Deploying to gh-pages from @ db08740 🚀

Author: Cidree

articles/benchmark.html

@@ -111,8 +111,8 @@ memo_diff <- round(as.numeric(temp$mem_alloc[2] / temp$mem_alloc[1]),1)
 time_diff <- (1 - round(as.numeric(temp$median[1] / temp$median[2]),2))*100
 ```
 
-In this example working with 1 million points, {duckspatial} was 60%
-faster and used 5.4 times less memory than {sf}. Not bad.
+In this example working with 1 million points, {duckspatial} was 57%
+faster and used 5.5 times less memory than {sf}. Not bad.
 
 ``` r
 ggplot(data = df_bench_join) +
@@ -171,7 +171,7 @@ memo_diff <- round(as.numeric(temp$mem_alloc[2] / temp$mem_alloc[1]),1)
 time_diff <- (1 - round(as.numeric(temp$median[1] / temp$median[2]),2))*100
 ```
 
-In this example working with 1 million points, {duckspatial} was 69%
+In this example working with 1 million points, {duckspatial} was 76%
 faster and used 2.6 times less memory than {sf}.
 
 plot
@@ -236,8 +236,6 @@ df_bench_distance <- lapply(
     dplyr::bind_rows()
 #> Warning: Some expressions had a GC in every iteration; so filtering is
 #> disabled.
-#> Warning: Some expressions had a GC in every iteration; so filtering is
-#> disabled.
 
 
 # calculate difference in performance
@@ -249,7 +247,7 @@ time_diff <- (1 - round(as.numeric(temp$median[1] / temp$median[2]),2))*100
 ```
 
 In this example calculating the distance between 10K points,
-{duckspatial} was 86% faster, but used 2 times more memory than {sf}.
+{duckspatial} was 87% faster, but used 2 times more memory than {sf}.
 Mind you that {sf} is still more efficient when calculating Euclidean
 distances.
 

articles/benchmark.md

@@ -1,5 +1,26 @@
 # duckspatial
 
+> Important note: The {duckspatial} R package is undergoing major
+> changes in preparation for v1.0.0. This release will align with DuckDB
+> v1.5 and its native CRS support in the Spatial extension.
+>
+> Previously, {duckspatial} relied on a workaround to store CRS
+> information in a dedicated column (named “crs_duckspatial” by default
+> and referenced via the crs_column argument). With native CRS support
+> now available, this workaround is no longer needed and will be
+> removed. As a result, the crs and crs_column arguments are deprecated
+> as of v0.9.0.
+>
+> The upcoming version also introduces a new native spatial class,
+> duckspatial_df. This represents a lazy spatial table backed by a
+> temporary DuckDB view, allowing workflows to remain fully lazy until
+> results are explicitly materialized with ddbs_collect(). In practice,
+> this means you can work with {duckspatial} without eagerly loading
+> data into R.
+>
+> Finally, v1.0.0 will include many new functions and broader API
+> improvements.
+
 The **{duckspatial}** package provides fast and memory-efficient
 functions to analyze and manipulate large spatial vector datasets in R.
 It allows R users to benefit directly from the analytical power of

articles/benchmark_files/figure-html/unnamed-chunk-4-1.png

@@ -1,5 +1,26 @@
 # duckspatial
 
+> Important note: The {duckspatial} R package is undergoing major
+> changes in preparation for v1.0.0. This release will align with DuckDB
+> v1.5 and its native CRS support in the Spatial extension.
+>
+> Previously, {duckspatial} relied on a workaround to store CRS
+> information in a dedicated column (named “crs_duckspatial” by default
+> and referenced via the crs_column argument). With native CRS support
+> now available, this workaround is no longer needed and will be
+> removed. As a result, the crs and crs_column arguments are deprecated
+> as of v0.9.0.
+>
+> The upcoming version also introduces a new native spatial class,
+> duckspatial_df. This represents a lazy spatial table backed by a
+> temporary DuckDB view, allowing workflows to remain fully lazy until
+> results are explicitly materialized with ddbs_collect(). In practice,
+> this means you can work with {duckspatial} without eagerly loading
+> data into R.
+>
+> Finally, v1.0.0 will include many new functions and broader API
+> improvements.
+
 The **{duckspatial}** package provides fast and memory-efficient
 functions to analyze and manipulate large spatial vector datasets in R.
 It allows R users to benefit directly from the analytical power of

articles/benchmark_files/figure-html/unnamed-chunk-6-1.png

@@ -6,7 +6,7 @@ articles:
   benchmark: benchmark.html
   duckspatial: duckspatial.html
   spatial_joins: spatial_joins.html
-last_built: 2026-02-10T16:13Z
+last_built: 2026-02-10T16:21Z
 urls:
   reference: https://cidree.github.io/duckspatial/reference
   article: https://cidree.github.io/duckspatial/articles