Fix remaining issues in #10 (#13)

* Fix remaining issues in arrow -> R-arrow rename

* Fix bug in rename code in report.R

The previous code was causing something wild to happen. The changed code is idiomatic code for replacing values in a data.frame based on a condition.

Author: amoeba

R-arrow/groupby-R-arrow.R

@@ -7,8 +7,8 @@ source("./_helpers/helpers.R")
 stopifnot(requireNamespace("bit64", quietly=TRUE)) # used in chk to sum numeric columns
 .libPaths("./arrow/r-arrow") # tidyverse/dplyr#4641 ## leave it like here in case if this affects arrow pkg as well
 suppressPackageStartupMessages({
-  library("arrow", lib.loc="./arrow/r-arrow", warn.conflicts=FALSE)
-  library("dplyr", lib.loc="./arrow/r-arrow", warn.conflicts=FALSE)
+  library("arrow", lib.loc="./R-arrow/r-arrow", warn.conflicts=FALSE)
+  library("dplyr", lib.loc="./R-arrow/r-arrow", warn.conflicts=FALSE)
 })
 ver = packageVersion("arrow")
 git = ""

R-arrow/join-R-arrow.R

@@ -6,8 +6,8 @@ source("./_helpers/helpers.R")
 
 .libPaths("./arrow/r-arrow") # tidyverse/dplyr#4641 ## leave it like here in case if this affects arrow pkg as well
 suppressPackageStartupMessages({
-  library("arrow", lib.loc="./arrow/r-arrow", warn.conflicts=FALSE)
-  library("dplyr", lib.loc="./arrow/r-arrow", warn.conflicts=FALSE)
+  library("arrow", lib.loc="./R-arrow/r-arrow", warn.conflicts=FALSE)
+  library("dplyr", lib.loc="./R-arrow/r-arrow", warn.conflicts=FALSE)
 })
 ver = packageVersion("arrow")
 git = ""

R-arrow/setup-R-arrow.sh

@@ -2,5 +2,5 @@
 set -e
 
 # install stable arrow
-mkdir -p ./arrow/r-arrow
+mkdir -p ./R-arrow/r-arrow
 Rscript -e 'install.packages(c("arrow","dplyr"), lib="./R-arrow/r-arrow")'

R-arrow/upg-R-arrow.sh

@@ -3,4 +3,4 @@ set -e
 
 # upgrade all packages in arrow library only if new arrow is out
 echo 'upgrading arrow...'
-Rscript -e 'ap=available.packages(); if (ap["arrow","Version"]!=packageVersion("arrow", lib.loc="./R-arrow/r-arrow")) update.packages(lib.loc="./arrow/r-arrow", ask=FALSE, checkBuilt=TRUE, quiet=TRUE)'
+Rscript -e 'ap=available.packages(); if (ap["arrow","Version"]!=packageVersion("arrow", lib.loc="./R-arrow/r-arrow")) update.packages(lib.loc="./R-arrow/r-arrow", ask=FALSE, checkBuilt=TRUE, quiet=TRUE)'

R-arrow/ver-R-arrow.sh

@@ -1,4 +1,4 @@
 #!/bin/bash
 set -e
 
-Rscript -e 'v=read.dcf(system.file(package="arrow", lib.loc="./R-arrow/r-arrow", "DESCRIPTION"), fields=c("Version","RemoteSha")); colnames(v)[colnames(v)=="RemoteSha"]="Revision"; cnafill=function(x) {x=c(x); x[is.na(x)]=""; x}; fw=function(f, v) writeLines(v, file.path("arrow", f)); invisible(mapply(fw, toupper(colnames(v)), cnafill(v)))'
+Rscript -e 'v=read.dcf(system.file(package="arrow", lib.loc="./R-arrow/r-arrow", "DESCRIPTION"), fields=c("Version","RemoteSha")); colnames(v)[colnames(v)=="RemoteSha"]="Revision"; cnafill=function(x) {x=c(x); x[is.na(x)]=""; x}; fw=function(f, v) writeLines(v, file.path("R-arrow", f)); invisible(mapply(fw, toupper(colnames(v)), cnafill(v)))'

_launcher/launcher.R

@@ -14,7 +14,7 @@ readret = function(x) {
 file.ext = function(x) {
   ans = switch(
     x,
-    "collapse"=, "data.table"=, "dplyr"=, "h2o"=, "arrow"=, "duckdb"="R", "duckdb-latest"="R",
+    "collapse"=, "data.table"=, "dplyr"=, "h2o"=, "R-arrow"=, "duckdb"="R", "duckdb-latest"="R",
     "pandas"=, "spark"=, "pydatatable"=, "modin"=, "dask"=, "datafusion"=, "polars"="py",
     "clickhouse"="sql",
     "juliadf"="jl", "juliads"="jl",

_launcher/solution.R

@@ -110,7 +110,7 @@ if ("quiet" %in% names(args)) {
 file.ext = function(x) {
   ans = switch(
     x,
-    "collapse"=, "data.table"=, "dplyr"=, "h2o"=, "arrow"=, "duckdb"="R", "duckdb-latest"="R",
+    "collapse"=, "data.table"=, "dplyr"=, "h2o"=, "R-arrow"=, "duckdb"="R", "duckdb-latest"="R",
     "pandas"="py", "spark"=, "pydatatable"=, "modin"=, "dask"=, "datafusion"=, "polars"="py",
     "clickhouse"="sql",
     "juliadf"="jl", "juliads"="jl"

_report/report.R

@@ -71,7 +71,7 @@ clean_time = function(d) {
   old_advanced_groupby_questions = c("median v3 sd v3 by id2 id4","max v1 - min v2 by id2 id4","largest two v3 by id2 id4","regression v1 v2 by id2 id4","sum v3 count by id1:id6")
 
   # replace arrow with R-arrow (see https://github.com/duckdblabs/db-benchmark/pull/66)
-  d[solution == "arrow", solution := "R-arrow"]
+  d[which(solution == "arrow"),c("solution")] == "R-arrow"
   d[!nzchar(git), git := NA_character_
     ][,"on_disk" := as.logical(on_disk)
       ][task=="groupby" & solution%in%c("pandas","dask","spark") & batch<1558106628, "out_cols" := NA_integer_

run.sh

@@ -72,7 +72,7 @@ if [[ "$RUN_SOLUTIONS" =~ "h2o" ]]; then ./h2o/ver-h2o.sh; fi;
 if [[ "$DO_UPGRADE" == true && "$RUN_SOLUTIONS" =~ "polars" ]]; then ./polars/upg-polars.sh; fi;
 if [[ "$RUN_SOLUTIONS" =~ "polars" ]]; then ./polars/ver-polars.sh; fi;
 if [[ "$DO_UPGRADE" == true && "$RUN_SOLUTIONS" =~ "R-arrow" ]]; then ./R-arrow/R-upg-arrow.sh; fi;
-if [[ "$RUN_SOLUTIONS" =~ "R-arrow" ]]; then ./R-arrow/R-ver-arrow.sh; fi;
+if [[ "$RUN_SOLUTIONS" =~ "R-arrow" ]]; then ./R-arrow/ver-R-arrow.sh; fi;
 if [[ "$DO_UPGRADE" == true && "$RUN_SOLUTIONS" == "duckdb" ]]; then ./duckdb/upg-duckdb.sh; fi;
 if [[ "$RUN_SOLUTIONS" == "duckdb" ]]; then ./duckdb/ver-duckdb.sh; fi;
 if [[ "$DO_UPGRADE" == true && "$RUN_SOLUTIONS" == "duckdb-latest" ]]; then ./duckdb-latest/setup-duckdb-latest.sh; fi;