Support equeation and table head in markdown #1

Author: selfboot

examples/insert_text.ipynb

@@ -48,7 +48,7 @@
     "notion = Client(auth=notion_api_key)\n",
     "parent = {\"type\": \"database_id\", \"database_id\": database_id}\n",
     "properties = {\"Title\": {\"title\": [{\"text\": {\"content\": \"Page with multi text\"}}]}}\n",
-    "text_content = \"Some words\" * 200 + \"Any\"\n",
+    "text_content = \"Some words\" \n",
     "text_block = {\n",
     "    \"type\": \"text\",\n",
     "    \"text\": {\n",
@@ -66,11 +66,18 @@
     "    \"href\": None\n",
     "}\n",
     "\n",
+    "equation = {\n",
+    "  \"type\": \"equation\",\n",
+    "  \"equation\": {\n",
+    "    \"expression\": \"e=mc^2\"\n",
+    "  }\n",
+    "}\n",
+    "\n",
     "children = [{\n",
     "    \"object\": \"block\",\n",
     "    \"type\": \"paragraph\",\n",
     "    \"paragraph\": {\n",
-    "        \"rich_text\": [text_block]\n",
+    "        \"rich_text\": [text_block, equation]\n",
     "    }\n",
     "}]\n",
     "\n",
@@ -84,35 +91,6 @@
     "created_page = cast(Dict[str, Any], created_page)\n",
     "print(f'page_id = {created_page[\"id\"]}')"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "text_content = \"Some words\" * 200 + \"Any\"\n",
-    "from html2notion.translate.html2json_base import Html2JsonBase\n",
-    "\n",
-    "text_obj = Html2JsonBase.generate_text(plain_text=text_content)\n",
-    "children_2 = [{\n",
-    "    \"object\": \"block\",\n",
-    "    \"type\": \"paragraph\",\n",
-    "    \"paragraph\": {\n",
-    "        \"rich_text\": [text_obj]\n",
-    "    }\n",
-    "}]\n",
-    "\n",
-    "created_page_2 = notion.pages.create(\n",
-    "    parent=parent,\n",
-    "    properties=properties,\n",
-    "    children=children_2\n",
-    ")\n",
-    "\n",
-    "from typing import Dict, Any, cast\n",
-    "created_page_2 = cast(Dict[str, Any], created_page)\n",
-    "print(f'page_id = {created_page_2[\"id\"]}')"
-   ]
   }
  ],
  "metadata": {

html2notion/translate/html2json_base.py

@@ -160,7 +160,6 @@ def generate_inline_obj(self, tag: PageElement):
         return res_obj
 
 
-    
     def generate_link(self, **kwargs):
         plain_text = kwargs.get("plain_text", "")
         if not plain_text:
@@ -425,7 +424,7 @@ def _convert_one_list_item(self, soup, list_type):
     """
     <div>
     <div><br /></div>
-    <table> <tbody> <tr> <td> </td> </tr> </tbody>
+    <table> <thead> </thead><tbody> <tr> <td> </td> </tr> </tbody> </table>
     <div><br /></div>
     </div>
     """
@@ -438,10 +437,12 @@ def convert_table(self, soup):
             return
 
         table_width = len(tr_tags[0].find_all('td'))
+        has_header = False
         for tr in tr_tags:
             td_tags = tr.find_all('td')
             if not td_tags:
-                continue
+                td_tags = tr.find_all('th')
+                has_header = True
             table_width = max(table_width, len(td_tags))
             one_row = {
                 "type": "table_row",
@@ -457,7 +458,7 @@ def convert_table(self, soup):
         table_obj = {
             "table": {
                 "has_row_header": False,
-                "has_column_header": False,
+                "has_column_header": has_header,
                 "table_width": table_width,
                 "children": table_rows,
             }

html2notion/translate/html2json_markdown.py

@@ -130,7 +130,7 @@ def convert_code(self, soup):
 
         css_dict = Html2JsonBase.get_tag_style(code_tag)
         language = css_dict.get('language', 'plain text')
-        json_obj["code"]["language"] = language
+        json_obj["code"]["language"] = Html2JsonBase.get_valid_language(language)
         json_obj["code"]["rich_text"] = self.merge_rich_text(rich_text)
         return json_obj
 
@@ -148,6 +148,24 @@ def convert_quote(self, soup):
             rich_text.extend(text_obj)
         return json_obj
 
+    def convert_equation(self, soup: Tag):
+        json_obj = {
+            "object": "block",
+            "type": "paragraph",
+            "paragraph": {
+                "rich_text": []
+            }
+        }
+        expression = soup.get_text()
+        equation = json_obj["paragraph"]["rich_text"]
+        equation.append({
+            "type": "equation",
+            "equation": {
+                "expression": expression
+            }
+        })
+        return json_obj
+    
     def convert_to_do(self, soup: Tag):
         li_tags = soup.find_all('li', recursive=True)
         childs = li_tags if li_tags else [soup]
@@ -183,7 +201,7 @@ def _is_checkbox(self, soup):
     def _extract_code_blocks(self):
         code_pattern = re.compile(r'```(\w+)?\n(.*?)```', re.DOTALL)
         matches = code_pattern.findall(self.markdown)
-        code_blocks = [{'language': Html2JsonBase.get_valid_language(match[0]), 'code': match[1]} for match in matches]
+        code_blocks = [{'language': match[0], 'code': match[1].rstrip('\n')} for match in matches]
         return code_blocks
 
     def _replace_pre_code(self, soup):