X-Git-Url: http://jsfdemo.indexdata.com/?a=blobdiff_plain;f=dict%2Fdicttest.c;h=6b6ff00636f6216a91cddac88e41d76552bc9229;hb=76474239da7b68e6dd05b3cc7528b85cb3d29b84;hp=0a2ad73b66f4602da52c735380be6c431ea29608;hpb=2f0932393fb205b2e9b941db50cf5aebfba5c385;p=idzebra-moved-to-github.git diff --git a/dict/dicttest.c b/dict/dicttest.c index 0a2ad73..6b6ff00 100644 --- a/dict/dicttest.c +++ b/dict/dicttest.c @@ -4,7 +4,36 @@ * Sebastian Hammer, Adam Dickmeiss * * $Log: dicttest.c,v $ - * Revision 1.2 1994-08-18 12:40:54 adam + * Revision 1.10 1994-09-26 10:17:24 adam + * Minor changes. + * + * Revision 1.9 1994/09/22 14:43:56 adam + * First functional version of lookup with error correction. A 'range' + * specified the maximum number of insertions+deletions+substitutions. + * + * Revision 1.8 1994/09/22 10:43:44 adam + * Two versions of depend. Type 1 is the tail-type compatible with + * all make programs. Type 2 is the GNU make with include facility. + * Type 2 is default. depend rule chooses current rule. + * + * Revision 1.7 1994/09/19 16:34:26 adam + * Depend rule change. Minor changes in dicttest.c + * + * Revision 1.6 1994/09/16 15:39:12 adam + * Initial code of lookup - not tested yet. + * + * Revision 1.5 1994/09/06 13:05:14 adam + * Further development of insertion. Some special cases are + * not properly handled yet! assert(0) are put here. The + * binary search in each page definitely reduce usr CPU. + * + * Revision 1.4 1994/09/01 17:49:37 adam + * Removed stupid line. Work on insertion in dictionary. Not finished yet. + * + * Revision 1.3 1994/09/01 17:44:06 adam + * depend include change. + * + * Revision 1.2 1994/08/18 12:40:54 adam * Some development of dictionary. Not finished at all! * * Revision 1.1 1994/08/16 16:26:47 adam @@ -15,39 +44,69 @@ #include #include #include +#include #include char *prog; -Dict dict; +static Dict dict; + +static int look_hits; + +static int lookup_handle (Dict_char *name) +{ + look_hits++; + printf ("%s\n", name); + return 0; +} int main (int argc, char **argv) { const char *name = NULL; const char *inputfile = NULL; + const char *base = NULL; + int range = -1; int rw = 0; + int infosize = 4; int cache = 10; int ret; + int unique = 0; char *arg; + int no_of_iterations = 0; + int no_of_new = 0, no_of_same = 0, no_of_change = 0; + int no_of_hits = 0, no_of_misses = 0; + prog = argv[0]; - log_init (LOG_DEFAULT_LEVEL, prog, NULL); if (argc < 2) { - fprintf (stderr, "usage:\n" - " %s [-v n] [-i f] [-w] [-c n] file\n", prog); + fprintf (stderr, "usage:\n " + " %s [-r n] [-u] [-s n] [-v n] [-i f] [-w] [-c n]" + " base file\n", + prog); exit (1); } - while ((ret = options ("v:i:wc:", argv, argc, &arg)) != -2) + while ((ret = options ("r:us:v:i:wc:", argv, argc, &arg)) != -2) { if (ret == 0) { - if (name) + if (!base) + base = arg; + else if (!name) + name = arg; + else { log (LOG_FATAL, "too many files specified\n"); exit (1); } - name = arg; + } + else if (ret == 'r') + { + range = atoi (arg); + } + else if (ret == 'u') + { + unique = 1; } else if (ret == 'c') { @@ -58,9 +117,10 @@ int main (int argc, char **argv) else if (ret == 'w') rw = 1; else if (ret == 'i') - { inputfile = arg; - rw = 1; + else if (ret == 's') + { + infosize = atoi(arg); } else if (ret == 'v') { @@ -72,23 +132,30 @@ int main (int argc, char **argv) exit (1); } } - if (!name) + if (!base || !name) + { + log (LOG_FATAL, "no base and/or dictionary specified"); + exit (1); + } + common_resource = res_open (base); + if (!common_resource) { - log (LOG_FATAL, "no dictionary file given"); + log (LOG_FATAL, "cannot open resource `%s'", base); exit (1); } dict = dict_open (name, cache, rw); if (!dict) { - log (LOG_FATAL, "dict_open fail"); + log (LOG_FATAL, "dict_open fail of `%s'", name); exit (1); } if (inputfile) { FILE *ipf; - char ipf_buf[256]; - char word[256]; - int i, line = 1; + char ipf_buf[1024]; + int line = 1; + char infobytes[120]; + memset (infobytes, 0, 120); if (!(ipf = fopen(inputfile, "r"))) { @@ -96,20 +163,81 @@ int main (int argc, char **argv) exit (1); } - while (fgets (ipf_buf, 255, ipf)) + while (fgets (ipf_buf, 1023, ipf)) { - for (i=0; i<255; i++) - if (ipf_buf[i] > ' ') - word[i] = ipf_buf[i]; - else - break; - word[i] = 0; - if (i) - dict_insert (dict, word, &line); + char *ipf_ptr = ipf_buf; + sprintf (infobytes, "%d", line); + for (;*ipf_ptr && *ipf_ptr != '\n';ipf_ptr++) + { + if (isalpha(*ipf_ptr) || *ipf_ptr == '_') + { + int i = 1; + while (ipf_ptr[i] && (isalnum(ipf_ptr[i]) || + ipf_ptr[i] == '_')) + i++; + if (ipf_ptr[i]) + ipf_ptr[i++] = '\0'; + if (rw) + { + switch(dict_insert (dict, ipf_ptr, + infosize, infobytes)) + { + case 0: + no_of_new++; + break; + case 1: + no_of_change++; + if (unique) + log (LOG_LOG, "%s change\n", ipf_ptr); + break; + case 2: + if (unique) + log (LOG_LOG, "%s duplicate\n", ipf_ptr); + no_of_same++; + break; + } + } + else if(range < 0) + { + char *cp; + + cp = dict_lookup (dict, ipf_ptr); + if (cp && *cp) + no_of_hits++; + else + no_of_misses++; + } + else + { + look_hits = 0; + dict_lookup_ec (dict, ipf_ptr, range, lookup_handle); + if (look_hits) + no_of_hits++; + else + no_of_misses++; + } + ++no_of_iterations; + ipf_ptr += (i-1); + } + } ++line; } fclose (ipf); } + if (rw) + { + log (LOG_LOG, "Insertions.... %d", no_of_iterations); + log (LOG_LOG, "No of new..... %d", no_of_new); + log (LOG_LOG, "No of change.. %d", no_of_change); + log (LOG_LOG, "No of same.... %d", no_of_same); + } + else + { + log (LOG_LOG, "Lookups....... %d", no_of_iterations); + log (LOG_LOG, "No of hits.... %d", no_of_hits); + log (LOG_LOG, "No of misses.. %d", no_of_misses); + } dict_close (dict); + res_close (common_resource); return 0; }