Get Duckdb-latest to test in regressions again

_report/report.R

@@ -6,7 +6,7 @@ get_report_status_file = function(path=getwd()) {
   file.path(path, "report-done")
 }
 get_report_solutions = function() {
-  c("collapse", "data.table", "dplyr", "pandas", "pydatatable", "spark", "dask", "juliadf", "juliads", "clickhouse", "cudf", "polars", "duckdb", "datafusion", "arrow", "R-arrow")
+  c("duckdb-latest", "collapse", "data.table", "dplyr", "pandas", "pydatatable", "spark", "dask", "juliadf", "juliads", "clickhouse", "cudf", "polars", "duckdb", "datafusion", "arrow", "R-arrow")
 }
 get_data_levels = function() {
   ## groupby
@@ -273,11 +273,9 @@ time_logs = function(path=getwd()) {
   lt <- load_time(path=getwd())
 
   ct = clean_time(lt)
-
-  # ct = ct %>% filter(machine_type != 'c6id.metal')
 
   # remove duckdb-latest for now
-  ct = ct %>% filter(!(solution == 'duckdb-latest'))
+  ct = ct %>% filter(!(solution == 'duckdb-latest'))  
   d = model_time(ct)
   ll <- load_logs(path=path)
 

duckdb-latest/VERSION

@@ -1 +1 @@
-1.0.99.9000
+1.3.1

duckdb-latest/setup-duckdb-latest.sh

@@ -12,8 +12,11 @@ Rscript -e 'install.packages("DBI", lib="./duckdb-latest/r-duckdb-latest", repos
 cd duckdb-latest
 rm -rf duckdb-r
 git clone https://github.com/duckdb/duckdb-r.git
+cd duckdb-r
+git checkout next
+cd ..
 ncores=`python3 -c 'import multiprocessing as mp; print(mp.cpu_count())'`
-MAKE="make -j$ncores" R CMD INSTALL -l "./r-duckdb-latest" duckdb-r
+MAKEFLAGS="-j$ncores" R CMD INSTALL -l "./r-duckdb-latest" duckdb-r
 rm -rf duckdb-r
 cd ..
 

duckdb/groupby-duckdb.R

@@ -39,6 +39,8 @@ if (machine_type == 'c6id.4xlarge' && on_disk) {
   dbExecute(con, "pragma memory_limit='25G'")
 }
 
+dbExecute(con, "SET enable_progress_bar = false;")
+
 ncores = parallel::detectCores()
 invisible(dbExecute(con, sprintf("PRAGMA THREADS=%d", ncores)))
 git = dbGetQuery(con, "SELECT source_id FROM pragma_version()")[[1L]]

duckdb/join-duckdb.R

@@ -209,6 +209,10 @@ invisible(dbExecute(con, "DROP TABLE IF EXISTS ans"))
 
 dbDisconnect(con, shutdown=TRUE)
 
+if (on_disk) {
+  unlink(db_file)
+}
+
 cat(sprintf("joining finished, took %.0fs\n", proc.time()[["elapsed"]]-task_init))
 
 if( !interactive() ) q("no", status=0)