copy one multibyte character */
106  buflen -= clen;
107  while (clen--)
108  *buf++ = *name++;
109 
110  /* Skip to end-of-word */
111  for (; *name; name += clen)
112  {
113  clen = mutt_mb_charlen(name, NULL);
114  if (clen < 1)
115  return false;
116  if ((clen == 1) && (isspace(*name) || (*name == '-')))
117  break;
118  }
119 
120  /* Skip any whitespace, or hyphens */
121  while (*name && (isspace(*name) || (*name == '-')))
122  name++;
123  }
124 
125  *buf = '\0';
126  return true;
127 }
int mutt_mb_charlen(const char *s, int *width)
Count the bytes in a (multibyte) character.
Definition: mbyte.c:54
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ mutt_mb_width()

int mutt_mb_width ( const char *  str,
int  col,
bool  display 
)

Measure a string's display width (in screen columns)

Parameters
strString to measure
colDisplay column (used for expanding tabs)
displaywill this be displayed to the user?
Return values
numStrings width in screen columns

This is like wcwidth(), but gets const char* not wchar_t*.

Definition at line 138 of file mbyte.c.

139 {
140  wchar_t wc;
141  int l, w = 0, nl = 0;
142  const char *p = str;
143 
144  while (p && *p)
145  {
146  if (mbtowc(&wc, p, MB_CUR_MAX) >= 0)
147  {
148  l = wcwidth(wc);
149  if (l < 0)
150  l = 1;
151  /* correctly calc tab stop, even for sending as the
152  * line should look pretty on the receiving end */
153  if ((wc == L'\t') || (nl && (wc == L' ')))
154  {
155  nl = 0;
156  l = 8 - (col % 8);
157  }
158  /* track newlines for display-case: if we have a space
159  * after a newline, assume 8 spaces as for display we
160  * always tab-fold */
161  else if (display && (wc == '\n'))
162  nl = 1;
163  }
164  else
165  l = 1;
166  w += l;
167  p++;
168  }
169  return w;
170 }
+ Here is the caller graph for this function:

◆ mutt_mb_wcwidth()

int mutt_mb_wcwidth ( wchar_t  wc)

Measure the screen width of a character.

Parameters
wcCharacter to examine
Return values
numWidth in screen columns

Definition at line 177 of file mbyte.c.

178 {
179  int n = wcwidth(wc);
180  if (IsWPrint(wc) && (n > 0))
181  return n;
182  if (!(wc & ~0x7f))
183  return 2;
184  if (!(wc & ~0xffff))
185  return 6;
186  return 10;
187 }
#define IsWPrint(wc)
Definition: mbyte.h:39
+ Here is the caller graph for this function:

◆ mutt_mb_wcswidth()

int mutt_mb_wcswidth ( const wchar_t *  s,
size_t  n 
)

Measure the screen width of a string.

Parameters
sString to measure
nLength of string in characters
Return values
numWidth in screen columns

Definition at line 195 of file mbyte.c.

196 {
197  if (!s)
198  return 0;
199 
200  int w = 0;
201  while (n--)
202  w += mutt_mb_wcwidth(*s++);
203  return w;
204 }
int mutt_mb_wcwidth(wchar_t wc)
Measure the screen width of a character.
Definition: mbyte.c:177
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ mutt_mb_width_ceiling()

size_t mutt_mb_width_ceiling ( const wchar_t *  s,
size_t  n,
int  w1 
)

Keep the end of the string on-screen.

Parameters
sString being displayed
nLength of string in characters
w1Width limit
Return values
numChars to skip

Given a string and a width, determine how many characters from the beginning of the string should be skipped so that the string fits.

Definition at line 216 of file mbyte.c.

217 {
218  if (!s)
219  return 0;
220 
221  const wchar_t *s0 = s;
222  int w = 0;
223  for (; n; s++, n--)
224  if ((w += mutt_mb_wcwidth(*s)) > w1)
225  break;
226  return s - s0;
227 }
int mutt_mb_wcwidth(wchar_t wc)
Measure the screen width of a character.
Definition: mbyte.c:177
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ mutt_mb_wcstombs()

void mutt_mb_wcstombs ( char *  dest,
size_t  dlen,
const wchar_t *  src,
size_t  slen 
)

Convert a string from wide to multibyte characters.

Parameters
destBuffer for the result
dlenLength of the result buffer
srcSource string to convert
slenLength of the source string

Definition at line 236 of file mbyte.c.

237 {
238  if (!dest || !src)
239  return;
240 
241  mbstate_t st;
242  size_t k;
243 
244  /* First convert directly into the destination buffer */
245  memset(&st, 0, sizeof(st));
246  for (; slen && dlen >= MB_LEN_MAX; dest += k, dlen -= k, src++, slen--)
247  {
248  k = wcrtomb(dest, *src, &st);
249  if (k == (size_t) (-1))
250  break;
251  }
252 
253  /* If this works, we can stop now */
254  if (dlen >= MB_LEN_MAX)
255  {
256  dest += wcrtomb(dest, 0, &st);
257  return;
258  }
259 
260  /* Otherwise convert any remaining data into a local buffer */
261  {
262  char buf[3 * MB_LEN_MAX];
263  char *p = buf;
264 
265  for (; slen && p - buf < dlen; p += k, src++, slen--)
266  {
267  k = wcrtomb(p, *src, &st);
268  if (k == (size_t) (-1))
269  break;
270  }
271  p += wcrtomb(p, 0, &st);
272 
273  /* If it fits into the destination buffer, we can stop now */
274  if (p - buf <= dlen)
275  {
276  memcpy(dest, buf, p - buf);
277  return;
278  }
279 
280  /* Otherwise we truncate the string in an ugly fashion */
281  memcpy(dest, buf, dlen);
282  dest[dlen - 1] = '\0'; /* assume original dlen > 0 */
283  }
284 }
+ Here is the caller graph for this function:

◆ mutt_mb_mbstowcs()

size_t mutt_mb_mbstowcs ( wchar_t **  pwbuf,
size_t *  pwbuflen,
size_t  i,
const char *  buf 
)

Convert a string from multibyte to wide characters.

Parameters
[out]pwbufBuffer for the result
[out]pwbuflenLength of the result buffer
[in]iStarting index into the result buffer
[in]bufString to convert
Return values
numFirst character after the result

Definition at line 294 of file mbyte.c.

295 {
296  if (!pwbuf || !pwbuflen || !buf)
297  return 0;
298 
299  wchar_t wc;
300  mbstate_t st;
301  size_t k;
302  wchar_t *wbuf = *pwbuf;
303  size_t wbuflen = *pwbuflen;
304 
305  while (*buf != '\0')
306  {
307  memset(&st, 0, sizeof(st));
308  for (; (k = mbrtowc(&wc, buf, MB_LEN_MAX, &st)) && k != (size_t) (-1) &&
309  k != (size_t) (-2);
310  buf += k)
311  {
312  if (i >= wbuflen)
313  {
314  wbuflen = i + 20;
315  mutt_mem_realloc(&wbuf, wbuflen * sizeof(*wbuf));
316  }
317  wbuf[i++] = wc;
318  }
319  if ((*buf != '\0') && ((k == (size_t) -1) || (k == (size_t) -2)))
320  {
321  if (i >= wbuflen)
322  {
323  wbuflen = i + 20;
324  mutt_mem_realloc(&wbuf, wbuflen * sizeof(*wbuf));
325  }
326  wbuf[i++] = ReplacementChar;
327  buf++;
328  }
329  }
330  *pwbuf = wbuf;
331  *pwbuflen = wbuflen;
332  return i;
333 }
void mutt_mem_realloc(void *ptr, size_t size)
Resize a block of memory on the heap.
Definition: memory.c:114
wchar_t ReplacementChar
When a Unicode character can&#39;t be displayed, use this instead.
Definition: charset.c:57
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

◆ mutt_mb_is_shell_char()

bool mutt_mb_is_shell_char ( wchar_t  ch)

Is character not typically part of a pathname.

Parameters
chCharacter to examine
Return values
trueCharacter is not typically part of a pathname
falseCharacter is typically part of a pathname
Note
The name is very confusing.

Definition at line 343 of file mbyte.c.

344 {
345  static const wchar_t shell_chars[] = L"<>&()$?*;{}| "; /* ! not included because it can be part of a pathname in NeoMutt */
346  return wcschr(shell_chars, ch);
347 }
+ Here is the caller graph for this function:

◆ mutt_mb_is_lower()

bool mutt_mb_is_lower ( const char *  s)

Does a multi-byte string contain only lowercase characters?

Parameters
sString to check
Return values
trueString contains no uppercase characters
falseError, or contains some uppercase characters

Non-alphabetic characters are considered lowercase.

Definition at line 357 of file mbyte.c.

358 {
359  if (!s)
360  return false;
361 
362  wchar_t w;
363  mbstate_t mb;
364  size_t l;
365 
366  memset(&mb, 0, sizeof(mb));
367 
368  for (; (l = mbrtowc(&w, s, MB_CUR_MAX, &mb)) != 0; s += l)
369  {
370  if (l == (size_t) -2)
371  continue; /* shift sequences */
372  if (l == (size_t) -1)
373  return false;
374  if (iswalpha((wint_t) w) && iswupper((wint_t) w))
375  return false;
376  }
377 
378  return true;
379 }
+ Here is the caller graph for this function:

◆ mutt_mb_is_display_corrupting_utf8()

bool mutt_mb_is_display_corrupting_utf8 ( wchar_t  wc)

Will this character corrupt the display?

Parameters
wcCharacter to examine
Return values
trueCharacter would corrupt the display
falseCharacter is safe to display
Note
This list isn't complete.

Definition at line 389 of file mbyte.c.

390 {
391  if ((wc == (wchar_t) 0x00ad) || /* soft hyphen */
392  (wc == (wchar_t) 0x200e) || /* left-to-right mark */
393  (wc == (wchar_t) 0x200f) || /* right-to-left mark */
394  (wc == (wchar_t) 0xfeff)) /* zero width no-break space */
395  {
396  return true;
397  }
398 
399  /* left-to-right isolate, right-to-left isolate, first strong isolate,
400  * pop directional isolate */
401  if ((wc >= (wchar_t) 0x2066) && (wc <= (wchar_t) 0x2069))
402  return true;
403 
404  /* left-to-right embedding, right-to-left embedding, pop directional formatting,
405  * left-to-right override, right-to-left override */
406  if ((wc >= (wchar_t) 0x202a) && (wc <= (wchar_t) 0x202e))
407  return true;
408 
409  return false;
410 }
+ Here is the caller graph for this function:

◆ mutt_mb_filter_unprintable()

int mutt_mb_filter_unprintable ( char **  s)

Replace unprintable characters.

Parameters
[in,out]sString to modify
Return values
0Success
-1Error

Unprintable characters will be replaced with ReplacementChar.

Note
The source string will be freed and a newly allocated string will be returned in its place. The caller should free the returned string.

Definition at line 423 of file mbyte.c.

424 {
425  if (!s || !*s)
426  return -1;
427 
428  wchar_t wc;
429  size_t k, k2;
430  char scratch[MB_LEN_MAX + 1];
431  char *p = *s;
432  mbstate_t mbstate1, mbstate2;
433 
434  struct Buffer buf = mutt_buffer_make(0);
435  memset(&mbstate1, 0, sizeof(mbstate1));
436  memset(&mbstate2, 0, sizeof(mbstate2));
437  for (; (k = mbrtowc(&wc, p, MB_LEN_MAX, &mbstate1)); p += k)
438  {
439  if ((k == (size_t) -1) || (k == (size_t) -2))
440  {
441  k = 1;
442  memset(&mbstate1, 0, sizeof(mbstate1));
443  wc = ReplacementChar;
444  }
445  if (!IsWPrint(wc))
446  wc = '?';
448  continue;
449  k2 = wcrtomb(scratch, wc, &mbstate2);
450  scratch[k2] = '\0';
451  mutt_buffer_addstr(&buf, scratch);
452  }
453  FREE(s);
454  *s = buf.data ? buf.data : mutt_mem_calloc(1, 1);
455  return 0;
456 }
void * mutt_mem_calloc(size_t nmemb, size_t size)
Allocate zeroed memory on the heap.
Definition: memory.c:50
bool mutt_mb_is_display_corrupting_utf8(wchar_t wc)
Will this character corrupt the display?
Definition: mbyte.c:389
#define IsWPrint(wc)
Definition: mbyte.h:39
struct Buffer mutt_buffer_make(size_t size)
Make a new buffer on the stack.
Definition: buffer.c:61
String manipulation buffer.
Definition: buffer.h:33
size_t mutt_buffer_addstr(struct Buffer *buf, const char *s)
Add a string to a Buffer.
Definition: buffer.c:225
char * data
Pointer to data.
Definition: buffer.h:35
wchar_t ReplacementChar
When a Unicode character can&#39;t be displayed, use this instead.
Definition: charset.c:57
bool CharsetIsUtf8
Is the user&#39;s current character set utf-8?
Definition: charset.c:62
#define FREE(x)
Definition: memory.h:40
+ Here is the call graph for this function:
+ Here is the caller graph for this function:

Variable Documentation

◆ OptLocales

bool OptLocales

(pseudo) set if user has valid locale definition

Definition at line 43 of file mbyte.c.