diff options
| author | Leonard Richardson <leonardr@segfault.org> | 2014-12-10 19:06:33 -0500 |
|---|---|---|
| committer | Leonard Richardson <leonardr@segfault.org> | 2014-12-10 19:06:33 -0500 |
| commit | a1d16f0b14a9f08f54677d933b4ddb8e32ffabb1 (patch) | |
| tree | c7889c0245b47aea083bdec279afb2c0194ca644 /bs4 | |
| parent | 5048db4ab06f6065d8d364500b38631456b58690 (diff) | |
The warning when you pass in a filename or URL as markup will now be
displayed correctly even if the filename or URL is a Unicode
string. [bug=1268888]
Diffstat (limited to 'bs4')
| -rw-r--r-- | bs4/__init__.py | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/bs4/__init__.py b/bs4/__init__.py index b74acee..a53048d 100644 --- a/bs4/__init__.py +++ b/bs4/__init__.py @@ -186,6 +186,8 @@ 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) if markup[:5] == "http:" or markup[:6] == "https:": @@ -193,6 +195,8 @@ class BeautifulSoup(Tag): # Python 3 otherwise. if ((isinstance(markup, bytes) and not b' ' in markup) or (isinstance(markup, unicode) and not u' ' in markup)): + if isinstance(markup, unicode): + markup = markup.encode("utf8") warnings.warn( '"%s" looks like a URL. Beautiful Soup is not an HTTP client. You should probably use an HTTP client to get the document behind the URL, and feed that document to Beautiful Soup.' % markup) |