From e8a1603f809138b645694f9280cd9eff26b6dfed Mon Sep 17 00:00:00 2001
From: Nigel <windymilla@users.noreply.github.com>
Date: Sun, 9 Aug 2026 16:44:25 +0100
Subject: [PATCH] Improve edge case for WF Hyphen search

Find "school teacher" in "Sunday-school teacher" when
clicking on WF hyphens report.

Fixes #1922
---
 src/guiguts/word_frequency.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/guiguts/word_frequency.py b/src/guiguts/word_frequency.py
index 11563201..26dfe6f9 100644
--- a/src/guiguts/word_frequency.py
+++ b/src/guiguts/word_frequency.py
@@ -785,14 +785,18 @@ def goto_word(
             # Also, boundary for an emdash is any non-emdash, but for other non-word characters the
             # boundary is a word/space character
             emdash_bound = "—" if newline_word[0] == "—" else r"[^\w\s]"
+            # If word contains space, then permit boundary character to be a hyphen
+            # (by not having it in the excluded characters)
+            # e.g. "school teacher" will match "Sunday-school teacher"
+            hyph_bound = "" if " " in newline_word else "-"
             left_boundary = (
-                r"(?<!([^\W_]|[\p{L}\p{N}][-'’]))"
+                r"(?<!([^\W_]|[\p{L}\p{N}][" + hyph_bound + r"'’]))"
                 if newline_word[0].isalnum()
                 else rf"(?<!{emdash_bound})"
             )
             emdash_bound = "—" if newline_word[-1] == "—" else r"[^\w\s]"
             right_boundary = (
-                r"(?!([^\W_]|[-'’][\p{L}\p{N}]))"
+                r"(?!([^\W_]|[" + hyph_bound + r"'’][\p{L}\p{N}]))"
                 if newline_word[-1].isalnum()
                 else rf"(?!{emdash_bound})"
             )