From b76af35586285461189eaf9ea141db8e9c1ba5d9 Mon Sep 17 00:00:00 2001
From: Matin Nuhamunada <matinnu@biosustain.dtu.dk>
Date: Wed, 31 Jan 2024 08:21:59 +0000
Subject: [PATCH] feat: add exporter script

---
 .pre-commit-config.yaml  |  8 ++++----
 README.md                | 16 +++++++++++++++
 scripts/export_duckdb.py | 43 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 63 insertions(+), 4 deletions(-)
 create mode 100644 scripts/export_duckdb.py

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 82e0c2b..463727c 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -1,6 +1,6 @@
 repos:
 -   repo: https://github.com/Lucas-C/pre-commit-hooks
-    rev: v1.1.13
+    rev: v1.5.4
     hooks:
     -   id: forbid-crlf
     -   id: remove-crlf
@@ -9,7 +9,7 @@ repos:
     -   id: remove-tabs
         exclude_types: [csv]
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.1.0
+    rev: v4.5.0
     hooks:
     - id: trailing-whitespace
     - id: end-of-file-fixer
@@ -21,7 +21,7 @@ repos:
     hooks:
     - id: isort
 -   repo: https://github.com/ambv/black
-    rev: 22.3.0
+    rev: 24.1.1
     hooks:
     - id: black
-      language_version: python3.8
+      language_version: python3.11
diff --git a/README.md b/README.md
index b73c71b..f56fc38 100644
--- a/README.md
+++ b/README.md
@@ -48,5 +48,21 @@ dbt docs generate
 dbt docs serve
 ```
 
+### Exporting to newer version of DuckDB
+Right now newer version of DuckDB is not backward compatible. To migrate the data to newer version, use the script [`export_duckdb.py`](scripts/export_duckdb.py):
+```bash
+$ python scripts/export_duckdb.py -h
+usage: export_duckdb.py [-h] [--database_filename DATABASE_FILENAME] [--export_directory EXPORT_DIRECTORY]
+
+Export a DuckDB database.
+
+options:
+  -h, --help            show this help message and exit
+  --database_filename DATABASE_FILENAME
+                        The filename of the DuckDB database to export.
+  --export_directory EXPORT_DIRECTORY
+                        The directory to save the exported database.
+```
+
 # Credits
 This dbt template was inspired adapted from [jaffle_shop_duckdb](https://github.com/dbt-labs/jaffle_shop_duckdb) example.
diff --git a/scripts/export_duckdb.py b/scripts/export_duckdb.py
new file mode 100644
index 0000000..ebebc2b
--- /dev/null
+++ b/scripts/export_duckdb.py
@@ -0,0 +1,43 @@
+import argparse
+import logging
+
+import duckdb
+
+# Set up logging
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
+)
+
+
+def export_database(database_filename, export_directory):
+    """
+    Export a DuckDB database to a specified directory.
+
+    Args:
+    database_filename (str): The filename of the DuckDB database to export.
+    export_directory (str): The directory to save the exported database.
+    """
+    logging.info(f"Connecting to database: {database_filename}")
+    conn = duckdb.connect(database_filename)
+
+    logging.info(f"Exporting database to directory: {export_directory}")
+    conn.execute(f"EXPORT DATABASE '{export_directory}' (FORMAT PARQUET)")
+
+    logging.info("Database export completed successfully")
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Export a DuckDB database.")
+    parser.add_argument(
+        "--database_filename",
+        help="The filename of the DuckDB database to export.",
+        default="dbt_bgcflow.duckdb",
+    )
+    parser.add_argument(
+        "--export_directory",
+        help="The directory to save the exported database.",
+        default="./exported_database",
+    )
+    args = parser.parse_args()
+
+    export_database(args.database_filename, args.export_directory)