diff options
author | Ulrich Drepper <drepper@redhat.com> | 2004-03-14 21:12:06 +0000 |
---|---|---|
committer | Ulrich Drepper <drepper@redhat.com> | 2004-03-14 21:12:06 +0000 |
commit | ccadf7b5346a3e21c692dfcbfcf38a63433bc36a (patch) | |
tree | a1e826af3345eee581925387513b1b6c76629c31 /string/strcoll_l.c | |
parent | 3bc9b83f49805969b7d139b52ef6b003e5136354 (diff) | |
download | glibc-ccadf7b5346a3e21c692dfcbfcf38a63433bc36a.tar.gz glibc-ccadf7b5346a3e21c692dfcbfcf38a63433bc36a.tar.xz glibc-ccadf7b5346a3e21c692dfcbfcf38a63433bc36a.zip |
Update.
2004-03-14 Ulrich Drepper <drepper@redhat.com> Make the non-_l functions wrappers around the _l functions. * include/monetary.h: Declare __vstrmon_l. * include/string.h: Add libc_hidden_proto for __strcoll_l and __strxfrm_l. * include/time.h: Define ptime_locale_status. Declare __strptime_internal. * include/wchar.h: Add libc_hidden_proto for __wcscoll_l and __wcsxfrm_l. * stdlib/strfmon.c: Move the code to strfmon_l.c. Add little wrapper around __vstrfmon_l. * stdlib/strfmon_l.c: Add real implementation. Split into new function __vstrfmon_l to allow calling it from strfmon. * stdlib/strtod.c: Move real code to strtod_l.c and add wrapper. * stdlib/strtod_l.c: Add real implementation. * stdlib/strtof.c: Adjust to changed strtod.c. * stdlib/strtof_l.c: Include strtod_l.c now. * stdlib/strtold.c: New file. * stdlib/strtold_l.c: Removed. * string/strcoll.c: Move real code to strcoll_l.c: Add wrapper. * string/strcoll_l.c: Add real implementation. * string/strxfrm.c: Move real code to strxfrm_l.c: Add wrapper. * string/strxfrm_l.c: Add real implementation. * sysdeps/generic/strtol.c: Move real implementation to strtol_l.c. Add wrappers. * sysdeps/generic/strtol_l.c: Add real implementation. * sysdeps/generic/strtold.c: Removed. * sysdeps/generic/strtold_l.c: New file. * sysdeps/generic/strtoll_l.c: Include strtol_l.c now. Adjust #defines. * sysdeps/generic/strtoul_l.c: Likewise. * sysdeps/generic/strtoull_l.c: Likewise. * sysdeps/generic/wcstol_l.c: Likewise. * sysdeps/generic/wcstoll_l.c: Likewise. * sysdeps/generic/wcstoul_l.c: Likewise. * sysdeps/generic/wcstoull_l.c: Likewise. * sysdeps/ieee754/ldbl-128/strtold.c: Removed. * sysdeps/ieee754/ldbl-128/strtold_l.c: New file. * sysdeps/ieee754/ldbl-96/strtold.c: Removed. * sysdeps/ieee754/ldbl-96/strtold_l.c: New file. * sysdeps/m68k/strtold.c: Removed. * sysdeps/m68k/strtold_l.c: New file. * time/strftime.c: Move real code to strftime_l.c. Add wrapper. * time/strftime_l.c: Add real implementation. * time/strptime.c: Move real code to strptime_l.c. Add wrapper. * time/strptime_l.c: Add real implementation. * time/wcsftime.c: Simplify since only wrappers are defined in strftime.c. * time/wcsftime_l.c: Include strftime_l.c. * wcsmbs/wcscoll.c: Simplify since the file is not used by wcscoll_l.c anymore. * wcsmbs/wcscoll_l.c: Include strcoll_l.c. * wcsmbs/wcsxfrm.c: Simplify since the file is not used by wcsxfrm_l.c anymore. * wcsmbs/wcsxfrm_l.c: Include strxfrm_l.c. * wcsmbs/wcstod.c: Prepare to include new strtod.c. * wcsmbs/wcstod_l.c: Include strtod_l.c. * wcsmbs/wcstof.c: Prepare to include new strtof.c. * wcsmbs/wcstof_l.c: Include strtof_l.c. * wcsmbs/wcstold.c: Prepare to include new strtold.c. * wcsmbs/wcstold_l.c: Include strtold_l.c. * locale/uselocale.c: Use _NL_CURRENT_LOCALE instead of __libc_tsd_get. * sysdeps/generic/strcasecmp.c: Optimize a bit. It's better to get a reference to the current locale and then use the _l functions. * sysdeps/generic/strncase.c: Likewise.
Diffstat (limited to 'string/strcoll_l.c')
-rw-r--r-- | string/strcoll_l.c | 514 |
1 files changed, 511 insertions, 3 deletions
diff --git a/string/strcoll_l.c b/string/strcoll_l.c index 6611589ed0..c46921dcc9 100644 --- a/string/strcoll_l.c +++ b/string/strcoll_l.c @@ -1,4 +1,4 @@ -/* Copyright (C) 1995,96,97,2002 Free Software Foundation, Inc. +/* Copyright (C) 1995,96,97,2002, 2004 Free Software Foundation, Inc. This file is part of the GNU C Library. Written by Ulrich Drepper <drepper@gnu.org>, 1995. @@ -17,7 +17,515 @@ Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. */ -#define USE_IN_EXTENDED_LOCALE_MODEL 1 -#include <strcoll.c> +#include <assert.h> +#include <langinfo.h> +#include <locale.h> +#include <stddef.h> +#include <stdint.h> +#include <stdlib.h> +#include <string.h> + +#ifndef STRING_TYPE +# define STRING_TYPE char +# define USTRING_TYPE unsigned char +# define STRCOLL __strcoll_l +# define STRCMP strcmp +# define STRLEN strlen +# define WEIGHT_H "../locale/weight.h" +# define SUFFIX MB +# define L(arg) arg +#endif + +#define CONCAT(a,b) CONCAT1(a,b) +#define CONCAT1(a,b) a##b + +#include "../locale/localeinfo.h" + +int +STRCOLL (s1, s2, l) + const STRING_TYPE *s1; + const STRING_TYPE *s2; + __locale_t l; +{ + struct locale_data *current = l->__locales[LC_COLLATE]; + uint_fast32_t nrules = current->values[_NL_ITEM_INDEX (_NL_COLLATE_NRULES)].word; + /* We don't assign the following values right away since it might be + unnecessary in case there are no rules. */ + const unsigned char *rulesets; + const int32_t *table; + const USTRING_TYPE *weights; + const USTRING_TYPE *extra; + const int32_t *indirect; + uint_fast32_t pass; + int result = 0; + const USTRING_TYPE *us1; + const USTRING_TYPE *us2; + size_t s1len; + size_t s2len; + int32_t *idx1arr; + int32_t *idx2arr; + unsigned char *rule1arr; + unsigned char *rule2arr; + size_t idx1max; + size_t idx2max; + size_t idx1cnt; + size_t idx2cnt; + size_t idx1now; + size_t idx2now; + size_t backw1_stop; + size_t backw2_stop; + size_t backw1; + size_t backw2; + int val1; + int val2; + int position; + int seq1len; + int seq2len; + int use_malloc; + +#include WEIGHT_H + + if (nrules == 0) + return STRCMP (s1, s2); + + rulesets = (const unsigned char *) + current->values[_NL_ITEM_INDEX (_NL_COLLATE_RULESETS)].string; + table = (const int32_t *) + current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_TABLE,SUFFIX))].string; + weights = (const USTRING_TYPE *) + current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_WEIGHT,SUFFIX))].string; + extra = (const USTRING_TYPE *) + current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_EXTRA,SUFFIX))].string; + indirect = (const int32_t *) + current->values[_NL_ITEM_INDEX (CONCAT(_NL_COLLATE_INDIRECT,SUFFIX))].string; + use_malloc = 0; + + assert (((uintptr_t) table) % __alignof__ (table[0]) == 0); + assert (((uintptr_t) weights) % __alignof__ (weights[0]) == 0); + assert (((uintptr_t) extra) % __alignof__ (extra[0]) == 0); + assert (((uintptr_t) indirect) % __alignof__ (indirect[0]) == 0); + + /* We need this a few times. */ + s1len = STRLEN (s1); + s2len = STRLEN (s2); + + /* Catch empty strings. */ + if (__builtin_expect (s1len == 0, 0) || __builtin_expect (s2len == 0, 0)) + return (s1len != 0) - (s2len != 0); + + /* We need the elements of the strings as unsigned values since they + are used as indeces. */ + us1 = (const USTRING_TYPE *) s1; + us2 = (const USTRING_TYPE *) s2; + + /* Perform the first pass over the string and while doing this find + and store the weights for each character. Since we want this to + be as fast as possible we are using `alloca' to store the temporary + values. But since there is no limit on the length of the string + we have to use `malloc' if the string is too long. We should be + very conservative here. + + Please note that the localedef programs makes sure that `position' + is not used at the first level. */ + if (! __libc_use_alloca (s1len + s2len)) + { + idx1arr = (int32_t *) malloc ((s1len + s2len) * (sizeof (int32_t) + 1)); + idx2arr = &idx1arr[s1len]; + rule1arr = (unsigned char *) &idx2arr[s2len]; + rule2arr = &rule1arr[s1len]; + + if (idx1arr == NULL) + /* No memory. Well, go with the stack then. + + XXX Once this implementation is stable we will handle this + differently. Instead of precomputing the indeces we will + do this in time. This means, though, that this happens for + every pass again. */ + goto try_stack; + use_malloc = 1; + } + else + { + try_stack: + idx1arr = (int32_t *) alloca (s1len * sizeof (int32_t)); + idx2arr = (int32_t *) alloca (s2len * sizeof (int32_t)); + rule1arr = (unsigned char *) alloca (s1len); + rule2arr = (unsigned char *) alloca (s2len); + } + + idx1cnt = 0; + idx2cnt = 0; + idx1max = 0; + idx2max = 0; + idx1now = 0; + idx2now = 0; + backw1_stop = ~0ul; + backw2_stop = ~0ul; + backw1 = ~0ul; + backw2 = ~0ul; + seq1len = 0; + seq2len = 0; + position = rulesets[0] & sort_position; + while (1) + { + val1 = 0; + val2 = 0; + + /* Get the next non-IGNOREd element for string `s1'. */ + if (seq1len == 0) + do + { + ++val1; + + if (backw1_stop != ~0ul) + { + /* The is something pushed. */ + if (backw1 == backw1_stop) + { + /* The last pushed character was handled. Continue + with forward characters. */ + if (idx1cnt < idx1max) + idx1now = idx1cnt; + else + /* Nothing anymore. The backward sequence ended with + the last sequence in the string. Note that seq1len + is still zero. */ + break; + } + else + idx1now = --backw1; + } + else + { + backw1_stop = idx1max; + + while (*us1 != L('\0')) + { + int32_t tmp = findidx (&us1); + rule1arr[idx1max] = tmp >> 24; + idx1arr[idx1max] = tmp & 0xffffff; + idx1cnt = idx1max++; + + if ((rulesets[rule1arr[idx1cnt] * nrules] + & sort_backward) == 0) + /* No more backward characters to push. */ + break; + ++idx1cnt; + } + + if (backw1_stop >= idx1cnt) + { + /* No sequence at all or just one. */ + if (idx1cnt == idx1max || backw1_stop > idx1cnt) + /* Note that seq1len is still zero. */ + break; + + backw1_stop = ~0ul; + idx1now = idx1cnt; + } + else + /* We pushed backward sequences. */ + idx1now = backw1 = idx1cnt - 1; + } + } + while ((seq1len = weights[idx1arr[idx1now]++]) == 0); + + /* And the same for string `s2'. */ + if (seq2len == 0) + do + { + ++val2; + + if (backw2_stop != ~0ul) + { + /* The is something pushed. */ + if (backw2 == backw2_stop) + { + /* The last pushed character was handled. Continue + with forward characters. */ + if (idx2cnt < idx2max) + idx2now = idx2cnt; + else + /* Nothing anymore. The backward sequence ended with + the last sequence in the string. Note that seq2len + is still zero. */ + break; + } + else + idx2now = --backw2; + } + else + { + backw2_stop = idx2max; + + while (*us2 != L('\0')) + { + int32_t tmp = findidx (&us2); + rule2arr[idx2max] = tmp >> 24; + idx2arr[idx2max] = tmp & 0xffffff; + idx2cnt = idx2max++; + + if ((rulesets[rule2arr[idx2cnt] * nrules] + & sort_backward) == 0) + /* No more backward characters to push. */ + break; + ++idx2cnt; + } + + if (backw2_stop >= idx2cnt) + { + /* No sequence at all or just one. */ + if (idx2cnt == idx2max || backw2_stop > idx2cnt) + /* Note that seq1len is still zero. */ + break; + + backw2_stop = ~0ul; + idx2now = idx2cnt; + } + else + /* We pushed backward sequences. */ + idx2now = backw2 = idx2cnt - 1; + } + } + while ((seq2len = weights[idx2arr[idx2now]++]) == 0); + + /* See whether any or both strings are empty. */ + if (seq1len == 0 || seq2len == 0) + { + if (seq1len == seq2len) + /* Both ended. So far so good, both strings are equal at the + first level. */ + break; + + /* This means one string is shorter than the other. Find out + which one and return an appropriate value. */ + result = seq1len == 0 ? -1 : 1; + goto free_and_return; + } + + /* Test for position if necessary. */ + if (position && val1 != val2) + { + result = val1 - val2; + goto free_and_return; + } + + /* Compare the two sequences. */ + do + { + if (weights[idx1arr[idx1now]] != weights[idx2arr[idx2now]]) + { + /* The sequences differ. */ + result = weights[idx1arr[idx1now]] - weights[idx2arr[idx2now]]; + goto free_and_return; + } + + /* Increment the offsets. */ + ++idx1arr[idx1now]; + ++idx2arr[idx2now]; + + --seq1len; + --seq2len; + } + while (seq1len > 0 && seq2len > 0); + + if (position && seq1len != seq2len) + { + result = seq1len - seq2len; + goto free_and_return; + } + } + + /* Now the remaining passes over the weights. We now use the + indeces we found before. */ + for (pass = 1; pass < nrules; ++pass) + { + /* We assume that if a rule has defined `position' in one section + this is true for all of them. */ + idx1cnt = 0; + idx2cnt = 0; + backw1_stop = ~0ul; + backw2_stop = ~0ul; + backw1 = ~0ul; + backw2 = ~0ul; + position = rulesets[rule1arr[0] * nrules + pass] & sort_position; + + while (1) + { + val1 = 0; + val2 = 0; + + /* Get the next non-IGNOREd element for string `s1'. */ + if (seq1len == 0) + do + { + ++val1; + + if (backw1_stop != ~0ul) + { + /* The is something pushed. */ + if (backw1 == backw1_stop) + { + /* The last pushed character was handled. Continue + with forward characters. */ + if (idx1cnt < idx1max) + idx1now = idx1cnt; + else + { + /* Nothing anymore. The backward sequence + ended with the last sequence in the string. */ + idx1now = ~0ul; + break; + } + } + else + idx1now = --backw1; + } + else + { + backw1_stop = idx1cnt; + + while (idx1cnt < idx1max) + { + if ((rulesets[rule1arr[idx1cnt] * nrules + pass] + & sort_backward) == 0) + /* No more backward characters to push. */ + break; + ++idx1cnt; + } + + if (backw1_stop == idx1cnt) + { + /* No sequence at all or just one. */ + if (idx1cnt == idx1max) + /* Note that seq1len is still zero. */ + break; + + backw1_stop = ~0ul; + idx1now = idx1cnt++; + } + else + /* We pushed backward sequences. */ + idx1now = backw1 = idx1cnt - 1; + } + } + while ((seq1len = weights[idx1arr[idx1now]++]) == 0); + + /* And the same for string `s2'. */ + if (seq2len == 0) + do + { + ++val2; + + if (backw2_stop != ~0ul) + { + /* The is something pushed. */ + if (backw2 == backw2_stop) + { + /* The last pushed character was handled. Continue + with forward characters. */ + if (idx2cnt < idx2max) + idx2now = idx2cnt; + else + { + /* Nothing anymore. The backward sequence + ended with the last sequence in the string. */ + idx2now = ~0ul; + break; + } + } + else + idx2now = --backw2; + } + else + { + backw2_stop = idx2cnt; + + while (idx2cnt < idx2max) + { + if ((rulesets[rule2arr[idx2cnt] * nrules + pass] + & sort_backward) == 0) + /* No more backward characters to push. */ + break; + ++idx2cnt; + } + + if (backw2_stop == idx2cnt) + { + /* No sequence at all or just one. */ + if (idx2cnt == idx2max) + /* Note that seq2len is still zero. */ + break; + + backw2_stop = ~0ul; + idx2now = idx2cnt++; + } + else + /* We pushed backward sequences. */ + idx2now = backw2 = idx2cnt - 1; + } + } + while ((seq2len = weights[idx2arr[idx2now]++]) == 0); + + /* See whether any or both strings are empty. */ + if (seq1len == 0 || seq2len == 0) + { + if (seq1len == seq2len) + /* Both ended. So far so good, both strings are equal + at this level. */ + break; + + /* This means one string is shorter than the other. Find out + which one and return an appropriate value. */ + result = seq1len == 0 ? -1 : 1; + goto free_and_return; + } + + /* Test for position if necessary. */ + if (position && val1 != val2) + { + result = val1 - val2; + goto free_and_return; + } + + /* Compare the two sequences. */ + do + { + if (weights[idx1arr[idx1now]] != weights[idx2arr[idx2now]]) + { + /* The sequences differ. */ + result = (weights[idx1arr[idx1now]] + - weights[idx2arr[idx2now]]); + goto free_and_return; + } + + /* Increment the offsets. */ + ++idx1arr[idx1now]; + ++idx2arr[idx2now]; + + --seq1len; + --seq2len; + } + while (seq1len > 0 && seq2len > 0); + + if (position && seq1len != seq2len) + { + result = seq1len - seq2len; + goto free_and_return; + } + } + } + + /* Free the memory if needed. */ + free_and_return: + if (use_malloc) + free (idx1arr); + + return result; +} +libc_hidden_def (STRCOLL) + +#ifndef WIDE_CHAR_VERSION weak_alias (__strcoll_l, strcoll_l) +#endif |