diff options
| -rw-r--r-- | src/library.js | 19 | ||||
| -rw-r--r-- | system/lib/libc/musl/src/string/strstr.c | 156 | ||||
| -rw-r--r-- | system/lib/libcextra.symbols | 1 | ||||
| -rw-r--r-- | tools/system_libs.py | 1 |
4 files changed, 158 insertions, 19 deletions
diff --git a/src/library.js b/src/library.js index 543f6f04..fbef0d15 100644 --- a/src/library.js +++ b/src/library.js @@ -3639,25 +3639,6 @@ LibraryManager.library = { return pdest|0; }, - strstr: function(ptr1, ptr2) { - var check = 0, start; - do { - if (!check) { - start = ptr1; - check = ptr2; - } - var curr1 = {{{ makeGetValue('ptr1++', 0, 'i8') }}}; - var curr2 = {{{ makeGetValue('check++', 0, 'i8') }}}; - if (curr2 == 0) return start; - if (curr2 != curr1) { - // rewind to one character after start, to find ez in eeez - ptr1 = start + 1; - check = 0; - } - } while (curr1); - return 0; - }, - strchr: function(ptr, chr) { ptr--; do { diff --git a/system/lib/libc/musl/src/string/strstr.c b/system/lib/libc/musl/src/string/strstr.c new file mode 100644 index 00000000..06491748 --- /dev/null +++ b/system/lib/libc/musl/src/string/strstr.c @@ -0,0 +1,156 @@ +#include <string.h> +#include <stdlib.h> +#include <stdint.h> + +static char *twobyte_strstr(const unsigned char *h, const unsigned char *n) +{ + uint16_t nw = n[0]<<8 | n[1], hw = h[0]<<8 | h[1]; + for (h++; *h && hw != nw; hw = hw<<8 | *++h); + return *h ? (char *)h-1 : 0; +} + +static char *threebyte_strstr(const unsigned char *h, const unsigned char *n) +{ + uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8; + uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8; + for (h+=2; *h && hw != nw; hw = (hw|*++h)<<8); + return *h ? (char *)h-2 : 0; +} + +static char *fourbyte_strstr(const unsigned char *h, const unsigned char *n) +{ + uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8 | n[3]; + uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8 | h[3]; + for (h+=3; *h && hw != nw; hw = hw<<8 | *++h); + return *h ? (char *)h-3 : 0; +} + +#define MAX(a,b) ((a)>(b)?(a):(b)) +#define MIN(a,b) ((a)<(b)?(a):(b)) + +#define BITOP(a,b,op) \ + ((a)[(size_t)(b)/(8*sizeof *(a))] op (size_t)1<<((size_t)(b)%(8*sizeof *(a)))) + +static char *twoway_strstr(const unsigned char *h, const unsigned char *n) +{ + const unsigned char *z; + size_t l, ip, jp, k, p, ms, p0, mem, mem0; + size_t byteset[32 / sizeof(size_t)] = { 0 }; + size_t shift[256]; + + /* Computing length of needle and fill shift table */ + for (l=0; n[l] && h[l]; l++) + BITOP(byteset, n[l], |=), shift[n[l]] = l+1; + if (n[l]) return 0; /* hit the end of h */ + + /* Compute maximal suffix */ + ip = -1; jp = 0; k = p = 1; + while (jp+k<l) { + if (n[ip+k] == n[jp+k]) { + if (k == p) { + jp += p; + k = 1; + } else k++; + } else if (n[ip+k] > n[jp+k]) { + jp += k; + k = 1; + p = jp - ip; + } else { + ip = jp++; + k = p = 1; + } + } + ms = ip; + p0 = p; + + /* And with the opposite comparison */ + ip = -1; jp = 0; k = p = 1; + while (jp+k<l) { + if (n[ip+k] == n[jp+k]) { + if (k == p) { + jp += p; + k = 1; + } else k++; + } else if (n[ip+k] < n[jp+k]) { + jp += k; + k = 1; + p = jp - ip; + } else { + ip = jp++; + k = p = 1; + } + } + if (ip+1 > ms+1) ms = ip; + else p = p0; + + /* Periodic needle? */ + if (memcmp(n, n+p, ms+1)) { + mem0 = 0; + p = MAX(ms, l-ms-1) + 1; + } else mem0 = l-p; + mem = 0; + + /* Initialize incremental end-of-haystack pointer */ + z = h; + + /* Search loop */ + for (;;) { + /* Update incremental end-of-haystack pointer */ + if (z-h < l) { + /* Fast estimate for MIN(l,63) */ + size_t grow = l | 63; + const unsigned char *z2 = memchr(z, 0, grow); + if (z2) { + z = z2; + if (z-h < l) return 0; + } else z += grow; + } + + /* Check last byte first; advance by shift on mismatch */ + if (BITOP(byteset, h[l-1], &)) { + k = l-shift[h[l-1]]; + //printf("adv by %zu (on %c) at [%s] (%zu;l=%zu)\n", k, h[l-1], h, shift[h[l-1]], l); + if (k) { + if (mem0 && mem && k < p) k = l-p; + h += k; + mem = 0; + continue; + } + } else { + h += l; + mem = 0; + continue; + } + + /* Compare right half */ + for (k=MAX(ms+1,mem); n[k] && n[k] == h[k]; k++); + if (n[k]) { + h += k-ms; + mem = 0; + continue; + } + /* Compare left half */ + for (k=ms+1; k>mem && n[k-1] == h[k-1]; k--); + if (k == mem) return (char *)h; + h += p; + mem = mem0; + } +} + +char *strstr(const char *h, const char *n) +{ + /* Return immediately on empty needle */ + if (!n[0]) return (char *)h; + + /* Use faster algorithms for short needles */ + h = strchr(h, *n); + if (!h || !n[1]) return (char *)h; + if (!h[1]) return 0; + if (!n[2]) return twobyte_strstr((void *)h, (void *)n); + if (!h[2]) return 0; + if (!n[3]) return threebyte_strstr((void *)h, (void *)n); + if (!h[3]) return 0; + if (!n[4]) return fourbyte_strstr((void *)h, (void *)n); + + return twoway_strstr((void *)h, (void *)n); +} diff --git a/system/lib/libcextra.symbols b/system/lib/libcextra.symbols index 14b6906e..bc69bb65 100644 --- a/system/lib/libcextra.symbols +++ b/system/lib/libcextra.symbols @@ -123,6 +123,7 @@ T strnlen T strsep T strspn + T strstr T strverscmp T strxfrm W strxfrm_l diff --git a/tools/system_libs.py b/tools/system_libs.py index 2564460a..5a30547f 100644 --- a/tools/system_libs.py +++ b/tools/system_libs.py @@ -251,6 +251,7 @@ def calculate(temp_files, in_temp, stdout, stderr): 'strnlen.c', 'strsep.c', 'strspn.c', + 'strstr.c', 'strverscmp.c', 'wcpcpy.c', 'wcpncpy.c', |