Fix superscripts

VikParuchuri · Feb 26, 2025 · e2a2861 · e2a2861
1 parent f8a3c8c
commit e2a2861
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 3 deletions.
diff --git a/marker/providers/pdf.py b/marker/providers/pdf.py
@@ -205,10 +205,14 @@ def pdftext_extraction(self, doc: PdfDocument) -> ProviderPageLines:
                         font_size = span["font"]["size"] or 0
                         polygon = PolygonBox.from_bbox(span["bbox"], ensure_nonzero_area=True)
                         span_chars = [Char(char=c['char'], polygon=PolygonBox.from_bbox(c['bbox'], ensure_nonzero_area=True), char_idx=c['char_idx']) for c in span["chars"]]
+                        superscript = span.get("superscript", False)
+                        text = self.normalize_spaces(fix_text(span["text"]))
+                        if superscript:
+                            text = text.strip()
                         spans.append(
                             SpanClass(
                                 polygon=polygon,
-                                text=self.normalize_spaces(fix_text(span["text"])),
+                                text=text,
                                 font=font_name,
                                 font_weight=font_weight,
                                 font_size=font_size,
@@ -218,7 +222,7 @@ def pdftext_extraction(self, doc: PdfDocument) -> ProviderPageLines:
                                 page_id=page_id,
                                 text_extraction_method="pdftext",
                                 url=span.get("url"),
-                                has_superscript=span.get("superscript", False),
+                                has_superscript=superscript,
                             )
                         )
                         chars.append(span_chars)

diff --git a/marker/schema/text/span.py b/marker/schema/text/span.py
@@ -66,7 +66,6 @@ def assemble_html(self, document, child_blocks, parent_structure):
 
             # Handle full block superscript
             if "<sup>" not in text:
-                print(text)
                 text = f"<sup>{text}</sup>"
 
         if self.url: