From 6b90169eb0b5c21a374ee098cba1b5d326230486 Mon Sep 17 00:00:00 2001
From: Ivor Bosloper <ivorbosloper@gmail.com>
Date: Thu, 30 Oct 2025 14:22:41 +0100
Subject: [PATCH] Use zstd with compression 15 by default, fixes #12

---
 CHANGELOG.md                       |  2 ++
 vecorel_cli/cli/options.py         | 10 +++++++++-
 vecorel_cli/conversion/base.py     |  2 ++
 vecorel_cli/convert.py             |  9 ++++++++-
 vecorel_cli/encoding/geoparquet.py | 18 ++++++++----------
 5 files changed, 29 insertions(+), 12 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 910cfdb..c07260a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 ## [Unreleased]
 
+- Change default compression to zstd
+- Add option to set compression level, zstd defaults to 15
 - Change default temporal property to datetime
 - Enable Converter.columns list and tuple types
 - Add BaseConverter get_columns hook to customize columns after reading the file
diff --git a/vecorel_cli/cli/options.py b/vecorel_cli/cli/options.py
index eb4d67c..085cb04 100644
--- a/vecorel_cli/cli/options.py
+++ b/vecorel_cli/cli/options.py
@@ -31,7 +31,15 @@ def CRS(default_value):
     type=click.Choice(COMPRESSION_METHODS),
     help="GeoParquet only: Compression method",
     show_default=True,
-    default="brotli",
+    default="zstd",
+)
+
+GEOPARQUET_COMPRESSION_LEVEL = click.option(
+    "--compression_level",
+    "-pcl",
+    type=click.IntRange(min=0, max=255),
+    help="GeoParquet only: Compression level",
+    default=None,
 )
 
 GEOPARQUET_VERSION = click.option(
diff --git a/vecorel_cli/conversion/base.py b/vecorel_cli/conversion/base.py
index 8185276..e427396 100644
--- a/vecorel_cli/conversion/base.py
+++ b/vecorel_cli/conversion/base.py
@@ -298,6 +298,7 @@ def convert(
         input_files=None,
         variant=None,
         compression=None,
+        compression_level=None,
         geoparquet_version=None,
         original_geometries=False,
         **kwargs,
@@ -425,6 +426,7 @@ def convert(
             gdf,
             properties=columns,
             compression=compression,
+            compression_level=compression_level,
             geoparquet_version=geoparquet_version,
         )
 
diff --git a/vecorel_cli/convert.py b/vecorel_cli/convert.py
index f627e8c..c9aa7e0 100644
--- a/vecorel_cli/convert.py
+++ b/vecorel_cli/convert.py
@@ -1,7 +1,13 @@
 import click
 
 from .basecommand import BaseCommand, runnable
-from .cli.options import GEOPARQUET_COMPRESSION, GEOPARQUET_VERSION, PY_PACKAGE, VECOREL_TARGET
+from .cli.options import (
+    GEOPARQUET_COMPRESSION,
+    GEOPARQUET_COMPRESSION_LEVEL,
+    GEOPARQUET_VERSION,
+    PY_PACKAGE,
+    VECOREL_TARGET,
+)
 from .cli.util import parse_converter_input_files
 from .converters import Converters
 from .registry import Registry
@@ -61,6 +67,7 @@ def get_cli_args():
                 default=None,
             ),
             "compression": GEOPARQUET_COMPRESSION,
+            "compression_level": GEOPARQUET_COMPRESSION_LEVEL,
             "geoparquet_version": GEOPARQUET_VERSION,
             "mapping_file": click.option(
                 "--mapping-file",
diff --git a/vecorel_cli/encoding/geoparquet.py b/vecorel_cli/encoding/geoparquet.py
index 5567d95..251ecdc 100644
--- a/vecorel_cli/encoding/geoparquet.py
+++ b/vecorel_cli/encoding/geoparquet.py
@@ -131,22 +131,22 @@ def get_compression(self) -> Optional[str]:
     # geoparquet_version: bool, optional, default False
     #     If True, writes the data in GeoParquet 1.0.0 format,
     #     otherwise in GeoParquet 1.1.0 format.
-    # compression: str, optional, default "brotli"
-    #     Compression algorithm to use, defaults to "brotli".
-    #     Other options are "snappy", "gzip", "lz4", "zstd", etc.
+    # compression: str, optional, default "zstd"
+    #     Compression algorithm to use, defaults to "zstd".
+    #     Other options are "snappy", "gzip", "lz4", "brotli", etc.
     def write(
         self,
         data: GeoDataFrame,
         properties: Optional[list[str]] = None,
         schema_map: SchemaMapping = {},
         dehydrate: bool = True,
-        compression: Optional[str] = None,
+        compression: Optional[str] = "zstd",
+        compression_level=None,  # default level for compression
         geoparquet_version: Optional[str] = None,
         **kwargs,  # capture unknown arguments
     ) -> bool:
-        if compression is None:
-            compression = "brotli"
-
+        if compression == "zstd" and compression_level is None:
+            compression_level = 15
         if geoparquet_version not in GEOPARQUET_VERSIONS:
             geoparquet_version = GEOPARQUET_DEFAULT_VERSION
         self.uri.parent.mkdir(parents=True, exist_ok=True)
@@ -240,9 +240,6 @@ def write(
             }
         )
 
-        if compression is None:
-            compression = "brotli"
-
         # Write the data to the Parquet file
         to_parquet(
             data,
@@ -254,6 +251,7 @@ def write(
             schema_version=geoparquet_version,
             row_group_size=self.row_group_size,
             write_covering_bbox=bool(geoparquet_version != "1.0.0"),
+            compression_level=compression_level,
         )
 
         return True