From 0e0681676824534589231032090e2035965e1744 Mon Sep 17 00:00:00 2001
From: dricottone <dominic.ricottone@gmail.com>
Date: Thu, 22 Jul 2021 13:09:33 -0400
Subject: [PATCH] Add count_nonascii

---
 count_nonascii.ado | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)
 create mode 100644 count_nonascii.ado

diff --git a/count_nonascii.ado b/count_nonascii.ado
new file mode 100644
index 0000000..5d15b74
--- /dev/null
+++ b/count_nonascii.ado
@@ -0,0 +1,25 @@
+
+program count_nonascii
+	syntax [varlist(string default=none)]
+
+	if "`varlist'"=="" {
+		quietly ds, has(type string)
+		local varlist `r(varlist)'
+	}
+
+	local affected 0
+	local total 0
+
+	foreach v of varlist `varlist' {
+		quietly count if ustrregexm(`v',"[^ -~]")
+		local affected = `affected' + r(N)
+
+		quietly generate int _nonascii = ustrlen(`v') - ustrlen(ustrregexra(`v',"[^ -~]",""))
+		quietly egen _subtotal = total(_nonascii)
+		local total = `total' + _subtotal[1]
+
+		quietly drop _nonascii _subtotal
+	}
+
+	display "There are `total' Unicode characters in `affected' values"
+end
-- 
2.45.2