Fixed a bug that happened when passing a Unicode filename containing

  non-ASCII characters as markup into Beautiful Soup, on a system that
  allows Unicode filenames. [bug=1866717]

1 files changed, 20 insertions, 10 deletions

diff --git a/bs4/__init__.py b/bs4/__init__.py
index 3cd2e15..f828cd2 100644
--- a/bs4/__init__.py
+++ b/bs4/__init__.py
@@ -306,12 +306,11 @@ class BeautifulSoup(Tag):
                 # system. Just let it go.
                 pass
             if is_file:
-                if isinstance(markup, unicode):
-                    markup = markup.encode("utf8")
                 warnings.warn(
                     '"%s" looks like a filename, not markup. You should'
                     ' probably open this file and pass the filehandle into'
-                    ' Beautiful Soup.' % markup)
+                    ' Beautiful Soup.' % self._decode_markup(markup)
+                )
             self._check_markup_is_url(markup)
 
         rejections = []
@@ -360,8 +359,21 @@ class BeautifulSoup(Tag):
             d['builder'] = None
         return d
 
-    @staticmethod
-    def _check_markup_is_url(markup):
+    @classmethod
+    def _decode_markup(cls, markup):
+        """Ensure `markup` is bytes so it's safe to send into warnings.warn.
+
+        TODO: warnings.warn had this problem back in 2010 but it might not
+        anymore.
+        """
+        if isinstance(markup, bytes):
+            decoded = markup.decode('utf-8', 'replace')
+        else:
+            decoded = markup
+        return decoded
+
+    @classmethod
+    def _check_markup_is_url(cls, markup):
         """Error-handling method to raise a warning if incoming markup looks
         like a URL.
 
@@ -378,15 +390,13 @@ class BeautifulSoup(Tag):
 
         if any(markup.startswith(prefix) for prefix in cant_start_with):
             if not space in markup:
-                if isinstance(markup, bytes):
-                    decoded_markup = markup.decode('utf-8', 'replace')
-                else:
-                    decoded_markup = markup
                 warnings.warn(
                     '"%s" looks like a URL. Beautiful Soup is not an'
                     ' HTTP client. You should probably use an HTTP client like'
                     ' requests to get the document behind the URL, and feed'
-                    ' that document to Beautiful Soup.' % decoded_markup
+                    ' that document to Beautiful Soup.' % cls._decode_markup(
+                        markup
+                    )
                 )
 
     def _feed(self):