Add initial legacy Presto/Athena compatibility udfs (#316)

* Add initial legacy Presto/Athena compatibility udfs

* fixup! Add initial legacy Presto/Athena compatibility udfs

README.md

@@ -53,6 +53,8 @@ Recommended practices
   - Each file must define a single function using `CREATE TEMP FUNCTION` syntax
   - SQL UDFs must be defined in the `udf/` directory and JS UDFs must be defined
     in the `udf_js` directory
+    - The `udf_legacy/` directory is an exception which should only contain
+      compatibility functions for queries migrated from Athena/Presto.
   - The function must be named as `<dir_name>_<file_name_without_suffix>`
     so `udf/mode_last.sql` must define a function `udf_mode_last`
 - Must be defined as temporary UDFs

udf_legacy/README.md

@@ -0,0 +1,4 @@
+Legacy UDFs
+===
+
+This directory contains compatibility functions for query migrations from Athena/Presto, and is named `udf_legacy` to discourage their ongoing use.

udf_legacy/contains.sql

@@ -0,0 +1,16 @@
+/*
+
+Returns true if the array arr contains the element el
+
+*/
+CREATE TEMP FUNCTION
+  udf_legacy_contains(arr ANY TYPE, el ANY TYPE)
+  RETURNS BOOLEAN
+  AS (
+    el IN UNNEST(arr)
+  );
+
+-- Tests
+
+assert_true(udf_legacy_contains([1, 2, 3], 1)),
+asset_false(udf_legacy_contains([1, 2, 3], 5))

udf_legacy/date_format.sql

@@ -0,0 +1,15 @@
+/*
+
+Shim for Presto/Athena's DATE_FORMAT to BigQuery's FORMAT_DATE
+
+Note that the format strings for FORMAT_DATE vs DATE_FORMAT are not
+identical, but usages found in existing scheduled queries had format
+strings that were portable between the two
+
+*/
+CREATE TEMP FUNCTION
+  udf_legacy_date_format(d DATE, f STRING)
+  RETURNS STRING
+  AS (
+    FORMAT_DATE(f, d)
+  );

udf_legacy/to_iso8601.sql

@@ -0,0 +1,16 @@
+/*
+
+Takes in a DATE type and returns a `YYYY-MM-DD` formatted string. This
+function does _not_ attempt to replicate the entirety of the
+functionality of Presto/Athena's `to_iso8601()` (which takes in a DATE,
+DATETIME or TIMESTAMP type and returns either an iso8601-formatted date
+string or a date-time string depending on the input type) since all use
+cases found in legacy scheduled queries were for the DATE case.
+
+*/
+CREATE TEMP FUNCTION
+  udf_legacy_to_iso8601(x DATE)
+  RETURNS STRING
+  AS (
+    FORMAT_DATE('%F', x)
+  );