feat: use new spec to represent title and paragraph

drunkpig · drunkpig · commit fe4adac274cb · 2025-11-24T15:48:40.000+08:00
diff --git a/llm_web_kit/extractor/html/recognizer/text.py b/llm_web_kit/extractor/html/recognizer/text.py
@@ -93,7 +93,7 @@ def to_content_list_node(self, base_url: str, parsed_content: HtmlElement, raw_h
         el = parsed_content
         node = {
             'type': DocElementType.PARAGRAPH,
-            'raw_content': raw_html_segment,
+            # 'raw_content': raw_html_segment,
             'content': json.loads(el.text),
         }
         return node
diff --git a/llm_web_kit/extractor/html/recognizer/title.py b/llm_web_kit/extractor/html/recognizer/title.py
@@ -36,7 +36,7 @@ def to_content_list_node(self, base_url: str, parsed_content: HtmlElement, raw_h
             return None
         cctitle_content_node = {
             'type': DocElementType.TITLE,
-            'raw_content': raw_html_segment,
+            # 'raw_content': raw_html_segment,
             'content': {
                 'title_content': text,
                 'level': level

Original file line number	Diff line number	Diff line change
`@@ -93,7 +93,7 @@ def to_content_list_node(self, base_url: str, parsed_content: HtmlElement, raw_h`
`93`	`93`	`el = parsed_content`
`94`	`94`	`node = {`
`95`	`95`	`'type': DocElementType.PARAGRAPH,`
`96`		`- 'raw_content': raw_html_segment,`
	`96`	`+ # 'raw_content': raw_html_segment,`
`97`	`97`	`'content': json.loads(el.text),`
`98`	`98`	`}`
`99`	`99`	`return node`