Don't validate <uwaga> contents

[librarian.git] / src / librarian / pdf.py

diff --git a/src/librarian/pdf.py b/src/librarian/pdf.py
index 31dfe1e..b32395f 100644 (file)
--- a/src/librarian/pdf.py
+++ b/src/librarian/pdf.py
@@ -425,7 +425,7 @@ def load_including_children(wldoc=None, provider=None, uri=None):
     """
 
     if uri and provider:
-        f = provider.by_uri(uri)
+        f = provider.by_slug(uri.slug)
         text = f.read().decode('utf-8')
         f.close()
     elif wldoc is not None:
@@ -436,7 +436,10 @@ def load_including_children(wldoc=None, provider=None, uri=None):
             'Neither a WLDocument, nor provider and URI were provided.'
         )
 
+    # Cyrrilic
     text = re.sub(r"([\u0400-\u04ff]+)", r"<alien>\1</alien>", text)
+    # Geometric shapes.
+    text = re.sub(r"([\u25a0-\u25ff]+)", r"<alien>\1</alien>", text)
 
     document = WLDocument.from_bytes(text.encode('utf-8'),
                                      parse_dublincore=True, provider=provider)