1 /* $Header: /p/tcsh/cvsroot/tcsh/tc.str.c,v 3.42 2012/01/10 21:34:31 christos Exp $ */
3 * tc.str.c: Short string package
4 * This has been a lesson of how to write buggy code!
7 * Copyright (c) 1980, 1991 The Regents of the University of California.
10 * Redistribution and use in source and binary forms, with or without
11 * modification, are permitted provided that the following conditions
13 * 1. Redistributions of source code must retain the above copyright
14 * notice, this list of conditions and the following disclaimer.
15 * 2. Redistributions in binary form must reproduce the above copyright
16 * notice, this list of conditions and the following disclaimer in the
17 * documentation and/or other materials provided with the distribution.
18 * 3. Neither the name of the University nor the names of its contributors
19 * may be used to endorse or promote products derived from this software
20 * without specific prior written permission.
22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
39 RCSID("$tcsh: tc.str.c,v 3.42 2012/01/10 21:34:31 christos Exp $")
41 #define MALLOC_INCR 128
43 #define MALLOC_SURPLUS MB_LEN_MAX /* Space for one multibyte character */
45 #define MALLOC_SURPLUS 0
50 one_mbtowc(Char *pwc, const char *s, size_t n)
54 len = rt_mbtowc(pwc, s, n);
57 *pwc = (unsigned char)*s | INVALID_BYTE;
65 one_wctomb(char *s, Char wchar)
69 if (wchar & INVALID_BYTE) {
74 if (wchar >= 0x10000) {
75 /* UTF-16 systems can't handle these values directly in calls to
76 wctomb. Convert value to UTF-16 surrogate and call wcstombs to
77 convert the "string" to the correct multibyte representation,
81 ws[0] = 0xd800 | (wchar >> 10);
82 ws[1] = 0xdc00 | (wchar & 0x3ff);
84 /* The return value of wcstombs excludes the trailing 0, so len is
85 the correct number of multibytes for the Unicode char. */
86 len = wcstombs (s, ws, MB_CUR_MAX + 1);
89 len = wctomb(s, (wchar_t) wchar);
99 rt_mbtowc(Char *pwc, const char *s, size_t n)
102 char back[MB_LEN_MAX];
104 #if defined(UTF16_STRINGS) && defined(HAVE_MBRTOWC)
105 # if defined(AUTOSET_KANJI)
106 static mbstate_t mb_zero, mb;
108 * Workaround the Shift-JIS endcoding that translates unshifted 7 bit ASCII!
110 if (!adrof(STRnokanji) && n && pwc && s && (*s == '\\' || *s == '~') &&
111 !memcmp(&mb, &mb_zero, sizeof(mb)))
120 memset (&mb, 0, sizeof mb);
121 ret = mbrtowc(&tmp, s, n, &mb);
123 ret = mbtowc(&tmp, s, n);
127 #if defined(UTF16_STRINGS) && defined(HAVE_MBRTOWC)
128 if (tmp >= 0xd800 && tmp <= 0xdbff) {
129 /* UTF-16 surrogate pair. Fetch second half and compute
130 UTF-32 value. Dispense with the inverse test in this case. */
131 size_t n2 = mbrtowc(&tmp, s + ret, n - ret, &mb);
132 if (n2 == 0 || n2 == (size_t)-1 || n2 == (size_t)-2)
135 *pwc = (((*pwc & 0x3ff) << 10) | (tmp & 0x3ff)) + 0x10000;
140 if (wctomb(back, *pwc) != ret || memcmp(s, back, ret) != 0)
143 } else if (ret == -2)
154 blk2short(char **src)
162 for (n = 0; src[n] != NULL; n++)
164 sdst = dst = xmalloc((n + 1) * sizeof(Char *));
166 for (; *src != NULL; src++)
173 short2blk(Char **src)
181 for (n = 0; src[n] != NULL; n++)
183 sdst = dst = xmalloc((n + 1) * sizeof(char *));
185 for (; *src != NULL; src++)
186 *dst++ = strsave(short2str(*src));
192 str2short(const char *src)
194 static struct Strbuf buf; /* = Strbuf_INIT; */
203 src += one_mbtowc(&wc, src, MB_LEN_MAX);
204 Strbuf_append1(&buf, wc);
206 Strbuf_terminate(&buf);
211 short2str(const Char *src)
213 static char *sdst = NULL;
214 static size_t dstsize = 0;
221 dstsize = MALLOC_INCR;
222 sdst = xmalloc((dstsize + MALLOC_SURPLUS) * sizeof(char));
225 edst = &dst[dstsize];
227 dst += one_wctomb(dst, *src & CHAR);
233 dstsize += MALLOC_INCR;
234 sdst = xrealloc(sdst, (dstsize + MALLOC_SURPLUS) * sizeof(char));
235 edst = &sdst[dstsize];
236 dst = &edst[-MALLOC_INCR];
237 while (wdst > wedst) {
247 #if !defined (WIDE_STRINGS) || defined (UTF16_STRINGS)
249 s_strcpy(Char *dst, const Char *src)
254 while ((*dst++ = *src++) != '\0')
260 s_strncpy(Char *dst, const Char *src, size_t n)
269 if ((*dst++ = *src++) == '\0') {
279 s_strcat(Char *dst, const Char *src)
281 Strcpy(Strend(dst), src);
287 s_strncat(Char *dst, const Char *src, size_t n)
300 if ((*dst++ = *src++) == '\0')
312 s_strchr(const Char *str, int ch)
316 return ((Char *)(intptr_t)str);
322 s_strrchr(const Char *str, int ch)
331 return ((Char *)(intptr_t)rstr);
335 s_strlen(const Char *str)
339 for (n = 0; *str++; n++)
345 s_strcmp(const Char *str1, const Char *str2)
347 for (; *str1 && *str1 == *str2; str1++, str2++)
350 * The following case analysis is necessary so that characters which look
351 * negative collate low against normal characters but high against the
354 if (*str1 == '\0' && *str2 == '\0')
356 else if (*str1 == '\0')
358 else if (*str2 == '\0')
361 return (*str1 - *str2);
365 s_strncmp(const Char *str1, const Char *str2, size_t n)
370 if (*str1 != *str2) {
372 * The following case analysis is necessary so that characters
373 * which look negative collate low against normal characters
374 * but high against the end-of-string NUL.
378 else if (*str2 == '\0')
381 return (*str1 - *str2);
389 #endif /* not WIDE_STRINGS */
392 s_strcasecmp(const Char *str1, const Char *str2)
395 wint_t l1 = 0, l2 = 0;
396 for (; *str1; str1++, str2++)
399 else if ((l1 = towlower(*str1)) != (l2 = towlower(*str2)))
402 unsigned char l1 = 0, l2 = 0;
403 for (; *str1; str1++, str2++)
406 else if ((l1 = tolower((unsigned char)*str1)) !=
407 (l2 = tolower((unsigned char)*str2)))
411 * The following case analysis is necessary so that characters which look
412 * negative collate low against normal characters but high against the
415 if (*str1 == '\0' && *str2 == '\0')
417 else if (*str1 == '\0')
419 else if (*str2 == '\0')
421 else if (l1 == l2) /* They are zero when they are equal */
422 return (*str1 - *str2);
428 s_strnsave(const Char *s, size_t len)
432 n = xmalloc((len + 1) * sizeof (*n));
433 memcpy(n, s, len * sizeof (*n));
439 s_strsave(const Char *s)
446 size = (Strlen(s) + 1) * sizeof(*n);
453 s_strspl(const Char *cp, const Char *dp)
466 res = xmalloc(((p - cp) + (q - dp) - 1) * sizeof(Char));
467 for (ep = res, q = cp; (*ep++ = *q++) != '\0';)
469 for (ep--, q = dp; (*ep++ = *q++) != '\0';)
475 s_strend(const Char *cp)
478 return ((Char *)(intptr_t) cp);
481 return ((Char *)(intptr_t) cp);
485 s_strstr(const Char *s, const Char *t)
493 return ((Char *)(intptr_t) s);
494 while (*ss++ == *tt++);
495 } while (*s++ != '\0');
499 #else /* !SHORT_STRINGS */
501 caching_strip(const char *s)
503 static char *buf = NULL;
504 static size_t buf_size = 0;
509 size = strlen(s) + 1;
510 if (buf_size < size) {
511 buf = xrealloc(buf, size);
514 memcpy(buf, s, size);
521 short2qstr(const Char *src)
523 static char *sdst = NULL;
524 static size_t dstsize = 0;
531 dstsize = MALLOC_INCR;
532 sdst = xmalloc((dstsize + MALLOC_SURPLUS) * sizeof(char));
535 edst = &dst[dstsize];
540 dstsize += MALLOC_INCR;
541 sdst = xrealloc(sdst,
542 (dstsize + MALLOC_SURPLUS) * sizeof(char));
543 edst = &sdst[dstsize];
544 dst = &edst[-MALLOC_INCR];
547 dst += one_wctomb(dst, *src & CHAR);
550 ptrdiff_t i = dst - edst;
551 dstsize += MALLOC_INCR;
552 sdst = xrealloc(sdst, (dstsize + MALLOC_SURPLUS) * sizeof(char));
553 edst = &sdst[dstsize];
554 dst = &edst[-MALLOC_INCR + i];
564 return xcalloc(1, sizeof(struct blk_buf));
568 bb_store(struct blk_buf *bb, Char *str)
570 if (bb->len == bb->size) { /* Keep space for terminating NULL */
572 bb->size = 16; /* Arbitrary */
575 bb->vec = xrealloc(bb->vec, bb->size * sizeof (*bb->vec));
577 bb->vec[bb->len] = str;
581 bb_append(struct blk_buf *bb, Char *str)
588 bb_cleanup(void *xbb)
594 for (i = 0; i < bb->len; i++)
607 bb_finish(struct blk_buf *bb)
610 return xrealloc(bb->vec, (bb->len + 1) * sizeof (*bb->vec));
613 #define DO_STRBUF(STRBUF, CHAR, STRLEN) \
616 STRBUF##_alloc(void) \
618 return xcalloc(1, sizeof(struct STRBUF)); \
622 STRBUF##_store1(struct STRBUF *buf, CHAR c) \
624 if (buf->size == buf->len) { \
625 if (buf->size == 0) \
626 buf->size = 64; /* Arbitrary */ \
629 buf->s = xrealloc(buf->s, buf->size * sizeof(*buf->s)); \
632 buf->s[buf->len] = c; \
635 /* Like strbuf_append1(buf, '\0'), but don't advance len */ \
637 STRBUF##_terminate(struct STRBUF *buf) \
639 STRBUF##_store1(buf, '\0'); \
643 STRBUF##_append1(struct STRBUF *buf, CHAR c) \
645 STRBUF##_store1(buf, c); \
650 STRBUF##_appendn(struct STRBUF *buf, const CHAR *s, size_t len) \
652 if (buf->size < buf->len + len) { \
653 if (buf->size == 0) \
654 buf->size = 64; /* Arbitrary */ \
655 while (buf->size < buf->len + len) \
657 buf->s = xrealloc(buf->s, buf->size * sizeof(*buf->s)); \
659 memcpy(buf->s + buf->len, s, len * sizeof(*buf->s)); \
664 STRBUF##_append(struct STRBUF *buf, const CHAR *s) \
666 STRBUF##_appendn(buf, s, STRLEN(s)); \
670 STRBUF##_finish(struct STRBUF *buf) \
672 STRBUF##_append1(buf, 0); \
673 return xrealloc(buf->s, buf->len * sizeof(*buf->s)); \
677 STRBUF##_cleanup(void *xbuf) \
679 struct STRBUF *buf; \
686 STRBUF##_free(void *xbuf) \
688 STRBUF##_cleanup(xbuf); \
692 const struct STRBUF STRBUF##_init /* = STRBUF##_INIT; */
694 DO_STRBUF(strbuf, char, strlen);
695 DO_STRBUF(Strbuf, Char, Strlen);