Sql server count big

NEWS.md

@@ -17,6 +17,8 @@
 
 * `across(everything())` doesn't select grouping columns created via `.by` in
   `summarise()` (@mgirlich, #1493).
+*  Use `COUNT_BIG` instead of `COUNT` for SQL server so that `tally()` and
+  `count()` work regardless of size of the data (@edward-burn, #1498).
 
 *  New translations of clock function `date_count_between()` for SQL server, Redshift, Snowflake, Postgres, and Spark (@edward-burn, #1495).
 

R/backend-mssql.R

@@ -425,6 +425,8 @@ simulate_mssql <- function(version = "15.0") {
   sql_variant(
     mssql_scalar,
     sql_translator(.parent = base_odbc_agg,
+     # https://learn.microsoft.com/en-us/sql/t-sql/functions/count-big-transact-sql?view=sql-server-ver17
+      n           = function() sql("COUNT_BIG(*)"),
       sd            = sql_aggregate("STDEV", "sd"),
       var           = sql_aggregate("VAR", "var"),
       str_flatten = function(x, collapse = "") sql_expr(string_agg(!!x, !!collapse)),

tests/testthat/_snaps/backend-mssql.md

@@ -497,6 +497,38 @@
       FROM `df`
       ORDER BY `y`
 
+# count_big
+
+    Code
+      dplyr::count(mf)
+    Output
+      <SQL>
+      SELECT COUNT_BIG(*) AS `n`
+      FROM `df`
+
+---
+
+    Code
+      dplyr::tally(mf)
+    Output
+      <SQL>
+      SELECT COUNT_BIG(*) AS `n`
+      FROM `df`
+
+# can copy_to() and compute() with temporary tables (#438)
+
+    Code
+      db <- copy_to(con, df, name = unique_table_name(), temporary = TRUE)
+    Message
+      Created a temporary table named #dbplyr_{tmp}
+
+---
+
+    Code
+      db2 <- db %>% mutate(y = x + 1) %>% compute()
+    Message
+      Created a temporary table named #dbplyr_{tmp}
+
 # add prefix to temporary table
 
     Code