From 2e38eb5badd0d949c25e48227daa9b75991472c0 Mon Sep 17 00:00:00 2001
From: James Crake-Merani <James.Crake-Merani@stfc.ac.uk>
Date: Thu, 12 Sep 2024 10:10:44 +0100
Subject: [PATCH] parse unit strs function.

---
 sasdata/quantities/unit_parser.py | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/sasdata/quantities/unit_parser.py b/sasdata/quantities/unit_parser.py
index c53af08..7241e11 100644
--- a/sasdata/quantities/unit_parser.py
+++ b/sasdata/quantities/unit_parser.py
@@ -21,6 +21,17 @@ def parse_single_unit(unit_str: str) -> tuple[Unit | None, str]:
     remaining_str = unit_str[string_pos::]
     return (symbol_lookup[current_unit], remaining_str)
 
+def parse_unit_strs(unit_str: str, current_units: list[Unit] | None=None) -> list[Unit]:
+    if current_units is None:
+        current_units = []
+    if unit_str == '':
+        return current_units
+    parsed_unit, remaining_str = parse_single_unit(unit_str)
+    if not parsed_unit is None:
+        current_units += [parsed_unit]
+    return parse_unit_strs(remaining_str, current_units)
+
+
 # Its probably useful to work out the unit first, and then later work out if a named unit exists for it. Hence why there
 # are two functions.