Fix strip existing ocr bug

marker/providers/pdf.py

@@ -287,7 +287,7 @@ def check_page(self, page_id: int, doc: PdfDocument) -> bool:
 
                 # we also skip pages without embedded fonts and fonts without names
                 non_embedded_fonts.append(pdfium_c.FPDFFont_GetIsEmbedded(font) == 0)
-                empty_fonts.append(not font_name or font_name == "GlyphLessFont")
+                empty_fonts.append("glyphless" in font_name.lower()) # Add font name check back in when we bump pypdfium2
                 if font_name not in font_map:
                     font_map[font_name or 'Unknown'] = font
 
@@ -363,7 +363,7 @@ def _get_fontname(font) -> str:
                 font_name_buffer = ctypes.create_string_buffer(length)
                 pdfium_c.FPDFFont_GetBaseFontName(font, font_name_buffer, length)
                 font_name = font_name_buffer.value.decode("utf-8")
-        except:
+        except Exception as e:
             pass
 
         return font_name

marker/scripts/streamlit_app.py

@@ -41,7 +41,7 @@
 ]
 
 with open(
-    os.path.join(os.path.dirname(__file__), "streamlit_app_blocks_viz.html")
+    os.path.join(os.path.dirname(__file__), "streamlit_app_blocks_viz.html"), encoding="utf-8"
 ) as f:
     BLOCKS_VIZ_TMPL = string.Template(f.read())
 
@@ -179,7 +179,7 @@ def block_display(image: Image, blocks: dict | None = None, dpi=96):
 Find the project [here](https://github.com/VikParuchuri/marker).
 """)
 
-in_file: UploadedFile = st.sidebar.file_uploader("PDF or image file:", type=["pdf", "png", "jpg", "jpeg", "gif"])
+in_file: UploadedFile = st.sidebar.file_uploader("PDF, document, or image file:", type=["pdf", "png", "jpg", "jpeg", "gif", "pptx", "docx", "xlsx", "html", "epub"])
 
 if in_file is None:
     st.stop()

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "marker-pdf"
-version = "1.6.0"
+version = "1.6.1"
 description = "Convert documents to markdown with high speed and accuracy."
 authors = ["Vik Paruchuri <[email protected]>"]
 readme = "README.md"