Merge branch 'main' of https://github.com/Franz-Kafka/duckdb-web

Author: Franz-Kafka

_includes/list_of_community_extensions.md

@@ -240,20 +240,13 @@ A separate blog post will be coming.
 
 ## Performance and Optimizations
 
-### Sorting Rework
-
-[Laurens (@lnkuiper)](https://github.com/lnkuiper) [rewrote DuckDB’s sorting implementation](https://github.com/duckdb/duckdb/pull/17584#thread-scaling-performance) ([again](https://github.com/duckdb/duckdb/pull/1561)). This new implementation uses a k-way merge sort to reduce data movement. It is also adaptive to pre-sorted data and uses a new API that makes it possible to use this new sorting code elsewhere in DuckDB, for example in window functions. We are seeing much better thread scaling performance with this implementation. We will publish a separate blog post with more detailed performance measurements.
-
-### Materializing Common Table Expressions
-
-Common table expressions (CTEs) are now materialized by default (instead of inlining them). This both improves performance and resolves some correctness bugs that happened due to inlining.
-This feature was [implemented](https://github.com/duckdb/duckdb/pull/17459) by [Denis Hirn (kryonix)](https://github.com/kryonix), who [contributed support for recursive CTEs](https://github.com/duckdb/duckdb/pull/404) back in 2020.
+DuckDB v1.4.0 received a number of performance optimizations.
 
 ### Checkpointing In-Memory Tables
 
 In-memory tables now support [checkpointing](https://github.com/duckdb/duckdb/pull/18348). This has two key benefits:
 
-* In-memory tables now support compression. This is disabled by default – you can turn it on using:
+* In-memory tables now support compression, which can have a significant (5-10×) performance boost for some queries. This is disabled by default – you can turn it on using:
 
   ```sql
   ATTACH ':memory:' AS memory_compressed (COMPRESS);
@@ -262,6 +255,21 @@ In-memory tables now support [checkpointing](https://github.com/duckdb/duckdb/pu
 
 * Checkpointing triggers vacuuming deleted rows, allowing space to be reclaimed after deletes/truncation.
 
+### Sorting Rework
+
+[Laurens (@lnkuiper)](https://github.com/lnkuiper) [rewrote DuckDB’s sorting implementation](https://github.com/duckdb/duckdb/pull/17584#thread-scaling-performance) ([again](https://github.com/duckdb/duckdb/pull/1561)). This new implementation uses a k-way merge sort to reduce data movement. It is also adaptive to pre-sorted data and uses a new API that makes it possible to use this new sorting code elsewhere in DuckDB, for example in window functions. We are seeing much better thread scaling performance with this implementation. We will publish a separate blog post with more detailed performance measurements.
+
+> Update We have now covered this in a [separate blog post]({% post_url 2025-09-24-sorting-again %}).
+
+### Materializing Common Table Expressions
+
+Common table expressions (CTEs) are now materialized by default (instead of inlining them). This both improves performance and resolves some correctness bugs that happened due to inlining.
+This feature was [implemented](https://github.com/duckdb/duckdb/pull/17459) by [Denis Hirn (kryonix)](https://github.com/kryonix), who [contributed support for recursive CTEs](https://github.com/duckdb/duckdb/pull/404) back in 2020.
+
+### Additional Optimizations
+
+DuckBB v1.4.0 fixes some [scaling issues](https://github.com/duckdb/duckdb/pull/17985), supports [caching for hashes of string dictionaries](https://github.com/duckdb/duckdb/pull/18580) and has several [aggregation optimizations](https://github.com/duckdb/duckdb/pull/17718).
+
 ## Distribution 
 
 ### macOS Notarization

_posts/2025-09-16-announcing-duckdb-140.md

@@ -8,7 +8,7 @@ excerpt: "Today we are releasing DuckDB 1.4.1, the first bugfix release of our L
 tags: ["release"]
 ---
 
-In this blog post, we highlight a few important fixes and convenience improvements in DuckDB v1.4.1 LTS.
+In this blog post, we highlight a few important fixes and convenience improvements in DuckDB v1.4.1, the first bugfix release in [DuckDB's 1.4 LTS line]({% post_url 2025-09-16-announcing-duckdb-140 %}).
 You can find the complete [release notes on GitHub](https://github.com/duckdb/duckdb/releases/tag/v1.4.1).
 
 To install the new version, please visit the [installation page]({% link install/index.html %}).

_posts/2025-10-07-announcing-duckdb-141.md

@@ -0,0 +1,39 @@
+---
+layout: post
+title: "Benchmark Results for DuckDB v1.4 LTS"
+author: "The DuckDB team"
+thumb: "/images/blog/thumbs/benchmark-results-1-4-lts.svg"
+image: "/images/blog/thumbs/benchmark-results-1-4-lts.png"
+excerpt: "DuckDB v1.4 LTS is both fast and scalable. In in-memory mode, it is the fastest system on ClickBench. In disk-based mode, it can run complex analytical queries on a dataset equivalent to 100 TB CSV files on a single machine."
+tags: ["release"]
+---
+
+## ClickBench
+
+Today, DuckDB hit #1 on the popular [ClickBench database benchmark](https://benchmark.clickhouse.com/):
+
+<img src="/images/blog/clickbench-top1-light.png"
+     alt="ClickBench results as of October 9, 2025"
+     width="800"
+     class="lightmode-img"
+     />
+<img src="/images/blog/clickbench-top1-dark.png"
+     alt="ClickBench results as of October 9, 2025"
+     width="800"
+     class="darkmode-img"
+     />
+
+This result was made possible due several [performance optimizations]({% post_url 2025-09-16-announcing-duckdb-140 %}#performance-and-optimizations) in DuckDB v1.4.
+
+## TPC-H SF100,000
+
+DuckDB is not only fast but it is also scalable. We have recently run the queries of the [TPC-H workload]({% link docs/stable/core_extensions/tpch.md %}) on the SF100,000 dataset, which is equivalent to 100,000 GB of CSV files.
+
+We ran the experiment on an [`i8g.48xlarge` EC2 instance](https://aws.amazon.com/ec2/instance-types/i8g/), which has 1.5 TB of RAM and 192 CPU cores (AWS Graviton4, Arm64). This instance has 12 NVMe SSD disks, each 3750 GB in size. We created a RAID-0 array from them to have a single 45 TB partition and formatted it using [XFS]({% link docs/stable/guides/performance/environment.md %}#local-disk).
+
+We generated the dataset with the [tpchgen-rs](https://github.com/clflushopt/tpchgen-rs/) tool, a pure Rust implementation of the TPC-H generator. We configured the generator to produce chunks of Parquet files and loaded them into DuckDB. The final DuckDB database was about 27 TB in size (as a single file!).
+
+DuckDB completed all 22 queries of the benchmark using its [larger-than-memory processing]({% post_url 2024-07-09-memory-management %}). For some queries, this required spilling _about 7 terabytes of data_ to disk.
+The median query runtime was 1.19 hours and the geometric mean runtime was 1.13 hours.
+
+We will publish a detailed write-up on this experiment in the coming weeks.

_posts/2025-10-09-benchmark-results-14-lts.md

@@ -701,7 +701,7 @@ body.landing{
 		right: 0px;
 	}
 
-	section.clickbench {
+	section.most_scalable {
 		.flexwrap {
 			display: flex;
 			flex-wrap: wrap;
@@ -789,7 +789,7 @@ body.landing{
 
 
 @media only screen and (max-width: 1100px) {
-	body.landing section.clickbench {
+	body.landing section.most_scalable {
 		.flexwrap {
 			display: flex;
 			flex-wrap: wrap;
@@ -921,7 +921,7 @@ body.landing{
 			max-width: unset;
 		}	
 	}
-	body.landing section.clickbench .box > div:first-child{
+	body.landing section.most_scalable .box > div:first-child{
 		max-width: 100px;
 	}
 }

_sass/pages/_home.scss

@@ -0,0 +1,67 @@
+---
+warning: DO NOT CHANGE THIS MANUALLY, THIS IS GENERATED BY https://github/duckdb/community-extensions repository, check README there
+title: a5
+excerpt: |
+  DuckDB Community Extensions
+  Hierarchical pentagonial indexing for geospatial data that is equal-area and millimeter accurate.
+
+docs:
+  extended_description: |
+    For more information regarding usage, see the [documentation](https://query.farm/duckdb_extension_a5.html).
+extension:
+  build: cmake
+  description: Hierarchical pentagonial indexing for geospatial data that is equal-area and millimeter accurate.
+  language: C++
+  license: Apache-2.0
+  maintainers:
+    - rustyconover
+  name: a5
+  requires_toolchains: rust
+  excluded_platforms: "wasm_mvp;wasm_eh;wasm_threads;"
+  version: 2025101102
+repo:
+  github: query-farm/a5
+  ref: b61ebc04de40e9e36d150c9743f4e80a518ca99a
+
+extension_star_count: 2
+extension_star_count_pretty: 2
+extension_download_count: null
+extension_download_count_pretty: n/a
+image: '/images/community_extensions/social_preview/preview_community_extension_a5.png'
+layout: community_extension_doc
+---
+
+### Installing and Loading
+```sql
+INSTALL {{ page.extension.name }} FROM community;
+LOAD {{ page.extension.name }};
+```
+
+{% if page.docs.hello_world %}
+### Example
+```sql
+{{ page.docs.hello_world }}```
+{% endif %}
+
+{% if page.docs.extended_description %}
+### About {{ page.extension.name }}
+{{ page.docs.extended_description }}
+{% endif %}
+
+### Added Functions
+
+<div class="extension_functions_table"></div>
+
+| function_name | function_type | description | comment | examples |
+|---------------|---------------|-------------|---------|----------|
+| a5_area       | scalar        | NULL        | NULL    | NULL     |
+| a5_boundary   | scalar        | NULL        | NULL    | NULL     |
+| a5_cell       | scalar        | NULL        | NULL    | NULL     |
+| a5_children   | scalar        | NULL        | NULL    | NULL     |
+| a5_lon_lat    | scalar        | NULL        | NULL    | NULL     |
+| a5_num_cells  | scalar        | NULL        | NULL    | NULL     |
+| a5_parent     | scalar        | NULL        | NULL    | NULL     |
+| a5_res0_cells | scalar        | NULL        | NULL    | NULL     |
+| a5_resolution | scalar        | NULL        | NULL    | NULL     |
+
+

community_extensions/extensions/a5.md

@@ -6,8 +6,7 @@ excerpt: |
   The Airport extension brings Arrow Flight support to DuckDB, enabling DuckDB to query, modify, and store data from Arrow Flight servers.
 
 docs:
-  extended_description: |
-    For more information regarding usage, see the [documentation](https://query.farm/duckdb_extension_airport.html).
+  extended_description: For more information regarding usage, see the [documentation](https://query.farm/duckdb_extension_airport.html).
 extension:
   build: cmake
   description: The Airport extension brings Arrow Flight support to DuckDB, enabling
@@ -21,13 +20,13 @@ extension:
   requires_toolchains: parser_tools
   test_config: "{\"test_env_variables\":\n  {\n    \"AIRPORT_TEST_SERVER\": \"grpc+tls://airport-ci.query.farm\"\
     \n  }\n}\n"
-  version: 2025091801
+  version: '2025101201'
 repo:
   github: query-farm/airport
-  ref: 0d7b873a0fdf276e9a42e5716fdd872497c4d048
+  ref: 5872b7de86764abe0daeeeccfd3ad4495f5509c5
 
-extension_star_count: 302
-extension_star_count_pretty: 302
+extension_star_count: 304
+extension_star_count_pretty: 304
 extension_download_count: 409
 extension_download_count_pretty: 409
 image: '/images/community_extensions/social_preview/preview_community_extension_airport.png'

community_extensions/extensions/airport.md

@@ -43,8 +43,8 @@ docs:
 
 
 
-extension_star_count: 47
-extension_star_count_pretty: 47
+extension_star_count: 48
+extension_star_count_pretty: 48
 extension_download_count: 5427
 extension_download_count_pretty: 5.4k
 image: '/images/community_extensions/social_preview/preview_community_extension_nanoarrow.png'

community_extensions/extensions/arrow.md

@@ -8,7 +8,7 @@ excerpt: |
 extension:
   name: bigquery
   description: Integrates DuckDB with Google BigQuery, allowing direct querying and management of BigQuery datasets
-  version: 0.5.0
+  version: 0.5.1
   language: C++
   build: cmake
   license: MIT
@@ -20,7 +20,7 @@ extension:
 
 repo:
   github: hafenkran/duckdb-bigquery
-  ref: 621dbb517c11d82c0d7693c362591d156ab3da0c
+  ref: 385d9c46d388e75a7ef620104bf8b543fd351029
 
 docs:
   hello_world: |

community_extensions/extensions/bigquery.md

@@ -6,20 +6,21 @@ excerpt: |
   Provides high-performance, space-efficient probabilistic data structures—including quotient, XOR, and binary fuse filters—for fast approximate set membership testing with no false negatives and configurable false positive rates.
 
 docs:
-  extended_description: |
-    For more information regarding usage, see the [documentation](https://query.farm/duckdb_extension_bitfilters.html).
+  extended_description: For more information regarding usage, see the [documentation](https://query.farm/duckdb_extension_bitfilters.html).
 extension:
   build: cmake
-  description: Provides high-performance, space-efficient probabilistic data structures—including quotient, XOR, and binary fuse filters—for fast approximate set membership testing with no false negatives and configurable false positive rates.
+  description: "Provides high-performance, space-efficient probabilistic data structures\u2014\
+    including quotient, XOR, and binary fuse filters\u2014for fast approximate set\
+    \ membership testing with no false negatives and configurable false positive rates."
   language: C++
   license: MIT
   maintainers:
-    - rustyconover
+  - rustyconover
   name: bitfilters
-  version: 2025091601
+  version: '2025101201'
 repo:
   github: query-farm/bitfilters
-  ref: cfcf8cf24d7299aa07d0a24c234e5c1f31d04065
+  ref: 002ab751fc30d94d005b2db27c9f68271b8a36b5
 
 extension_star_count: 2
 extension_star_count_pretty: 2