From 31a6527fe830734191a7fc9ff78b6c0a5140688f Mon Sep 17 00:00:00 2001
From: Charlotte Koch <charlotte@magentastripe.com>
Date: Sun, 6 Apr 2025 13:04:30 -0700
Subject: More nuanced way of handling ellipses

---
 script/unicodify.sed | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/script/unicodify.sed b/script/unicodify.sed
index 4668732..138003e 100644
--- a/script/unicodify.sed
+++ b/script/unicodify.sed
@@ -1,6 +1,12 @@
-# Remove spaces on either end of an em-dash or ellipsis.
-s,[[:space:]]--[[:space:]],\&\#8212;,g
-s,[[:space:]]\.\.\.[[:space:]],\&\#8230;,g
+# Remove Asciidoc comments now, for the sake of getting more accurate
+# wordcounts.
+s,//.*,,g
+
+# Remove spaces on either end of an em-dash. 
+s,[[:space:]]*--[[:space:]]*,\&\#8212;,g
+
+# Remove spaces before an ellipsis, while ensuring one space after.
+s,[[:space:]]*\.\.\.[[:space:]]*,\&\#8230;\&nbsp;,g
 
 # Explicitly handle curly double quotes before curly single quotes.
 s,"`,\&\#8220;,g
@@ -14,6 +20,7 @@ s,\&egrave;,\&\#232;,g
 s,\&eacute;,\&\#233;,g
 s,\&iuml;,\&\#239;,g
 
-# Remove Asciidoc comments now, for the sake of getting more accurate
-# wordcounts.
-s,//.*,,g
+# Remove spaces before a close-quote, which might have accidentally been
+# introduced while converting ellipses earlier.
+s,[[:space:]]*\&\#8221;,\&\#8221;,g
+s,\&nbsp;\&\#8221;,\&\#8221;,g
-- 
cgit v1.2.3