diff options
Diffstat (limited to 'strings/conf_to_src.c')
-rw-r--r-- | strings/conf_to_src.c | 342 |
1 files changed, 253 insertions, 89 deletions
diff --git a/strings/conf_to_src.c b/strings/conf_to_src.c index 2f88c9ad22e..69dccbca0a7 100644 --- a/strings/conf_to_src.c +++ b/strings/conf_to_src.c @@ -14,129 +14,293 @@ along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -/* can't use -lmysys because this prog is used to create -lstrings */ - #include <ctype.h> #include <stdio.h> #include <stdlib.h> #include <string.h> #include <unistd.h> +#include <fcntl.h> -#define CHARSETS_SUBDIR "sql/share/charsets" -#define CTYPE_TABLE_SIZE 257 -#define TO_LOWER_TABLE_SIZE 256 -#define TO_UPPER_TABLE_SIZE 256 -#define SORT_ORDER_TABLE_SIZE 256 -#define ROW_LEN 16 +#include <my_global.h> +#include <m_ctype.h> +#include <my_xml.h> -void print_arrays_for(char *set); -char *prog; -char buf[1024], *p, *endptr; +#define ROW_LEN 16 +#define ROW16_LEN 8 +#define MAX_BUF 16*1024 -int -main(int argc, char **argv) +static CHARSET_INFO all_charsets[256]; + + +void +print_array(FILE *f, const char *set, const char *name, uchar *a, int n) { - prog = *argv; + int i; - if (argc < 2) { - fprintf(stderr, "usage: %s source-dir [charset [, charset]]\n", prog); - exit(EXIT_FAILURE); + fprintf(f,"uchar %s_%s[] = {\n", name, set); + + for (i=0 ;i<n ; i++) + { + fprintf(f,"0x%02X",a[i]); + fprintf(f, (i+1<n) ? "," :"" ); + fprintf(f, ((i+1) % ROW_LEN == n % ROW_LEN) ? "\n" : "" ); } + fprintf(f,"};\n\n"); +} - --argc; ++argv; /* skip program name */ - if (chdir(*argv) != 0) { - fprintf(stderr, "%s: can't cd to %s\n", prog, *argv); - exit(EXIT_FAILURE); - } - --argc; ++argv; +void +print_array16(FILE *f, const char *set, const char *name, uint16 *a, int n) +{ + int i; - if (chdir(CHARSETS_SUBDIR) != 0) { - fprintf(stderr, "%s: can't cd to %s\n", prog, CHARSETS_SUBDIR); - exit(EXIT_FAILURE); + fprintf(f,"uchar %s_%s[] = {\n", name, set); + + for (i=0 ;i<n ; i++) + { + fprintf(f,"0x%04X",a[i]); + fprintf(f, (i+1<n) ? "," :"" ); + fprintf(f, ((i+1) % ROW16_LEN == n % ROW16_LEN) ? "\n" : "" ); } + fprintf(f,"};\n\n"); +} - while (argc--) - print_arrays_for(*argv++); - exit(EXIT_SUCCESS); +static int get_charset_number(const char *charset_name) +{ + CHARSET_INFO *cs; + for (cs= all_charsets; cs < all_charsets+255; ++cs) + { + if ( cs->name && !strcmp(cs->name, charset_name)) + return cs->number; + } + return 0; } -void -print_array(FILE *f, const char *set, const char *name, int n) +char *mdup(const char *src, uint len) { - int i; - char val[100]; + char *dst=(char*)malloc(len); + memcpy(dst,src,len); + return dst; +} - printf("uchar %s_%s[] = {\n", name, set); +static void simple_cs_copy_data(CHARSET_INFO *to, CHARSET_INFO *from) +{ + to->number= from->number ? from->number : to->number; + to->state|= from->state; + + if (from->csname) + to->csname= strdup(from->csname); + + if (from->name) + to->name= strdup(from->name); + + if (from->ctype) + to->ctype= (uchar*) mdup((char*) from->ctype, MY_CS_CTYPE_TABLE_SIZE); + if (from->to_lower) + to->to_lower= (uchar*) mdup((char*) from->to_lower, MY_CS_TO_LOWER_TABLE_SIZE); + if (from->to_upper) + to->to_upper= (uchar*) mdup((char*) from->to_upper, MY_CS_TO_UPPER_TABLE_SIZE); + if (from->sort_order) + { + to->sort_order= (uchar*) mdup((char*) from->sort_order, MY_CS_SORT_ORDER_TABLE_SIZE); + /* + set_max_sort_char(to); + */ + } + if (from->tab_to_uni) + { + uint sz= MY_CS_TO_UNI_TABLE_SIZE*sizeof(uint16); + to->tab_to_uni= (uint16*) mdup((char*)from->tab_to_uni, sz); + /* + create_fromuni(to); + */ + } +} + +static my_bool simple_cs_is_full(CHARSET_INFO *cs) +{ + return ((cs->csname && cs->tab_to_uni && cs->ctype && cs->to_upper && + cs->to_lower) && + (cs->number && cs->name && cs->sort_order)); +} - p = buf; - *buf = '\0'; - for (i = 0; i < n; ++i) +static int add_collation(CHARSET_INFO *cs) +{ + if (cs->name && (cs->number || (cs->number=get_charset_number(cs->name)))) { - /* get a word from f */ - endptr = p; - for (;;) + if (!(all_charsets[cs->number].state & MY_CS_COMPILED)) { - while (isspace((* (unsigned char*) endptr))) - ++endptr; - if (*endptr && *endptr != '#') /* not comment */ - break; - if ((fgets(buf, sizeof(buf), f)) == NULL) - return; /* XXX: break silently */ - endptr = buf; + simple_cs_copy_data(&all_charsets[cs->number],cs); + } + + cs->number= 0; + cs->name= NULL; + cs->state= 0; + cs->sort_order= NULL; + cs->state= 0; + } + return MY_XML_OK; +} - p = val; - while (!isspace((* (unsigned char*) endptr))) - *p++ = *endptr++; - *p = '\0'; - p = endptr; - - /* write the value out */ - - if (i == 0 || i % ROW_LEN == n % ROW_LEN) - printf(" "); - - printf("%3d", (unsigned char) strtol(val, (char **) NULL, 16)); - - if (i < n - 1) - printf(","); - if ((i+1) % ROW_LEN == n % ROW_LEN) - printf("\n"); +static int my_read_charset_file(const char *filename) +{ + char buf[MAX_BUF]; + int fd; + uint len; + + if ((fd=open(filename,O_RDONLY)) < 0) + { + fprintf(stderr,"Can't open '%s'\n",filename); + return 1; } - - printf("};\n\n"); + + len=read(fd,buf,MAX_BUF); + close(fd); + + if (my_parse_charset_xml(buf,len,add_collation)) + { +#if 0 + printf("ERROR at line %d pos %d '%s'\n", + my_xml_error_lineno(&p)+1, + my_xml_error_pos(&p), + my_xml_error_string(&p)); +#endif + } + + return FALSE; } -void -print_arrays_for(char *set) +void dispcset(FILE *f,CHARSET_INFO *cs) { - FILE *f; + fprintf(f,"{\n"); + fprintf(f," %d,\n",cs->number); + fprintf(f," MY_CS_COMPILED,\n"); + + if (cs->name) + { + fprintf(f," \"%s\",\n",cs->name); + fprintf(f," \"%s\",\n",cs->csname); + fprintf(f," \"\",\n"); + fprintf(f," ctype_%s,\n",cs->name); + fprintf(f," to_lower_%s,\n",cs->name); + fprintf(f," to_upper_%s,\n",cs->name); + fprintf(f," sort_order_%s,\n",cs->name); + fprintf(f," to_uni_%s,\n",cs->name); + fprintf(f," from_uni_%s,\n",cs->name); + } + else + { + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + } + + fprintf(f," %d,\n",cs->strxfrm_multiply); + fprintf(f," my_strnncoll_simple,\n"); + fprintf(f," my_strnxfrm_simple,\n"); + fprintf(f," my_like_range_simple,\n"); + fprintf(f," my_wild_cmp_8bit,\n"); + fprintf(f," %d,\n",cs->mbmaxlen); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," NULL,\n"); + fprintf(f," my_mb_wc_8bit,\n"); + fprintf(f," my_wc_mb_8bit,\n"); + fprintf(f," my_caseup_str_8bit,\n"); + fprintf(f," my_casedn_str_8bit,\n"); + fprintf(f," my_caseup_8bit,\n"); + fprintf(f," my_casedn_8bit,\n"); + fprintf(f," my_tosort_8bit,\n"); + fprintf(f," my_strcasecmp_8bit,\n"); + fprintf(f," my_strncasecmp_8bit,\n"); + fprintf(f," my_hash_caseup_simple,\n"); + fprintf(f," my_hash_sort_simple,\n"); + fprintf(f," 0,\n"); + fprintf(f," my_snprintf_8bit,\n"); + fprintf(f," my_long10_to_str_8bit,\n"); + fprintf(f," my_longlong10_to_str_8bit,\n"); + fprintf(f," my_fill_8bit,\n"); + fprintf(f," my_strntol_8bit,\n"); + fprintf(f," my_strntoul_8bit,\n"); + fprintf(f," my_strntoll_8bit,\n"); + fprintf(f," my_strntoull_8bit,\n"); + fprintf(f," my_strntod_8bit,\n"); + fprintf(f," my_scan_8bit\n"); + fprintf(f,"}\n"); +} - sprintf(buf, "%s.conf", set); - if ((f = fopen(buf, "r")) == NULL) { - fprintf(stderr, "%s: can't read conf file for charset %s\n", prog, set); +int +main(int argc, char **argv __attribute__((unused))) +{ + CHARSET_INFO ncs; + CHARSET_INFO *cs; + char filename[256]; + FILE *f= stdout; + + if (argc < 2) + { + fprintf(stderr, "usage: %s source-dir\n", argv[0]); exit(EXIT_FAILURE); } - - printf("\ -/* The %s character set. Generated automatically by\n\ - * the %s program\n\ - */\n\n", - set, prog); - - /* it would be nice if this used the code in mysys/charset.c, but... */ - print_array(f, set, "ctype", CTYPE_TABLE_SIZE); - print_array(f, set, "to_lower", TO_LOWER_TABLE_SIZE); - print_array(f, set, "to_upper", TO_UPPER_TABLE_SIZE); - print_array(f, set, "sort_order", SORT_ORDER_TABLE_SIZE); - printf("\n"); - - fclose(f); - - return; + + bzero((void*)&ncs,sizeof(ncs)); + bzero((void*)&all_charsets,sizeof(all_charsets)); + + sprintf(filename,"%s/%s",argv[1],"Index.xml"); + my_read_charset_file(filename); + + for (cs=all_charsets; cs < all_charsets+256; cs++) + { + if (cs->number) + { + if ( (!simple_cs_is_full(cs)) && (cs->csname) ) + { + sprintf(filename,"%s/%s.xml",argv[1],cs->csname); + my_read_charset_file(filename); + } + } + } + + for (cs=all_charsets; cs < all_charsets+256; cs++) + { + if (simple_cs_is_full(cs)) + { + fprintf(f,"#ifdef HAVE_CHARSET_%s\n",cs->csname); + print_array(f, cs->name, "ctype", cs->ctype, MY_CS_CTYPE_TABLE_SIZE); + print_array(f, cs->name, "to_lower", cs->to_lower, MY_CS_TO_LOWER_TABLE_SIZE); + print_array(f, cs->name, "to_upper", cs->to_upper, MY_CS_TO_UPPER_TABLE_SIZE); + print_array(f, cs->name, "sort_order", cs->sort_order, MY_CS_SORT_ORDER_TABLE_SIZE); + print_array16(f, cs->name, "to_uni", cs->tab_to_uni, MY_CS_TO_UNI_TABLE_SIZE); + fprintf(f,"#endif\n"); + fprintf(f,"\n"); + } + } + + fprintf(f,"CHARSET_INFO compiled_charsets[] = {\n"); + for (cs=all_charsets; cs < all_charsets+256; cs++) + { + if (simple_cs_is_full(cs)) + { + fprintf(f,"#ifdef HAVE_CHARSET_%s\n",cs->csname); + dispcset(f,cs); + fprintf(f,",\n"); + fprintf(f,"#endif\n"); + } + } + + dispcset(f,&ncs); + fprintf(f,"};\n"); + + return 0; } |