From 39b79783297bee79db9853b63d199e120a009a8f Mon Sep 17 00:00:00 2001 From: Roberto Ierusalimschy Date: Fri, 23 Feb 2001 14:17:25 -0300 Subject: first (big) step to support wide chars --- lstrlib.c | 254 +++++++++++++++++++++++++++++++------------------------------- 1 file changed, 127 insertions(+), 127 deletions(-) (limited to 'lstrlib.c') diff --git a/lstrlib.c b/lstrlib.c index 5fb4de3e..10b57ea8 100644 --- a/lstrlib.c +++ b/lstrlib.c @@ -1,5 +1,5 @@ /* -** $Id: lstrlib.c,v 1.63 2001/02/22 17:15:18 roberto Exp roberto $ +** $Id: lstrlib.c,v 1.64 2001/02/22 18:59:59 roberto Exp roberto $ ** Standard library for string operations and pattern-matching ** See Copyright Notice in lua.h */ @@ -36,14 +36,14 @@ static sint32 posrelat (sint32 pos, size_t len) { static int str_sub (lua_State *L) { size_t l; - const char *s = luaL_check_lstr(L, 1, &l); + const l_char *s = luaL_check_lstr(L, 1, &l); sint32 start = posrelat(luaL_check_long(L, 2), l); sint32 end = posrelat(luaL_opt_long(L, 3, -1), l); if (start < 1) start = 1; if (end > (sint32)l) end = l; if (start <= end) lua_pushlstring(L, s+start-1, end-start+1); - else lua_pushliteral(L, ""); + else lua_pushliteral(L, l_s("")); return 1; } @@ -52,7 +52,7 @@ static int str_lower (lua_State *L) { size_t l; size_t i; luaL_Buffer b; - const char *s = luaL_check_lstr(L, 1, &l); + const l_char *s = luaL_check_lstr(L, 1, &l); luaL_buffinit(L, &b); for (i=0; i 0) @@ -88,9 +88,9 @@ static int str_rep (lua_State *L) { static int str_byte (lua_State *L) { size_t l; - const char *s = luaL_check_lstr(L, 1, &l); + const l_char *s = luaL_check_lstr(L, 1, &l); sint32 pos = posrelat(luaL_opt_long(L, 2, 1), l); - luaL_arg_check(L, 0= ms->level || ms->capture[l].len == CAP_UNFINISHED) - lua_error(ms->L, "invalid capture index"); + lua_error(ms->L, l_s("invalid capture index")); return l; } @@ -154,22 +154,22 @@ static int capture_to_close (MatchState *ms) { int level = ms->level; for (level--; level>=0; level--) if (ms->capture[level].len == CAP_UNFINISHED) return level; - lua_error(ms->L, "invalid pattern capture"); + lua_error(ms->L, l_s("invalid pattern capture")); return 0; /* to avoid warnings */ } -static const char *luaI_classend (MatchState *ms, const char *p) { +static const l_char *luaI_classend (MatchState *ms, const l_char *p) { switch (*p++) { case ESC: - if (*p == '\0') lua_error(ms->L, "malformed pattern (ends with `%')"); + if (*p == l_c('\0')) lua_error(ms->L, l_s("malformed pattern (ends with `%')")); return p+1; - case '[': - if (*p == '^') p++; + case l_c('['): + if (*p == l_c('^')) p++; do { /* look for a `]' */ - if (*p == '\0') lua_error(ms->L, "malformed pattern (missing `]')"); - if (*(p++) == ESC && *p != '\0') p++; /* skip escapes (e.g. `%]') */ - } while (*p != ']'); + if (*p == l_c('\0')) lua_error(ms->L, l_s("malformed pattern (missing `]')")); + if (*(p++) == ESC && *p != l_c('\0')) p++; /* skip escapes (e.g. `%]') */ + } while (*p != l_c(']')); return p+1; default: return p; @@ -177,28 +177,28 @@ static const char *luaI_classend (MatchState *ms, const char *p) { } -static int match_class (char c, char cl) { +static int match_class (l_char c, l_char cl) { int res; switch (tolower(uchar(cl))) { - case 'a' : res = isalpha(uchar(c)); break; - case 'c' : res = iscntrl(uchar(c)); break; - case 'd' : res = isdigit(uchar(c)); break; - case 'l' : res = islower(uchar(c)); break; - case 'p' : res = ispunct(uchar(c)); break; - case 's' : res = isspace(uchar(c)); break; - case 'u' : res = isupper(uchar(c)); break; - case 'w' : res = isalnum(uchar(c)); break; - case 'x' : res = isxdigit(uchar(c)); break; - case 'z' : res = (c == '\0'); break; + case l_c('a') : res = isalpha(uchar(c)); break; + case l_c('c') : res = iscntrl(uchar(c)); break; + case l_c('d') : res = isdigit(uchar(c)); break; + case l_c('l') : res = islower(uchar(c)); break; + case l_c('p') : res = ispunct(uchar(c)); break; + case l_c('s') : res = isspace(uchar(c)); break; + case l_c('u') : res = isupper(uchar(c)); break; + case l_c('w') : res = isalnum(uchar(c)); break; + case l_c('x') : res = isxdigit(uchar(c)); break; + case l_c('z') : res = (c == l_c('\0')); break; default: return (cl == c); } return (islower(uchar(cl)) ? res : !res); } -static int matchbracketclass (char c, const char *p, const char *endclass) { +static int matchbracketclass (l_char c, const l_char *p, const l_char *endclass) { int sig = 1; - if (*(p+1) == '^') { + if (*(p+1) == l_c('^')) { sig = 0; p++; /* skip the `^' */ } @@ -208,7 +208,7 @@ static int matchbracketclass (char c, const char *p, const char *endclass) { if (match_class(c, *p)) return sig; } - else if ((*(p+1) == '-') && (p+2 < endclass)) { + else if ((*(p+1) == l_c('-')) && (p+2 < endclass)) { p+=2; if (uchar(*(p-2)) <= uchar(c) && uchar(c) <= uchar(*p)) return sig; @@ -219,13 +219,13 @@ static int matchbracketclass (char c, const char *p, const char *endclass) { } -static int luaI_singlematch (char c, const char *p, const char *ep) { +static int luaI_singlematch (l_char c, const l_char *p, const l_char *ep) { switch (*p) { - case '.': /* matches any char */ + case l_c('.'): /* matches any char */ return 1; case ESC: return match_class(c, *(p+1)); - case '[': + case l_c('['): return matchbracketclass(c, p, ep-1); default: return (*p == c); @@ -233,12 +233,12 @@ static int luaI_singlematch (char c, const char *p, const char *ep) { } -static const char *match (MatchState *ms, const char *s, const char *p); +static const l_char *match (MatchState *ms, const l_char *s, const l_char *p); -static const char *matchbalance (MatchState *ms, const char *s, const char *p) { +static const l_char *matchbalance (MatchState *ms, const l_char *s, const l_char *p) { if (*p == 0 || *(p+1) == 0) - lua_error(ms->L, "unbalanced pattern"); + lua_error(ms->L, l_s("unbalanced pattern")); if (*s != *p) return NULL; else { int b = *p; @@ -255,14 +255,14 @@ static const char *matchbalance (MatchState *ms, const char *s, const char *p) { } -static const char *max_expand (MatchState *ms, const char *s, const char *p, - const char *ep) { +static const l_char *max_expand (MatchState *ms, const l_char *s, const l_char *p, + const l_char *ep) { sint32 i = 0; /* counts maximum expand for item */ while ((s+i)src_end && luaI_singlematch(*(s+i), p, ep)) i++; /* keeps trying to match with the maximum repetitions */ while (i>=0) { - const char *res = match(ms, (s+i), ep+1); + const l_char *res = match(ms, (s+i), ep+1); if (res) return res; i--; /* else didn't match; reduce 1 repetition to try again */ } @@ -270,10 +270,10 @@ static const char *max_expand (MatchState *ms, const char *s, const char *p, } -static const char *min_expand (MatchState *ms, const char *s, const char *p, - const char *ep) { +static const l_char *min_expand (MatchState *ms, const l_char *s, const l_char *p, + const l_char *ep) { for (;;) { - const char *res = match(ms, s, ep+1); + const l_char *res = match(ms, s, ep+1); if (res != NULL) return res; else if (ssrc_end && luaI_singlematch(*s, p, ep)) @@ -283,11 +283,11 @@ static const char *min_expand (MatchState *ms, const char *s, const char *p, } -static const char *start_capture (MatchState *ms, const char *s, - const char *p, int what) { - const char *res; +static const l_char *start_capture (MatchState *ms, const l_char *s, + const l_char *p, int what) { + const l_char *res; int level = ms->level; - if (level >= MAX_CAPTURES) lua_error(ms->L, "too many captures"); + if (level >= MAX_CAPTURES) lua_error(ms->L, l_s("too many captures")); ms->capture[level].init = s; ms->capture[level].len = what; ms->level = level+1; @@ -297,9 +297,9 @@ static const char *start_capture (MatchState *ms, const char *s, } -static const char *end_capture (MatchState *ms, const char *s, const char *p) { +static const l_char *end_capture (MatchState *ms, const l_char *s, const l_char *p) { int l = capture_to_close(ms); - const char *res; + const l_char *res; ms->capture[l].len = s - ms->capture[l].init; /* close capture */ if ((res = match(ms, s, p)) == NULL) /* match failed? */ ms->capture[l].len = CAP_UNFINISHED; /* undo capture */ @@ -307,7 +307,7 @@ static const char *end_capture (MatchState *ms, const char *s, const char *p) { } -static const char *match_capture (MatchState *ms, const char *s, int level) { +static const l_char *match_capture (MatchState *ms, const l_char *s, int level) { int l = check_capture(ms, level); size_t len = ms->capture[l].len; if ((size_t)(ms->src_end-s) >= len && @@ -317,15 +317,15 @@ static const char *match_capture (MatchState *ms, const char *s, int level) { } -static const char *match (MatchState *ms, const char *s, const char *p) { +static const l_char *match (MatchState *ms, const l_char *s, const l_char *p) { init: /* using goto's to optimize tail recursion */ switch (*p) { - case '(': /* start capture */ - if (*(p+1) == ')') /* position capture? */ + case l_c('('): /* start capture */ + if (*(p+1) == l_c(')')) /* position capture? */ return start_capture(ms, s, p+2, CAP_POSITION); else return start_capture(ms, s, p+1, CAP_UNFINISHED); - case ')': /* end capture */ + case l_c(')'): /* end capture */ return end_capture(ms, s, p+1); case ESC: /* may be %[0-9] or %b */ if (isdigit(uchar(*(p+1)))) { /* capture? */ @@ -333,33 +333,33 @@ static const char *match (MatchState *ms, const char *s, const char *p) { if (s == NULL) return NULL; p+=2; goto init; /* else return match(ms, s, p+2) */ } - else if (*(p+1) == 'b') { /* balanced string? */ + else if (*(p+1) == l_c('b')) { /* balanced string? */ s = matchbalance(ms, s, p+2); if (s == NULL) return NULL; p+=4; goto init; /* else return match(ms, s, p+4); */ } else goto dflt; /* case default */ - case '\0': /* end of pattern */ + case l_c('\0'): /* end of pattern */ return s; /* match succeeded */ - case '$': - if (*(p+1) == '\0') /* is the `$' the last char in pattern? */ + case l_c('$'): + if (*(p+1) == l_c('\0')) /* is the `$' the last char in pattern? */ return (s == ms->src_end) ? s : NULL; /* check end of string */ else goto dflt; default: dflt: { /* it is a pattern item */ - const char *ep = luaI_classend(ms, p); /* points to what is next */ + const l_char *ep = luaI_classend(ms, p); /* points to what is next */ int m = ssrc_end && luaI_singlematch(*s, p, ep); switch (*ep) { - case '?': { /* optional */ - const char *res; + case l_c('?'): { /* optional */ + const l_char *res; if (m && ((res=match(ms, s+1, ep+1)) != NULL)) return res; p=ep+1; goto init; /* else return match(ms, s, ep+1); */ } - case '*': /* 0 or more repetitions */ + case l_c('*'): /* 0 or more repetitions */ return max_expand(ms, s, p, ep); - case '+': /* 1 or more repetitions */ + case l_c('+'): /* 1 or more repetitions */ return (m ? max_expand(ms, s+1, p, ep) : NULL); - case '-': /* 0 or more repetitions (minimum) */ + case l_c('-'): /* 0 or more repetitions (minimum) */ return min_expand(ms, s, p, ep); default: if (!m) return NULL; @@ -371,15 +371,15 @@ static const char *match (MatchState *ms, const char *s, const char *p) { -static const char *lmemfind (const char *s1, size_t l1, - const char *s2, size_t l2) { +static const l_char *lmemfind (const l_char *s1, size_t l1, + const l_char *s2, size_t l2) { if (l2 == 0) return s1; /* empty strings are everywhere */ else if (l2 > l1) return NULL; /* avoids a negative `l1' */ else { - const char *init; /* to search for a `*s2' inside `s1' */ + const l_char *init; /* to search for a `*s2' inside `s1' */ l2--; /* 1st char will be checked by `memchr' */ l1 = l1-l2; /* `s2' cannot be found after that */ - while (l1 > 0 && (init = (const char *)memchr(s1, *s2, l1)) != NULL) { + while (l1 > 0 && (init = (const l_char *)memchr(s1, *s2, l1)) != NULL) { init++; /* 1st char is already checked */ if (memcmp(init, s2+1, l2) == 0) return init-1; @@ -395,7 +395,7 @@ static const char *lmemfind (const char *s1, size_t l1, static void push_onecapture (MatchState *ms, int i) { int l = ms->capture[i].len; - if (l == CAP_UNFINISHED) lua_error(ms->L, "unfinished capture"); + if (l == CAP_UNFINISHED) lua_error(ms->L, l_s("unfinished capture")); if (l == CAP_POSITION) lua_pushnumber(ms->L, ms->capture[i].init - ms->src_init + 1); else @@ -405,7 +405,7 @@ static void push_onecapture (MatchState *ms, int i) { static int push_captures (MatchState *ms) { int i; - luaL_checkstack(ms->L, ms->level, "too many captures"); + luaL_checkstack(ms->L, ms->level, l_s("too many captures")); for (i=0; ilevel; i++) push_onecapture(ms, i); return ms->level; /* number of strings pushed */ @@ -414,13 +414,13 @@ static int push_captures (MatchState *ms) { static int str_find (lua_State *L) { size_t l1, l2; - const char *s = luaL_check_lstr(L, 1, &l1); - const char *p = luaL_check_lstr(L, 2, &l2); + const l_char *s = luaL_check_lstr(L, 1, &l1); + const l_char *p = luaL_check_lstr(L, 2, &l2); sint32 init = posrelat(luaL_opt_long(L, 3, 1), l1) - 1; - luaL_arg_check(L, 0 <= init && (size_t)init <= l1, 3, "out of range"); + luaL_arg_check(L, 0 <= init && (size_t)init <= l1, 3, l_s("out of range")); if (lua_gettop(L) > 3 || /* extra argument? */ strpbrk(p, SPECIALS) == NULL) { /* or no special characters? */ - const char *s2 = lmemfind(s+init, l1-init, p, l2); + const l_char *s2 = lmemfind(s+init, l1-init, p, l2); if (s2) { lua_pushnumber(L, s2-s+1); lua_pushnumber(L, s2-s+l2); @@ -429,13 +429,13 @@ static int str_find (lua_State *L) { } else { MatchState ms; - int anchor = (*p == '^') ? (p++, 1) : 0; - const char *s1=s+init; + int anchor = (*p == l_c('^')) ? (p++, 1) : 0; + const l_char *s1=s+init; ms.L = L; ms.src_init = s; ms.src_end = s+l1; do { - const char *res; + const l_char *res; ms.level = 0; if ((res=match(&ms, s1, p)) != NULL) { lua_pushnumber(L, s1-s+1); /* start */ @@ -452,7 +452,7 @@ static int str_find (lua_State *L) { static void add_s (MatchState *ms, luaL_Buffer *b) { lua_State *L = ms->L; if (lua_isstring(L, 3)) { - const char *news = lua_tostring(L, 3); + const l_char *news = lua_tostring(L, 3); size_t l = lua_strlen(L, 3); size_t i; for (i=0; i= 3 && (lua_isstring(L, 3) || lua_isfunction(L, 3)), - 3, "string or function expected"); + 3, l_s("string or function expected")); luaL_buffinit(L, &b); ms.L = L; ms.src_init = src; ms.src_end = src+srcl; while (n < max_s) { - const char *e; + const l_char *e; ms.level = 0; e = match(&ms, src, p); if (e) { @@ -531,40 +531,40 @@ static int str_gsub (lua_State *L) { static void luaI_addquoted (lua_State *L, luaL_Buffer *b, int arg) { size_t l; - const char *s = luaL_check_lstr(L, arg, &l); - luaL_putchar(b, '"'); + const l_char *s = luaL_check_lstr(L, arg, &l); + luaL_putchar(b, l_c('"')); while (l--) { switch (*s) { - case '"': case '\\': case '\n': - luaL_putchar(b, '\\'); + case l_c('"'): case l_c('\\'): case l_c('\n'): + luaL_putchar(b, l_c('\\')); luaL_putchar(b, *s); break; - case '\0': luaL_addlstring(b, "\\000", 4); break; + case l_c('\0'): luaL_addlstring(b, l_s("\\000"), 4); break; default: luaL_putchar(b, *s); } s++; } - luaL_putchar(b, '"'); + luaL_putchar(b, l_c('"')); } -static const char *scanformat (lua_State *L, const char *strfrmt, char *form, +static const l_char *scanformat (lua_State *L, const l_char *strfrmt, l_char *form, int *hasprecision) { - const char *p = strfrmt; - while (strchr("-+ #0", *p)) p++; /* skip flags */ + const l_char *p = strfrmt; + while (strchr(l_s("-+ #0"), *p)) p++; /* skip flags */ if (isdigit(uchar(*p))) p++; /* skip width */ if (isdigit(uchar(*p))) p++; /* (2 digits at most) */ - if (*p == '.') { + if (*p == l_c('.')) { p++; *hasprecision = 1; if (isdigit(uchar(*p))) p++; /* skip precision */ if (isdigit(uchar(*p))) p++; /* (2 digits at most) */ } if (isdigit(uchar(*p))) - lua_error(L, "invalid format (width or precision too long)"); + lua_error(L, l_s("invalid format (width or precision too long)")); if (p-strfrmt+2 > MAX_FORMAT) /* +2 to include `%' and the specifier */ - lua_error(L, "invalid format (too long)"); - form[0] = '%'; + lua_error(L, l_s("invalid format (too long)")); + form[0] = l_c('%'); strncpy(form+1, strfrmt, p-strfrmt+1); form[p-strfrmt+2] = 0; return p; @@ -573,38 +573,38 @@ static const char *scanformat (lua_State *L, const char *strfrmt, char *form, static int str_format (lua_State *L) { int arg = 1; - const char *strfrmt = luaL_check_string(L, arg); + const l_char *strfrmt = luaL_check_string(L, arg); luaL_Buffer b; luaL_buffinit(L, &b); while (*strfrmt) { - if (*strfrmt != '%') + if (*strfrmt != l_c('%')) luaL_putchar(&b, *strfrmt++); - else if (*++strfrmt == '%') + else if (*++strfrmt == l_c('%')) luaL_putchar(&b, *strfrmt++); /* %% */ else { /* format item */ - char form[MAX_FORMAT]; /* to store the format (`%...') */ - char buff[MAX_ITEM]; /* to store the formatted item */ + l_char form[MAX_FORMAT]; /* to store the format (`%...') */ + l_char buff[MAX_ITEM]; /* to store the formatted item */ int hasprecision = 0; - if (isdigit(uchar(*strfrmt)) && *(strfrmt+1) == '$') - lua_error(L, "obsolete `format' option (d$)"); + if (isdigit(uchar(*strfrmt)) && *(strfrmt+1) == l_c('$')) + lua_error(L, l_s("obsolete `format' option (d$)")); arg++; strfrmt = scanformat(L, strfrmt, form, &hasprecision); switch (*strfrmt++) { - case 'c': case 'd': case 'i': + case l_c('c'): case l_c('d'): case l_c('i'): sprintf(buff, form, luaL_check_int(L, arg)); break; - case 'o': case 'u': case 'x': case 'X': + case l_c('o'): case l_c('u'): case l_c('x'): case l_c('X'): sprintf(buff, form, (unsigned int)luaL_check_number(L, arg)); break; - case 'e': case 'E': case 'f': case 'g': case 'G': + case l_c('e'): case l_c('E'): case l_c('f'): case l_c('g'): case l_c('G'): sprintf(buff, form, luaL_check_number(L, arg)); break; - case 'q': + case l_c('q'): luaI_addquoted(L, &b, arg); continue; /* skip the `addsize' at the end */ - case 's': { + case l_c('s'): { size_t l; - const char *s = luaL_check_lstr(L, arg, &l); + const l_char *s = luaL_check_lstr(L, arg, &l); if (!hasprecision && l >= 100) { /* no precision and string is too long to be formatted; keep original string */ @@ -618,7 +618,7 @@ static int str_format (lua_State *L) { } } default: /* also treat cases `pnLlh' */ - lua_error(L, "invalid option in `format'"); + lua_error(L, l_s("invalid option in `format'")); } luaL_addlstring(&b, buff, strlen(buff)); } @@ -629,16 +629,16 @@ static int str_format (lua_State *L) { static const luaL_reg strlib[] = { -{"strlen", str_len}, -{"strsub", str_sub}, -{"strlower", str_lower}, -{"strupper", str_upper}, -{"strchar", str_char}, -{"strrep", str_rep}, -{"strbyte", str_byte}, -{"format", str_format}, -{"strfind", str_find}, -{"gsub", str_gsub} +{l_s("strlen"), str_len}, +{l_s("strsub"), str_sub}, +{l_s("strlower"), str_lower}, +{l_s("strupper"), str_upper}, +{l_s("strchar"), str_char}, +{l_s("strrep"), str_rep}, +{l_s("strbyte"), str_byte}, +{l_s("format"), str_format}, +{l_s("strfind"), str_find}, +{l_s("gsub"), str_gsub} }; -- cgit v1.2.3-55-g6feb