#include "entity_ops.h" #include "../../../../cpp/vendor/sqlite3/sqlite3.h" #include #include #include #include #include namespace ge { // ---------------------------------------------------------------------------- // FNV1a-64 — debe coincidir con graph_sources.cpp // ---------------------------------------------------------------------------- static uint64_t fnv1a64(const char* s) { uint64_t h = 1469598103934665603ULL; for (; s && *s; ++s) { h ^= (uint8_t)*s; h *= 1099511628211ULL; } return h; } // ---------------------------------------------------------------------------- // Heuristicas — sin para evitar peso. Inspeccion lineal del string. // ---------------------------------------------------------------------------- static bool is_email(const char* s) { if (!s || !*s) return false; const char* at = std::strchr(s, '@'); if (!at || at == s) return false; if (std::strchr(at + 1, '@')) return false; // dos @ const char* dot = std::strchr(at + 1, '.'); if (!dot || dot[1] == 0) return false; if (std::strchr(s, ' ')) return false; return true; } static bool is_ipv4(const char* s) { if (!s || !*s) return false; int parts = 0, digits = 0, n = 0; for (const char* p = s; ; ++p) { if (*p >= '0' && *p <= '9') { n = n * 10 + (*p - '0'); if (++digits > 3 || n > 255) return false; } else if (*p == '.' || *p == 0) { if (digits == 0) return false; ++parts; digits = 0; n = 0; if (*p == 0) break; } else { return false; } } return parts == 4; } static bool is_url(const char* s) { if (!s) return false; return std::strncmp(s, "http://", 7) == 0 || std::strncmp(s, "https://", 8) == 0; } static bool is_domain(const char* s) { if (!s || !*s) return false; if (std::strchr(s, ' ') || std::strchr(s, '@') || std::strchr(s, '/')) return false; const char* dot = std::strchr(s, '.'); if (!dot || dot == s || dot[1] == 0) return false; // El TLD debe ser al menos 2 caracteres alfabeticos int tld = 0; for (const char* p = std::strrchr(s, '.') + 1; *p; ++p) { if (!std::isalpha((unsigned char)*p)) return false; ++tld; } return tld >= 2; } static bool is_phone(const char* s) { if (!s || !*s) return false; int digits = 0; for (const char* p = s; *p; ++p) { if (*p >= '0' && *p <= '9') ++digits; else if (*p == '+' || *p == ' ' || *p == '-' || *p == '(' || *p == ')' || *p == '.') {} else return false; } return digits >= 7 && digits <= 15; } DetectedType detect_type(const char* text) { if (!text || !*text) return DT_TEXT; if (is_email(text)) return DT_EMAIL; if (is_ipv4(text)) return DT_IP_ADDRESS; if (is_url(text)) return DT_URL; if (is_phone(text)) return DT_PHONE; // antes que domain (numeros con puntos) if (is_domain(text)) return DT_DOMAIN; return DT_TEXT; } const char* detected_type_name(DetectedType dt) { switch (dt) { case DT_EMAIL: return "email"; case DT_IP_ADDRESS: return "ip_address"; case DT_URL: return "url"; case DT_DOMAIN: return "domain"; case DT_PHONE: return "phone"; case DT_TEXT: default: return "text"; } } // ---------------------------------------------------------------------------- // SQLite helpers // ---------------------------------------------------------------------------- static bool exec_one(sqlite3* db, const char* sql, const char** params, int n_params) { sqlite3_stmt* st = nullptr; if (sqlite3_prepare_v2(db, sql, -1, &st, nullptr) != SQLITE_OK) return false; for (int i = 0; i < n_params; ++i) { sqlite3_bind_text(st, i + 1, params[i] ? params[i] : "", -1, SQLITE_TRANSIENT); } int rc = sqlite3_step(st); sqlite3_finalize(st); return rc == SQLITE_DONE; } static std::string now_iso() { using namespace std::chrono; auto t = system_clock::to_time_t(system_clock::now()); char buf[32]; std::strftime(buf, sizeof(buf), "%Y-%m-%dT%H:%M:%S", std::gmtime(&t)); return std::string(buf) + "Z"; } static long long now_ms() { using namespace std::chrono; return duration_cast(system_clock::now().time_since_epoch()).count(); } // ---------------------------------------------------------------------------- // CRUD // ---------------------------------------------------------------------------- bool entity_insert(const char* db_path, const char* name, const char* type_ref, char* out_id, size_t out_id_n) { if (!db_path || !name || !*name || !out_id || out_id_n < 32) return false; std::string tref; if (type_ref && *type_ref) { tref = type_ref; } else { tref = detected_type_name(detect_type(name)); } std::snprintf(out_id, out_id_n, "%s_%lld", tref.c_str(), now_ms()); std::string ts = now_iso(); sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READWRITE, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } const char* sql = "INSERT INTO entities (id, name, type_ref, source, created_at, updated_at) " "VALUES (?, ?, ?, 'manual', ?, ?)"; const char* params[5] = { out_id, name, tref.c_str(), ts.c_str(), ts.c_str() }; bool ok = exec_one(db, sql, params, 5); sqlite3_close(db); return ok; } bool entity_delete(const char* db_path, const char* id) { if (!db_path || !id) return false; sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READWRITE, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } const char* p_rel[2] = { id, id }; const char* p_ent[1] = { id }; exec_one(db, "DELETE FROM relations WHERE from_entity = ? OR to_entity = ?", p_rel, 2); bool ok = exec_one(db, "DELETE FROM entities WHERE id = ?", p_ent, 1); sqlite3_close(db); return ok; } bool entity_update_type(const char* db_path, const char* id, const char* new_type) { if (!db_path || !id || !new_type) return false; sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READWRITE, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } std::string ts = now_iso(); const char* p[3] = { new_type, ts.c_str(), id }; bool ok = exec_one(db, "UPDATE entities SET type_ref = ?, updated_at = ? WHERE id = ?", p, 3); sqlite3_close(db); return ok; } bool entity_duplicate(const char* db_path, const char* id, char* out_id, size_t out_id_n) { if (!db_path || !id || !out_id || out_id_n < 32) return false; sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READWRITE, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } sqlite3_stmt* st = nullptr; const char* sel = "SELECT name, type_ref, description, domain, tags, source, metadata, notes " "FROM entities WHERE id = ?"; if (sqlite3_prepare_v2(db, sel, -1, &st, nullptr) != SQLITE_OK) { sqlite3_close(db); return false; } sqlite3_bind_text(st, 1, id, -1, SQLITE_TRANSIENT); if (sqlite3_step(st) != SQLITE_ROW) { sqlite3_finalize(st); sqlite3_close(db); return false; } auto col = [&](int i) { const unsigned char* p = sqlite3_column_text(st, i); return std::string(p ? (const char*)p : ""); }; std::string name = col(0) + " (copia)"; std::string tref = col(1); std::string desc = col(2); std::string dom = col(3); std::string tags = col(4); std::string src = col(5); std::string meta = col(6); std::string notes = col(7); sqlite3_finalize(st); std::snprintf(out_id, out_id_n, "%s_%lld", tref.c_str(), now_ms()); std::string ts = now_iso(); const char* ins = "INSERT INTO entities (id, name, type_ref, description, domain, tags, source, " "metadata, notes, created_at, updated_at) " "VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)"; sqlite3_stmt* ist = nullptr; if (sqlite3_prepare_v2(db, ins, -1, &ist, nullptr) != SQLITE_OK) { sqlite3_close(db); return false; } sqlite3_bind_text(ist, 1, out_id, -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 2, name.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 3, tref.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 4, desc.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 5, dom.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 6, tags.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 7, src.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 8, meta.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 9, notes.c_str(),-1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 10, ts.c_str(), -1, SQLITE_TRANSIENT); sqlite3_bind_text(ist, 11, ts.c_str(), -1, SQLITE_TRANSIENT); bool ok = sqlite3_step(ist) == SQLITE_DONE; sqlite3_finalize(ist); sqlite3_close(db); return ok; } bool relation_insert(const char* db_path, const char* from_id, const char* to_id, const char* name) { if (!db_path || !from_id || !to_id) return false; const char* rel = (name && *name) ? name : k_default_relation_name; sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READWRITE, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } char id[64]; std::snprintf(id, sizeof(id), "rel_%lld", now_ms()); std::string ts = now_iso(); const char* sql = "INSERT INTO relations (id, name, from_entity, to_entity, created_at, updated_at) " "VALUES (?, ?, ?, ?, ?, ?)"; const char* params[6] = { id, rel, from_id, to_id, ts.c_str(), ts.c_str() }; bool ok = exec_one(db, sql, params, 6); sqlite3_close(db); return ok; } // ---------------------------------------------------------------------------- // Index user_data -> sql id // ---------------------------------------------------------------------------- bool entity_index_build(const char* db_path, EntityIndex* idx) { if (!db_path || !idx) return false; idx->by_hash.clear(); sqlite3* db = nullptr; if (sqlite3_open_v2(db_path, &db, SQLITE_OPEN_READONLY, nullptr) != SQLITE_OK) { if (db) sqlite3_close(db); return false; } sqlite3_stmt* st = nullptr; if (sqlite3_prepare_v2(db, "SELECT id FROM entities", -1, &st, nullptr) != SQLITE_OK) { sqlite3_close(db); return false; } while (sqlite3_step(st) == SQLITE_ROW) { const unsigned char* p = sqlite3_column_text(st, 0); if (!p) continue; const char* id = (const char*)p; idx->by_hash.emplace(fnv1a64(id), std::string(id)); } sqlite3_finalize(st); sqlite3_close(db); return true; } const char* entity_index_lookup(const EntityIndex& idx, uint64_t user_data) { auto it = idx.by_hash.find(user_data); return (it == idx.by_hash.end()) ? nullptr : it->second.c_str(); } } // namespace ge