notionary 0.1.9__tar.gz → 0.1.11__tar.gz

{notionary-0.1.9 → notionary-0.1.11}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: notionary
-Version: 0.1.9
+Version: 0.1.11
 Summary: A toolkit to convert between Markdown and Notion blocks
 Home-page: https://github.com/mathisarends/notionary
 Author: Mathis Arends

notionary-0.1.11/notionary/core/converters/elements/audio_element.py

@@ -0,0 +1,141 @@
+import re
+from typing import Dict, Any, Optional, List
+from notionary.core.converters.elements.notion_block_element import NotionBlockElement
+
+
+class AudioElement(NotionBlockElement):
+    """
+    Handles conversion between Markdown audio embeds and Notion audio blocks.
+
+    Markdown audio syntax (custom format since standard Markdown doesn't support audio):
+    - $[Caption](https://example.com/audio.mp3) - Basic audio with caption
+    - $[](https://example.com/audio.mp3) - Audio without caption
+    - $[Caption](https://storage.googleapis.com/audio_summaries/example.mp3) - CDN hosted audio
+
+    Supports various audio URLs including direct audio file links from CDNs and other sources.
+    """
+
+    # Regex pattern for audio syntax
+    PATTERN = re.compile(
+        r"^\$\[(.*?)\]"  # $[Caption] part
+        + r'\((https?://[^\s"]+)'  # (URL part
+        + r"\)$"  # closing parenthesis
+    )
+
+    # Audio file extensions
+    AUDIO_EXTENSIONS = [".mp3", ".wav", ".ogg", ".m4a", ".flac", ".aac"]
+
+    @staticmethod
+    def match_markdown(text: str) -> bool:
+        """Check if text is a markdown audio embed."""
+        text = text.strip()
+        return text.startswith("$[") and bool(AudioElement.PATTERN.match(text))
+
+    @staticmethod
+    def match_notion(block: Dict[str, Any]) -> bool:
+        """Check if block is a Notion audio."""
+        return block.get("type") == "audio"
+
+    @staticmethod
+    def is_audio_url(url: str) -> bool:
+        """Check if URL points to an audio file."""
+        return any(url.lower().endswith(ext) for ext in AudioElement.AUDIO_EXTENSIONS) or \
+               "audio" in url.lower() or \
+               "storage.googleapis.com/audio_summaries" in url.lower()
+
+    @staticmethod
+    def markdown_to_notion(text: str) -> Optional[Dict[str, Any]]:
+        """Convert markdown audio embed to Notion audio block."""
+        audio_match = AudioElement.PATTERN.match(text.strip())
+        if not audio_match:
+            return None
+
+        caption = audio_match.group(1)
+        url = audio_match.group(2)
+
+        if not url:
+            return None
+
+        # Make sure this is an audio URL
+        if not AudioElement.is_audio_url(url):
+            # If not obviously an audio URL, we'll still accept it as the user might know better
+            pass
+
+        # Prepare the audio block
+        audio_block = {
+            "type": "audio",
+            "audio": {"type": "external", "external": {"url": url}},
+        }
+
+        # Add caption if provided
+        if caption:
+            audio_block["audio"]["caption"] = [
+                {"type": "text", "text": {"content": caption}}
+            ]
+
+        return audio_block
+
+    @staticmethod
+    def notion_to_markdown(block: Dict[str, Any]) -> Optional[str]:
+        """Convert Notion audio block to markdown audio embed."""
+        if block.get("type") != "audio":
+            return None
+
+        audio_data = block.get("audio", {})
+
+        # Handle both external and file (uploaded) audios
+        if audio_data.get("type") == "external":
+            url = audio_data.get("external", {}).get("url", "")
+        elif audio_data.get("type") == "file":
+            url = audio_data.get("file", {}).get("url", "")
+        else:
+            return None
+
+        if not url:
+            return None
+
+        # Extract caption if available
+        caption = ""
+        caption_rich_text = audio_data.get("caption", [])
+        if caption_rich_text:
+            caption = AudioElement._extract_text_content(caption_rich_text)
+
+        return f"$[{caption}]({url})"
+
+    @staticmethod
+    def is_multiline() -> bool:
+        """Audio embeds are single-line elements."""
+        return False
+
+    @staticmethod
+    def _extract_text_content(rich_text: List[Dict[str, Any]]) -> str:
+        """Extract plain text content from Notion rich_text elements."""
+        result = ""
+        for text_obj in rich_text:
+            if text_obj.get("type") == "text":
+                result += text_obj.get("text", {}).get("content", "")
+            elif "plain_text" in text_obj:
+                result += text_obj.get("plain_text", "")
+        return result
+
+    @classmethod
+    def get_llm_prompt_content(cls) -> dict:
+        """Returns information for LLM prompts about this element."""
+        return {
+            "description": "Embeds audio content from external sources like CDNs or direct audio URLs.",
+            "when_to_use": "Use audio embeds when you want to include audio content directly in your document. Audio embeds are useful for podcasts, music, voice recordings, or any content that benefits from audio explanation.",
+            "syntax": [
+                "$[](https://example.com/audio.mp3) - Audio without caption",
+                "$[Caption text](https://example.com/audio.mp3) - Audio with caption",
+            ],
+            "supported_sources": [
+                "Direct links to audio files (.mp3, .wav, .ogg, etc.)",
+                "Google Cloud Storage links (storage.googleapis.com)",
+                "Other audio hosting platforms supported by Notion",
+            ],
+            "examples": [
+                "$[Podcast Episode](https://storage.googleapis.com/audio_summaries/ep_ai_summary_127d02ec-ca12-4312-a5ed-cb14b185480c.mp3)",
+                "$[Voice recording](https://example.com/audio/recording.mp3)",
+                "$[](https://storage.googleapis.com/audio_summaries/example.mp3)",
+            ],
+        }

notionary-0.1.11/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-0.1.9 → notionary-0.1.11}/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]
@@ -231,4 +233,4 @@ paragraphs, lists, quotes, etc.
         """
         element_docs = cls.generate_element_docs(element_classes)
 
-        return cls.SYSTEM_PROMPT_TEMPLATE.format(element_docs=element_docs)
+        return cls.SYSTEM_PROMPT_TEMPLATE.format(element_docs=element_docs)

{notionary-0.1.9 → notionary-0.1.11}/notionary/core/converters/registry/block_element_registry_builder.py

@@ -1,6 +1,8 @@
 from typing import List, Type
 from collections import OrderedDict
 
+from notionary.core.converters.elements.audio_element import AudioElement
+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 +92,8 @@ class BlockElementRegistryBuilder:
             .add_element(BookmarkElement)
             .add_element(ImageElement)
             .add_element(VideoElement)
+            .add_element(EmbedElement)
+            .add_element(AudioElement)
             .add_element(ParagraphElement)
         )  # Add paragraph last as fallback
 

{notionary-0.1.9 → notionary-0.1.11}/notionary/core/page/notion_page_manager.py

@@ -306,6 +306,17 @@ async def main():
     print("\nDemonstration abgeschlossen.")
 
 
+async def demo2():
+    url = "https://www.notion.so/Jarvis-Clipboard-1a3389d57bd380d7a507e67d1b25822c"
+    
+    page_manager = NotionPageManager(url=url)
+    
+    markdown = """
+$[Podcast Zusammenfassung](https://storage.googleapis.com/audio_summaries/ep_ai_summary_127d02ec-ca12-4312-a5ed-cb14b185480c.mp3)
+    """
+
+    await page_manager.append_markdown(markdown=markdown)
+
 if __name__ == "__main__":
-    asyncio.run(main())
-    print("\nDemonstration completed.")
+    asyncio.run(demo2())
+    print("\nDemonstration completed.")

{notionary-0.1.9 → notionary-0.1.11}/notionary.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: notionary
-Version: 0.1.9
+Version: 0.1.11
 Summary: A toolkit to convert between Markdown and Notion blocks
 Home-page: https://github.com/mathisarends/notionary
 Author: Mathis Arends

{notionary-0.1.9 → notionary-0.1.11}/notionary.egg-info/SOURCES.txt

@@ -11,11 +11,13 @@ notionary/core/notion_client.py
 notionary/core/converters/__init__.py
 notionary/core/converters/markdown_to_notion_converter.py
 notionary/core/converters/notion_to_markdown_converter.py
+notionary/core/converters/elements/audio_element.py
 notionary/core/converters/elements/bookmark_element.py
 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-0.1.9 → notionary-0.1.11}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_namespace_packages
 
 setup(
     name="notionary",
-    version="0.1.9",
+    version="0.1.11",
     packages=find_namespace_packages(include=["notionary*"]),
     install_requires=[
         "notion-client>=2.0.0",