Fix wrapping of protected links at hyphens

AUTHORS.rst

@@ -40,6 +40,7 @@ The AUTHORS/Contributors are (and/or have been):
 * Edward Ross <gh: EdwardJRoss>
 * Gregory Anders <gh: gpanders>
 * Alex Vandiver <gh: alexmv>
+* sjh9714
 
 Maintainer:
 

ChangeLog.rst

@@ -1,3 +1,10 @@
+Unreleased
+==========
+----
+
+* Fix #425: Keep protected URLs intact during wrapping, including at hyphens.
+  With ``protect_links`` enabled, hyphenated words are also kept together.
+
 2025.4.15
 =========
 ----

docs/usage.md

@@ -68,7 +68,8 @@ simple indications of their function.
     - BODY_WIDTH for wrapping long lines
     - SKIP_INTERNAL_LINKS to skip #local-anchor things
     - INLINE_LINKS for formatting images and links
-    - PROTECT_LINKS protect from line breaks
+    - PROTECT_LINKS protect URLs from line breaks by wrapping only at whitespace
+      (also keeps hyphenated words together)
     - GOOGLE_LIST_INDENT no of pixels to indent nested lists
     - IGNORE_ANCHORS
     - IGNORE_IMAGES

html2text/__init__.py

@@ -993,6 +993,7 @@ def optwrap(self, text: str) -> str:
                         para,
                         self.body_width,
                         break_long_words=False,
+                        break_on_hyphens=not self.protect_links,
                         subsequent_indent=indent,
                     )
                     result += "\n".join(wrapped)

test/protect_links.html

@@ -1 +1,3 @@
-<a href="http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html">foo</a>
+<p><a href="http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html">foo</a></p>
+<p>Before the link <a href="http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html">foo</a> The following text should wrap normally after the protected link, keeping every word in order.</p>
+<p><a href="http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html?query=another-long-hyphenated-value">http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html?query=another-long-hyphenated-value</a></p>

test/protect_links.md

@@ -1,2 +1,8 @@
-[foo](<http://im-a-very-very-very-very-very-very-very-very-very-very-
-long/link.html>)
+[foo](<http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html>)
+
+Before the link
+[foo](<http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html>)
+The following text should wrap normally after the protected link, keeping
+every word in order.
+
+<http://im-a-very-very-very-very-very-very-very-very-very-very-long/link.html?query=another-long-hyphenated-value>

test/test_html2text.py

@@ -254,3 +254,16 @@ def test_strong_emptied() -> None:
     h.strong_mark = ""
     string = "A <b>B</b> <i>C</i>."
     assert h.handle(string) == "A B _C_.\n\n"
+
+
+@pytest.mark.parametrize(
+    "protect_links,expected",
+    [
+        (False, "alpha-\nbeta-gamma\n\n"),
+        (True, "alpha-beta-gamma\n\n"),
+    ],
+)
+def test_hyphen_wrapping_with_protect_links(protect_links, expected):
+    h = html2text.HTML2Text(bodywidth=10)
+    h.protect_links = protect_links
+    assert h.handle("alpha-beta-gamma") == expected