POSTGRES_IMPORTER

PostgreSQL Importer Plugin

Stand: 5. Dezember 2025
Version: 1.0.0
Kategorie: Importers

---

Version: 1.0.0
Type: Data Importer
Format: pg_dump SQL

---

🎯 Übersicht

ThemisDB PostgreSQL Importer Plugin zum Importieren von PostgreSQL-Datenbanken via pg_dump SQL-Dumps.

Features

• ✅ pg_dump SQL-Format - Direkter Import von PostgreSQL-Dumps
• ✅ Schema-Parsing - CREATE TABLE, CREATE SCHEMA
• ✅ Daten-Import - INSERT und COPY statements
• ✅ Typ-Mapping - Automatische Konvertierung PostgreSQL → ThemisDB
• ✅ Batch-Processing - Konfigurierbare Batch-Größe
• ✅ Fortschritt-Tracking - Progress-Callbacks
• ✅ Dry-Run-Modus - Validierung ohne Import
• ✅ Filter - Include/Exclude Tabellen

---

📦 Installation

Als Plugin

# Plugin-Struktur
plugins/importers/postgres/
├── plugin.json                    # Manifest
├── plugin.json.sig                # Signatur
├── themis_import_postgres.dll     # Windows
├── themis_import_postgres.so      # Linux
└── themis_import_postgres.dylib   # macOS

Plugin laden

auto& pm = PluginManager::instance();
pm.scanPluginDirectory("./plugins");
auto* plugin = pm.loadPlugin("postgres_importer");
auto* importer = static_cast<IImporter*>(plugin->getInstance());

---

🚀 Verwendung

Basis-Import

#include "importers/importer_interface.h"

// Create importer
auto importer = /* get from plugin manager */;

// Import options
ImportOptions options;
options.batch_size = 1000;
options.default_namespace = "imported";
options.dry_run = false;

// Import data
auto stats = importer->importData("dump.sql", options);

// Check results
std::cout << "Imported: " << stats.imported_records << " records" << std::endl;
std::cout << "Failed: " << stats.failed_records << " records" << std::endl;

Mit Progress-Callback

ImportOptions options;

auto progress = [](const std::string& stage, size_t current, size_t total) {
    std::cout << stage << ": " << current << "/" << total << std::endl;
};

auto stats = importer->importData("dump.sql", options, progress);

---

📊 Typ-Mapping

PostgreSQL	ThemisDB
INTEGER, INT, SERIAL	integer
BIGINT, BIGSERIAL	long
SMALLINT	integer
REAL, FLOAT	double
DOUBLE PRECISION	double
NUMERIC, DECIMAL	double
BOOLEAN, BOOL	boolean
CHAR, VARCHAR, TEXT	string
TIMESTAMP	datetime
DATE	date
TIME	time
JSON, JSONB	json
UUID	string
BYTEA	binary

---

📁 pg_dump Format

Unterstützte Statements

DDL:

CREATE SCHEMA app;
CREATE TABLE users (
    id SERIAL PRIMARY KEY,
    username VARCHAR(50) NOT NULL,
    email VARCHAR(100),
    created_at TIMESTAMP DEFAULT NOW()
);

DML:

-- INSERT
INSERT INTO users (id, username, email) VALUES 
    (1, 'alice', 'alice@example.com'),
    (2, 'bob', 'bob@example.com');

-- COPY
COPY users (id, username, email) FROM stdin;
1	alice	alice@example.com
2	bob	bob@example.com
\.

pg_dump erstellen

# SQL-Format (empfohlen)
pg_dump -d mydb -f dump.sql --inserts

# Mit COPY (schneller)
pg_dump -d mydb -f dump.sql

# Nur Schema
pg_dump -d mydb -f schema.sql --schema-only

# Nur Daten
pg_dump -d mydb -f data.sql --data-only

---

🔧 Konfiguration

Import-Options

struct ImportOptions {
    // General
    bool dry_run = false;
    bool continue_on_error = true;
    size_t batch_size = 1000;
    
    // Schema
    bool auto_create_schema = true;
    std::string default_namespace = "imported";
    
    // Data
    bool preserve_ids = false;
    bool update_existing = false;
    bool skip_duplicates = true;
    
    // Filtering
    std::vector<std::string> include_tables;
    std::vector<std::string> exclude_tables;
    std::vector<std::string> include_schemas;
    
    // Transformations
    std::map<std::string, std::string> column_mappings;
    std::map<std::string, std::string> table_mappings;
};

---

🚨 Einschränkungen

Aktuell nicht unterstützt

• ❌ Live-Connection - Nur pg_dump Files
• ❌ Komplexe SQL - Nur Basic DDL/DML
• ❌ Foreign Keys - Werden ignoriert
• ❌ Triggers - Werden ignoriert
• ❌ Functions - Werden ignoriert
• ❌ Custom Types - Nur Standard-Typen

Geplante Features (v2.0)

• Live PostgreSQL Connection (via libpq)
• Incremental Import (delta updates)
• Foreign Key Preservation
• Custom Type Mapping
• Parallel Processing
• Resume after failure

---

Status: ✅ Production-Ready
Getestet mit: PostgreSQL 12, 13, 14, 15, 16
Dokumentation: docs/importers/POSTGRES_IMPORTER.md