Fix: Markdown table extractor (#11018)

### What problem does this PR solve?

Now markdown table extractor supports <table ...>. #10966 

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

deepdoc/parser/markdown_parser.py

@@ -70,6 +70,17 @@ class RAGFlowMarkdownParser:
             )
             working_text = replace_tables_with_rendered_html(no_border_table_pattern, tables)
 
+        # Replace any TAGS e.g. <table ...> to <table>
+        TAGS = ["table", "td", "tr", "th", "tbody", "thead", "div"]
+        table_with_attributes_pattern = re.compile(
+            rf"<(?:{'|'.join(TAGS)})[^>]*>", re.IGNORECASE
+        )
+        def replace_tag(m):
+            tag_name = re.match(r"<(\w+)", m.group()).group(1)
+            return "<{}>".format(tag_name)
+
+        working_text = re.sub(table_with_attributes_pattern, replace_tag, working_text)
+
         if "<table>" in working_text.lower():  # for optimize performance
             # HTML table extraction - handle possible html/body wrapper tags
             html_table_pattern = re.compile(

rag/app/naive.py

@@ -509,11 +509,11 @@ class Markdown(MarkdownParser):
                 txt = f.read()
 
         remainder, tables = self.extract_tables_and_remainder(f'{txt}\n', separate_tables=separate_tables)
-
+        # To eliminate duplicate tables in chunking result, uncomment code below and set separate_tables to True in line 410.
+        # extractor = MarkdownElementExtractor(remainder)
         extractor = MarkdownElementExtractor(txt)
         element_sections = extractor.extract_elements(delimiter)
         sections = [(element, "") for element in element_sections]
-
         tbls = []
         for table in tables:
             tbls.append(((None, markdown(table, extensions=['markdown.extensions.tables'])), ""))