/* * Command-line driver of the C stemmer core, used by test/core/compare.py. * * Reads one word per line from standard input and prints, tab-separated: * word, stem, segments joined with "|", suffix classes joined with " ", source * * Option: -l FILE a user lexicon (lines "wordn|v", as kaastem's lexicon files) * * Build: cc -std=c99 -O2 -Isrc test/core/kaacore_cli.c src/kaacore.c src/kaadata.c -o kaacore_cli */ #include #include #include #include "kaacore.h" static KaaExtra * load_lexicon(const char *path) { static char line[4096]; FILE *f = fopen(path, "r"); KaaExtra *x = kaa_extra_new(); if (!f || !x) return NULL; while (fgets(line, sizeof line, f)) { char *tab; line[strcspn(line, "#\r\n")] = '\0'; tab = strchr(line, '\t'); if (!tab) continue; *tab = '\0'; if (kaa_extra_add(x, line, strlen(line), tab[1] == 'v' ? 'v' : 'n') != 0) return NULL; } fclose(f); return x; } int main(int argc, char **argv) { static char line[1 << 16]; KaaExtra *extra = NULL; if (argc == 3 && strcmp(argv[1], "-l") == 0 && !(extra = load_lexicon(argv[2]))) { fprintf(stderr, "cannot load %s\n", argv[2]); return 2; } while (fgets(line, sizeof line, stdin)) { size_t n = strcspn(line, "\n"); KaaAnalysis a; int i; line[n] = '\0'; if (kaa_analyze_ex(line, n, extra, &a) != 0) { fprintf(stderr, "out of memory on: %s\n", line); return 2; } printf("%s\t%s\t", line, a.stem); for (i = 0; i < a.nsegments; i++) printf("%s%s", i ? "|" : "", a.segments[i]); printf("\t"); for (i = 0; i < a.nclasses; i++) printf("%s%s", i ? " " : "", kaa_class_names[a.classes[i]]); printf("\t%s\n", kaa_source_name(a.source)); kaa_analysis_free(&a); } kaa_extra_free(extra); return 0; }