notionary 0.1.8__tar.gz → 0.1.10__tar.gz

{notionary-0.1.8 → notionary-0.1.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: notionary
-Version: 0.1.8
+Version: 0.1.10
 Summary: A toolkit to convert between Markdown and Notion blocks
 Home-page: https://github.com/mathisarends/notionary
 Author: Mathis Arends

notionary-0.1.10/notionary/core/converters/elements/embed_element.py

@@ -0,0 +1,129 @@
+import re
+from typing import Dict, Any, Optional, List
+from notionary.core.converters.elements.notion_block_element import NotionBlockElement
+
+
+class EmbedElement(NotionBlockElement):
+    """
+    Handles conversion between Markdown embeds and Notion embed blocks.
+
+    Markdown embed syntax (custom format):
+    - <embed:Caption>(https://example.com) - Basic embed with caption
+    - <embed>(https://example.com) - Embed without caption
+
+    Supports various URL types including websites, PDFs, Google Maps, Google Drive,
+    Twitter/X posts, and other sources that Notion can embed.
+    """
+
+    PATTERN = re.compile(
+        r"^<embed(?:\:(.*?))?>(?:\s*)"
+        + r'\((https?://[^\s"]+)'
+        + r"\)$"
+    )
+
+    @staticmethod
+    def match_markdown(text: str) -> bool:
+        """Check if text is a markdown embed."""
+        text = text.strip()
+        return text.startswith("<embed") and bool(EmbedElement.PATTERN.match(text))
+
+    @staticmethod
+    def match_notion(block: Dict[str, Any]) -> bool:
+        """Check if block is a Notion embed."""
+        return block.get("type") == "embed"
+
+    @staticmethod
+    def markdown_to_notion(text: str) -> Optional[Dict[str, Any]]:
+        """Convert markdown embed to Notion embed block."""
+        embed_match = EmbedElement.PATTERN.match(text.strip())
+        if not embed_match:
+            return None
+
+        caption = embed_match.group(1) or ""
+        url = embed_match.group(2)
+
+        if not url:
+            return None
+
+        # Prepare the embed block
+        embed_block = {
+            "type": "embed",
+            "embed": {"url": url},
+        }
+
+        # Add caption if provided
+        if caption:
+            embed_block["embed"]["caption"] = [
+                {"type": "text", "text": {"content": caption}}
+            ]
+
+        return embed_block
+
+    @staticmethod
+    def notion_to_markdown(block: Dict[str, Any]) -> Optional[str]:
+        """Convert Notion embed block to markdown embed."""
+        if block.get("type") != "embed":
+            return None
+
+        embed_data = block.get("embed", {})
+        url = embed_data.get("url", "")
+
+        if not url:
+            return None
+
+        # Extract caption if available
+        caption = ""
+        caption_rich_text = embed_data.get("caption", [])
+        if caption_rich_text:
+            caption = EmbedElement._extract_text_content(caption_rich_text)
+
+        if caption:
+            return f"<embed:{caption}>({url})"
+        else:
+            return f"<embed>({url})"
+
+    @staticmethod
+    def is_multiline() -> bool:
+        """Embeds are single-line elements."""
+        return False
+
+    @staticmethod
+    def _extract_text_content(rich_text: List[Dict[str, Any]]) -> str:
+        """Extract plain text content from Notion rich_text elements."""
+        result = ""
+        for text_obj in rich_text:
+            if text_obj.get("type") == "text":
+                result += text_obj.get("text", {}).get("content", "")
+            elif "plain_text" in text_obj:
+                result += text_obj.get("plain_text", "")
+        return result
+
+    @classmethod
+    def get_llm_prompt_content(cls) -> dict:
+        """Returns information for LLM prompts about this element."""
+        return {
+            "description": "Embeds external content from websites, PDFs, Google Maps, and other sources directly in your document.",
+            "when_to_use": "Use embeds when you want to include external content that isn't just a video or image. Embeds are great for interactive content, reference materials, or live data sources.",
+            "syntax": [
+                "<embed>(https://example.com) - Embed without caption",
+                "<embed:Caption text>(https://example.com) - Embed with caption",
+            ],
+            "supported_sources": [
+                "Websites and web pages",
+                "PDFs and documents",
+                "Google Maps",
+                "Google Drive files",
+                "Twitter/X posts",
+                "GitHub repositories and code",
+                "Figma designs",
+                "Miro boards",
+                "Many other services supported by Notion's embed feature",
+            ],
+            "examples": [
+                "<embed:Course materials>(https://drive.google.com/file/d/123456/view)",
+                "<embed:Our office location>(https://www.google.com/maps?q=San+Francisco)",
+                "<embed:Latest announcement>(https://twitter.com/NotionHQ/status/1234567890)",
+                "<embed:Project documentation>(https://github.com/username/repo)",
+                "<embed>(https://example.com/important-reference.pdf)",
+            ],
+        }

{notionary-0.1.8 → notionary-0.1.10}/notionary/core/converters/registry/block_element_registry.py

@@ -95,6 +95,8 @@ class BlockElementRegistry:
         """
         # Create a copy of registered elements
         element_classes = self._elements.copy()
+        
+        # TODO: Das hier besser formattieren und über debug level lösen . ) 
         print("Elements in registry:", element_classes)
 
         formatter_names = [e.__name__ for e in element_classes]
@@ -232,3 +234,7 @@ paragraphs, lists, quotes, etc.
         element_docs = cls.generate_element_docs(element_classes)
 
         return cls.SYSTEM_PROMPT_TEMPLATE.format(element_docs=element_docs)
+
+
+
+# TODO: Testen ob der Inline Formatter hier überhaupt funktionieren tut:

{notionary-0.1.8 → notionary-0.1.10}/notionary/core/converters/registry/block_element_registry_builder.py

@@ -1,6 +1,7 @@
 from typing import List, Type
 from collections import OrderedDict
 
+from notionary.core.converters.elements.embed_element import EmbedElement
 from notionary.core.converters.elements.notion_block_element import NotionBlockElement
 from notionary.core.converters.registry.block_element_registry import (
     BlockElementRegistry,
@@ -90,6 +91,7 @@ class BlockElementRegistryBuilder:
             .add_element(BookmarkElement)
             .add_element(ImageElement)
             .add_element(VideoElement)
+            .add_element(EmbedElement)
             .add_element(ParagraphElement)
         )  # Add paragraph last as fallback
 

{notionary-0.1.8 → notionary-0.1.10}/notionary/core/page/notion_page_manager.py

@@ -305,6 +305,27 @@ async def main():
     
     print("\nDemonstration abgeschlossen.")
 
+
+async def demo2():
+    url = "https://www.notion.so/Jarvis-Clipboard-1a3389d57bd380d7a507e67d1b25822c"
+    
+    page_manager = NotionPageManager(url=url)
+    
+    # Beispiel mit einem Embed-Element im Transcript-Toggle
+    markdown = """
+## 💪 Muskelaufbau und Kraft
+- Regelmäßiges Training ist essentiell für den Muskelerhalt im Alter
+- Richtige Ernährung unterstützt die Regeneration nach dem Training
+
++++ Transcript
+    <embed:Listen to this highlight>(https://share.snipd.com/snip/ad3d95b2-d648-4fa9-9036-bd7df653ea32)
+    <!-- spacer -->
+    ... In diesem Teil des Podcasts erklärt der Sprecher, wie das Henneman-Größenprinzip funktioniert und wie Muskelfasern rekrutiert werden, basierend auf der Schwere des zu bewegenden Objekts.
+"""
+    
+    await page_manager.append_markdown(markdown)
+    print("Markdown wurde zur Notion-Seite hinzugefügt.")
+
 if __name__ == "__main__":
-    asyncio.run(main())
-    print("\nDemonstration completed.")
+    asyncio.run(demo2())
+    print("\nDemonstration completed.")

{notionary-0.1.8 → notionary-0.1.10}/notionary/util/page_id_utils.py

@@ -24,8 +24,6 @@ def format_uuid(value: str) -> Optional[str]:
     return extract_uuid(value)
 
 def extract_and_validate_page_id(page_id: Optional[str], url: Optional[str]) -> str:
-    print("page_id", page_id)
-    print("=====")
     if not page_id and not url:
         raise ValueError("Either page_id or url must be provided")
     

{notionary-0.1.8 → notionary-0.1.10}/notionary.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: notionary
-Version: 0.1.8
+Version: 0.1.10
 Summary: A toolkit to convert between Markdown and Notion blocks
 Home-page: https://github.com/mathisarends/notionary
 Author: Mathis Arends

{notionary-0.1.8 → notionary-0.1.10}/notionary.egg-info/SOURCES.txt

@@ -16,6 +16,7 @@ notionary/core/converters/elements/callout_element.py
 notionary/core/converters/elements/code_block_element.py
 notionary/core/converters/elements/column_element.py
 notionary/core/converters/elements/divider_element.py
+notionary/core/converters/elements/embed_element.py
 notionary/core/converters/elements/heading_element.py
 notionary/core/converters/elements/image_element.py
 notionary/core/converters/elements/list_element.py

{notionary-0.1.8 → notionary-0.1.10}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_namespace_packages
 
 setup(
     name="notionary",
-    version="0.1.8",
+    version="0.1.10",
     packages=find_namespace_packages(include=["notionary*"]),
     install_requires=[
         "notion-client>=2.0.0",