ORE Studio 0.0.4
Loading...
Searching...
No Matches
dataset_repository.hpp
1/* -*- mode: c++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*-
2 *
3 * Copyright (C) 2025 Marco Craveiro <marco.craveiro@gmail.com>
4 *
5 * This program is free software; you can redistribute it and/or modify it under
6 * the terms of the GNU General Public License as published by the Free Software
7 * Foundation; either version 3 of the License, or (at your option) any later
8 * version.
9 *
10 * This program is distributed in the hope that it will be useful, but WITHOUT
11 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
12 * FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
13 * details.
14 *
15 * You should have received a copy of the GNU General Public License along with
16 * this program; if not, write to the Free Software Foundation, Inc., 51
17 * Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 *
19 */
20#ifndef ORES_DQ_REPOSITORY_DATASET_REPOSITORY_HPP
21#define ORES_DQ_REPOSITORY_DATASET_REPOSITORY_HPP
22
23#include <string>
24#include <vector>
25#include <sqlgen/postgres.hpp>
26#include <boost/uuid/uuid.hpp>
27#include "ores.logging/make_logger.hpp"
28#include "ores.database/domain/context.hpp"
29#include "ores.dq/domain/dataset.hpp"
30
31namespace ores::dq::repository {
32
37private:
38 inline static std::string_view logger_name =
39 "ores.dq.repository.dataset_repository";
40
41 [[nodiscard]] static auto& lg() {
42 using namespace ores::logging;
43 static auto instance = make_logger(logger_name);
44 return instance;
45 }
46
47public:
49
50 explicit dataset_repository(context ctx);
51
52 std::string sql();
53
54 void write(const domain::dataset& dataset);
55 void write(const std::vector<domain::dataset>& datasets);
56
57 std::vector<domain::dataset> read_latest();
58 std::vector<domain::dataset> read_latest(const boost::uuids::uuid& id);
59 std::vector<domain::dataset>
60 read_latest(std::uint32_t offset, std::uint32_t limit);
61
65 std::vector<domain::dataset>
66 read_latest_by_catalog(const std::string& catalog_name);
67
71 std::vector<domain::dataset>
72 read_latest_by_subject_area(const std::string& subject_area_name,
73 const std::string& domain_name);
74
78 std::vector<domain::dataset>
79 read_latest_by_origin(const std::string& origin_code);
80
81 std::uint32_t get_total_count();
82 std::vector<domain::dataset> read_all(const boost::uuids::uuid& id);
83 void remove(const boost::uuids::uuid& id);
84
85private:
86 context ctx_;
87};
88
89}
90
91#endif
Implements logging infrastructure for ORE Studio.
Definition boost_severity.hpp:28
Context for the operations on a postgres database.
Definition context.hpp:30
Represents a data quality dataset with lineage tracking.
Definition dataset.hpp:36
Reads and writes datasets to data storage.
Definition dataset_repository.hpp:36
std::vector< domain::dataset > read_latest_by_origin(const std::string &origin_code)
Reads latest datasets by origin.
Definition dataset_repository.cpp:142
std::vector< domain::dataset > read_latest_by_catalog(const std::string &catalog_name)
Reads latest datasets by catalog.
Definition dataset_repository.cpp:106
std::vector< domain::dataset > read_latest_by_subject_area(const std::string &subject_area_name, const std::string &domain_name)
Reads latest datasets by subject area.
Definition dataset_repository.cpp:122