More robust detection of string concatenation for BigQuery

NEWS.md

@@ -7,6 +7,8 @@ Bugfixes:
 
 2. Fixed translation of `ALTER TABLE ALTER COLUMN` on PostgreSQL.
 
+3. More robust detection of string concatenation for BigQuery.
+
 
 SqlRender 1.17.0
 ================

inst/csv/replacementPatterns.csv

@@ -726,15 +726,24 @@ bigquery,"DATEADD(yy,@years,@date)","DATE_ADD(@date, INTERVAL @years YEAR)"
 bigquery,"DATEADD(yyyy,@years,@date)","DATE_ADD(@date, INTERVAL @years YEAR)"
 bigquery,"DATEADD(year,@years,@date)","DATE_ADD(@date, INTERVAL @years YEAR)"
 bigquery,INTERVAL @(-?[0-9]+)a.0,INTERVAL @a
-bigquery,@(([a-z]+\(.*\))|([a-z0-9_]+))a + '@b',"CONCAT(@a, '@b')"
-bigquery,'@a' + @(([a-z]+\(.*\))|([a-z0-9_]+))b,"CONCAT('@a', @b)"
-bigquery,CAST(@a AS VARCHAR(@b)) + @(([a-z]+\(.*\))|([a-z0-9_]+))c,"CONCAT(CAST(@a AS VARCHAR(@b)),@c)"
-bigquery,@(([a-z]+\(.*\))|([a-z0-9_]+))a + CAST(@b AS VARCHAR(@c)),"CONCAT(@a, CAST(@b AS VARCHAR(@c))"
-bigquery,CAST(@a AS VARCHAR) + @(([a-z]+\(.*\))|([a-z0-9_]+))b,"CONCAT(CAST(@a AS VARCHAR), @b)"
-bigquery,@(([a-z]+\(.*\))|([a-z0-9_]+))a + CAST(@b AS VARCHAR),"CONCAT(@a, CAST(@b AS VARCHAR))"
-bigquery,"CONCAT(@a, @b) + @(([a-z]+\(.*\))|([a-z0-9_]+))c","CONCAT(@a, @b, @c)"
-bigquery,"@(([a-z]+\(.*\))|([a-z0-9_]+))a + CONCAT(@b, @c)","CONCAT(@a, @b, @c)"
+bigquery,CAST(@a AS VARCHAR) + @b(@c),"CONCAT(CAST(@a AS VARCHAR), @b(@c))"
+bigquery,@([a-z]+)a(@b) + CAST(@c AS VARCHAR),"CONCAT(@a(@b), CAST(@c AS VARCHAR))"
+bigquery,CAST(@a AS VARCHAR(@n)) + @b(@c),"CONCAT(CAST(@a AS VARCHAR(@n)), @b(@c))"
+bigquery,@([a-z]+)a(@b) + CAST(@c AS VARCHAR(@n)),"CONCAT(@a(@b), CAST(@c AS VARCHAR(@n)))"
+bigquery,'@a' + @b(@c),"CONCAT('@a', @b(@c))"
+bigquery,@([a-z]+)a(@b) + '@c',"CONCAT(@a(@b), '@c')"
+bigquery,'@a' + @b FROM,"CONCAT('@a', @b) FROM"
+bigquery,@([a-z0-9_]+)a + '@b',"CONCAT(@a, '@b')"
+bigquery,CAST(@a AS VARCHAR) + @b FROM,"CONCAT(CAST(@a AS VARCHAR), @b) FROM"
+bigquery,@([a-z0-9_]+)a + CAST(@b AS VARCHAR),"CONCAT(@a, CAST(@b AS VARCHAR))"
+bigquery,CAST(@a AS VARCHAR(@n)) + @b FROM,"CONCAT(CAST(@a AS VARCHAR(@n)), @b) FROM"
+bigquery,@([a-z0-9_]+)a + CAST(@b AS VARCHAR(@n)),"CONCAT(@a, CAST(@b AS VARCHAR(@n)))"
+bigquery,CONCAT(@a) + @b(@c),"CONCAT(@a, @b(@c))"
+bigquery,@([a-z]+)a(@b) + CONCAT(@c),"CONCAT(@a(@b), @c)"
+bigquery,CONCAT(@a) + @b FROM,"CONCAT(@a, @b) FROM"
+bigquery,@([a-z0-9_]+)a + CONCAT(@b),"CONCAT(@a, @b)"
 bigquery,"CONCAT(@a, CONCAT(@b, @c))","CONCAT(@a, @b, @c)"
+bigquery,"CONCAT(CONCAT(@a, @b), @c)","CONCAT(@a, @b, @c)"
 bigquery,"CONCAT(CONCAT(@a, @b, @c))","CONCAT(@a, @b, @c)"
 bigquery,"STDEV(@a)","STDDEV(@a)"
 bigquery,"HASHBYTES('MD5',@a)","md5(@a)"

tests/testthat/test-translate-bigquery.R

@@ -396,11 +396,6 @@ test_that("translate sql server -> bigquery DATEFROMPARTS", {
   expect_equal_ignore_spaces(sql, "select DATE(2019,1,30)")
 })
 
-# test_that("translate sql server -> bigquery offset literal", {
-#   sql <- translate("create table test_that (\"offset\" STRING);", targetDialect = "bigquery")
-#   expect_equal_ignore_spaces(sql, "create table test_that (offset STRING);")
-# })
-
 test_that("translate sql server -> bigquery EOMONTH()", {
   sql <- translate("select eomonth(payer_plan_period_start_date)", targetDialect = "bigquery")
   expect_equal_ignore_spaces(
@@ -467,7 +462,7 @@ test_that("translate sql server -> String concatenation", {
   )
   expect_equal_ignore_spaces(
     sql,
-    "select CONCAT(last_name, ', ', first_name) from my_table;"
+    "select CONCAT(last_name, ', ', first_name) FROM my_table;"
   )
 })
 
@@ -477,7 +472,7 @@ test_that("translate sql server -> String concatenation", {
   )
   expect_equal_ignore_spaces(
     sql,
-    "select CONCAT(first_name, CAST(middle_initial AS STRING), last_name) from my_table;"
+    "select CONCAT(first_name, CAST(middle_initial AS STRING), last_name) FROM my_table;"
   )
 })
 
@@ -487,7 +482,7 @@ test_that("translate sql server -> String concatenation", {
   )
   expect_equal_ignore_spaces(
     sql,
-    "select CONCAT(first_name, CAST(middle_initial AS STRING), last_name) from my_table;"
+    "select CONCAT(first_name, CAST(middle_initial AS STRING), last_name) FROM my_table;"
   )
 })
 
@@ -497,7 +492,7 @@ test_that("translate sql server -> String concatenation", {
   )
   expect_equal_ignore_spaces(
     sql,
-    "select subgroup_id, CONCAT('Persons aged ', CONCAT(cast(age_low as STRING), 'to ', cast(age_high as STRING), 'with gender = '), gender_name) from subgroups;"
+    "select subgroup_id, CONCAT('Persons aged ', CAST(age_low  AS STRING), CONCAT('to ', cast(age_high as STRING)), 'with gender = ', gender_name ) FROM subgroups;"
   )
 })
 
@@ -549,3 +544,9 @@ test_that("translate sql server -> bigquery quotes", {
   sql <- translate("SELECT \"a\" from t;", targetDialect = "bigquery")
   expect_equal_ignore_spaces(sql, "select `a` from t;")
 })
+
+test_that("translate sql server -> bigquery RIGHT with implicit concat", {
+  sql <- translate("RIGHT('0' + CAST(p.month_of_birth AS VARCHAR), 2)", targetDialect = "bigquery")
+  expect_equal_ignore_spaces(sql, "SUBSTR(CONCAT('0', cast(p.month_of_birth as STRING)),-2)")
+})
+