Fix span id issue

VikParuchuri · Jan 24, 2025 · 2f14c52 · 2f14c52
1 parent d3c43d6
commit 2f14c52
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 4 deletions.
diff --git a/marker/renderers/markdown.py b/marker/renderers/markdown.py
@@ -168,11 +168,13 @@ def convert_table(self, el, text, convert_as_inline):
     def convert_a(self, el, text, convert_as_inline):
         text = self.escape(text)
         text = re.sub(r"([\[\]])", r"\\\1", text)
-        return super().convert_a(el, self.escape(text), convert_as_inline)
+        return super().convert_a(el, text, convert_as_inline)
 
     def convert_span(self, el, text, convert_as_inline):
-        return f'<span id="{el["id"]}"/>'
-
+        if el.get("id"):
+            return f'<span id="{el["id"]}">{text}</span>'
+        else:
+            return text
 
 class MarkdownOutput(BaseModel):
     markdown: str

diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "marker-pdf"
-version = "1.3.1"
+version = "1.3.2"
 description = "Convert PDF to markdown with high speed and accuracy."
 authors = ["Vik Paruchuri <[email protected]>"]
 readme = "README.md"