agno 2.4.2__tar.gz → 2.4.3__tar.gz

{agno-2.4.2 → agno-2.4.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agno
-Version: 2.4.2
+Version: 2.4.3
 Summary: Agno: a lightweight library for building Multi-Agent Systems
 Author-email: Ashpreet Bedi <ashpreet@agno.com>
 Project-URL: homepage, https://agno.com
@@ -260,8 +260,9 @@ Provides-Extra: text
 Requires-Dist: aiofiles; extra == "text"
 Provides-Extra: csv
 Requires-Dist: aiofiles; extra == "csv"
-Requires-Dist: openpyxl; extra == "csv"
-Requires-Dist: xlrd; extra == "csv"
+Provides-Extra: excel
+Requires-Dist: openpyxl; extra == "excel"
+Requires-Dist: xlrd; extra == "excel"
 Provides-Extra: markdown
 Requires-Dist: unstructured; extra == "markdown"
 Requires-Dist: markdown; extra == "markdown"
@@ -376,6 +377,7 @@ Requires-Dist: agno[docx]; extra == "knowledge"
 Requires-Dist: agno[pptx]; extra == "knowledge"
 Requires-Dist: agno[text]; extra == "knowledge"
 Requires-Dist: agno[csv]; extra == "knowledge"
+Requires-Dist: agno[excel]; extra == "knowledge"
 Requires-Dist: agno[markdown]; extra == "knowledge"
 Requires-Dist: agno[chonkie]; extra == "knowledge"
 Provides-Extra: embedders
@@ -450,7 +452,7 @@ Dynamic: license-file
   <span>&nbsp;•&nbsp;</span>
   <a href="https://community.agno.com/">Community</a>
   <span>&nbsp;•&nbsp;</span>
-  <a href="https://discord.gg/4MtYHHrgA8">Discord</a>
+  <a href="https://www.agno.com/discord">Discord</a>
 </div>
 
 ## What is Agno?

{agno-2.4.2 → agno-2.4.3}/README.md

@@ -19,7 +19,7 @@
   <span>&nbsp;•&nbsp;</span>
   <a href="https://community.agno.com/">Community</a>
   <span>&nbsp;•&nbsp;</span>
-  <a href="https://discord.gg/4MtYHHrgA8">Discord</a>
+  <a href="https://www.agno.com/discord">Discord</a>
 </div>
 
 ## What is Agno?

{agno-2.4.2 → agno-2.4.3}/agno/agent/agent.py

@@ -5181,6 +5181,8 @@ class Agent:
         functions: Optional[Dict[str, Function]] = None,
         stream_events: bool = False,
     ) -> Iterator[RunOutputEvent]:
+        from agno.run.agent import CustomEvent
+
         self.model = cast(Model, self.model)
         # Execute the tool
         function_call = self.model.get_function_call_to_run_from_tool_execution(tool, functions)
@@ -5222,6 +5224,10 @@ class Agent:
                                 events_to_skip=self.events_to_skip,  # type: ignore
                                 store_events=self.store_events,
                             )
+            # Yield CustomEvent instances from sync tool generators
+            elif isinstance(call_result, CustomEvent):
+                if stream_events:
+                    yield call_result  # type: ignore
 
         if len(function_call_results) > 0:
             run_messages.messages.extend(function_call_results)
@@ -5246,6 +5252,8 @@ class Agent:
         functions: Optional[Dict[str, Function]] = None,
         stream_events: bool = False,
     ) -> AsyncIterator[RunOutputEvent]:
+        from agno.run.agent import CustomEvent
+
         self.model = cast(Model, self.model)
 
         # Execute the tool
@@ -5288,6 +5296,11 @@ class Agent:
                                 events_to_skip=self.events_to_skip,  # type: ignore
                                 store_events=self.store_events,
                             )
+            # Yield CustomEvent instances from async tool generators
+            elif isinstance(call_result, CustomEvent):
+                if stream_events:
+                    yield call_result  # type: ignore
+
         if len(function_call_results) > 0:
             run_messages.messages.extend(function_call_results)
 

{agno-2.4.2 → agno-2.4.3}/agno/knowledge/embedder/openai.py

@@ -71,7 +71,8 @@ class OpenAIEmbedder(Embedder):
         }
         if self.user is not None:
             _request_params["user"] = self.user
-        if self.id.startswith("text-embedding-3"):
+        # Pass dimensions for text-embedding-3 models or when using custom base_url (third-party APIs)
+        if self.id.startswith("text-embedding-3") or self.base_url is not None:
             _request_params["dimensions"] = self.dimensions
         if self.request_params:
             _request_params.update(self.request_params)
@@ -106,7 +107,8 @@ class OpenAIEmbedder(Embedder):
         }
         if self.user is not None:
             req["user"] = self.user
-        if self.id.startswith("text-embedding-3"):
+        # Pass dimensions for text-embedding-3 models or when using custom base_url (third-party APIs)
+        if self.id.startswith("text-embedding-3") or self.base_url is not None:
             req["dimensions"] = self.dimensions
         if self.request_params:
             req.update(self.request_params)
@@ -126,7 +128,8 @@ class OpenAIEmbedder(Embedder):
         }
         if self.user is not None:
             req["user"] = self.user
-        if self.id.startswith("text-embedding-3"):
+        # Pass dimensions for text-embedding-3 models or when using custom base_url (third-party APIs)
+        if self.id.startswith("text-embedding-3") or self.base_url is not None:
             req["dimensions"] = self.dimensions
         if self.request_params:
             req.update(self.request_params)
@@ -166,7 +169,8 @@ class OpenAIEmbedder(Embedder):
             }
             if self.user is not None:
                 req["user"] = self.user
-            if self.id.startswith("text-embedding-3"):
+            # Pass dimensions for text-embedding-3 models or when using custom base_url (third-party APIs)
+            if self.id.startswith("text-embedding-3") or self.base_url is not None:
                 req["dimensions"] = self.dimensions
             if self.request_params:
                 req.update(self.request_params)

{agno-2.4.2 → agno-2.4.3}/agno/knowledge/knowledge.py

@@ -983,6 +983,11 @@ class Knowledge:
         """CSV reader - lazy loaded via factory."""
         return self._get_reader("csv")
 
+    @property
+    def excel_reader(self) -> Optional[Reader]:
+        """Excel reader - lazy loaded via factory."""
+        return self._get_reader("excel")
+
     @property
     def docx_reader(self) -> Optional[Reader]:
         """Docx reader - lazy loaded via factory."""
@@ -1123,6 +1128,8 @@ class Knowledge:
             return self.json_reader, ""
         elif file_extension == ".markdown":
             return self.markdown_reader, ""
+        elif file_extension in [".xlsx", ".xls"]:
+            return self.excel_reader, ""
         else:
             return self.text_reader, ""
 
@@ -1153,6 +1160,8 @@ class Knowledge:
             return self.json_reader
         elif uri_lower.endswith(".markdown"):
             return self.markdown_reader
+        elif uri_lower.endswith(".xlsx") or uri_lower.endswith(".xls"):
+            return self.excel_reader
         else:
             return self.text_reader
 
@@ -1284,6 +1293,10 @@ class Knowledge:
             if self._should_include_file(str(path), include, exclude):
                 log_debug(f"Adding file {path} due to include/exclude filters")
 
+                # Set name from path if not provided
+                if not content.name:
+                    content.name = path.name
+
                 await self._ainsert_contents_db(content)
                 if self._should_skip(content.content_hash, skip_if_exists):  # type: ignore[arg-type]
                     content.status = ContentStatus.COMPLETED
@@ -1365,6 +1378,10 @@ class Knowledge:
             if self._should_include_file(str(path), include, exclude):
                 log_debug(f"Adding file {path} due to include/exclude filters")
 
+                # Set name from path if not provided
+                if not content.name:
+                    content.name = path.name
+
                 self._insert_contents_db(content)
                 if self._should_skip(content.content_hash, skip_if_exists):  # type: ignore[arg-type]
                     content.status = ContentStatus.COMPLETED
@@ -1450,6 +1467,14 @@ class Knowledge:
         if not content.url:
             raise ValueError("No url provided")
 
+        # Set name from URL if not provided
+        if not content.name and content.url:
+            from urllib.parse import urlparse
+
+            parsed = urlparse(content.url)
+            url_path = Path(parsed.path)
+            content.name = url_path.name if url_path.name else content.url
+
         # 1. Add content to contents database
         await self._ainsert_contents_db(content)
         if self._should_skip(content.content_hash, skip_if_exists):  # type: ignore[arg-type]
@@ -1547,6 +1572,14 @@ class Knowledge:
         if not content.url:
             raise ValueError("No url provided")
 
+        # Set name from URL if not provided
+        if not content.name and content.url:
+            from urllib.parse import urlparse
+
+            parsed = urlparse(content.url)
+            url_path = Path(parsed.path)
+            content.name = url_path.name if url_path.name else content.url
+
         # 1. Add content to contents database
         self._insert_contents_db(content)
         if self._should_skip(content.content_hash, skip_if_exists):  # type: ignore[arg-type]
@@ -1633,6 +1666,8 @@ class Knowledge:
 
         if content.name:
             name = content.name
+        elif content.file_data and content.file_data.filename:
+            name = content.file_data.filename
         elif content.file_data and content.file_data.content:
             if isinstance(content.file_data.content, bytes):
                 name = content.file_data.content[:10].decode("utf-8", errors="ignore")
@@ -1696,9 +1731,17 @@ class Knowledge:
                     log_debug(f"Using reader: {content.reader.__class__.__name__} to read content")
                     reader = content.reader
                 else:
-                    reader = self._select_reader(content.file_data.type)
-                name = content.name if content.name else f"content_{content.file_data.type}"
-                read_documents = await reader.async_read(content_io, name=name)
+                    # Prefer filename extension over MIME type for reader selection
+                    # (browsers often send wrong MIME types for Excel files)
+                    reader_hint = content.file_data.type
+                    if content.file_data.filename:
+                        ext = Path(content.file_data.filename).suffix.lower()
+                        if ext:
+                            reader_hint = ext
+                    reader = self._select_reader(reader_hint)
+                # Use file_data.filename for reader (preserves extension for format detection)
+                reader_name = content.file_data.filename or content.name or f"content_{content.file_data.type}"
+                read_documents = await reader.async_read(content_io, name=reader_name)
                 if not content.id:
                     content.id = generate_id(content.content_hash or "")
                 self._prepare_documents_for_insert(read_documents, content.id, metadata=content.metadata)
@@ -1730,6 +1773,8 @@ class Knowledge:
 
         if content.name:
             name = content.name
+        elif content.file_data and content.file_data.filename:
+            name = content.file_data.filename
         elif content.file_data and content.file_data.content:
             if isinstance(content.file_data.content, bytes):
                 name = content.file_data.content[:10].decode("utf-8", errors="ignore")
@@ -1793,9 +1838,17 @@ class Knowledge:
                     log_debug(f"Using reader: {content.reader.__class__.__name__} to read content")
                     reader = content.reader
                 else:
-                    reader = self._select_reader(content.file_data.type)
-                name = content.name if content.name else f"content_{content.file_data.type}"
-                read_documents = reader.read(content_io, name=name)
+                    # Prefer filename extension over MIME type for reader selection
+                    # (browsers often send wrong MIME types for Excel files)
+                    reader_hint = content.file_data.type
+                    if content.file_data.filename:
+                        ext = Path(content.file_data.filename).suffix.lower()
+                        if ext:
+                            reader_hint = ext
+                    reader = self._select_reader(reader_hint)
+                # Use file_data.filename for reader (preserves extension for format detection)
+                reader_name = content.file_data.filename or content.name or f"content_{content.file_data.type}"
+                read_documents = reader.read(content_io, name=reader_name)
                 if not content.id:
                     content.id = generate_id(content.content_hash or "")
                 self._prepare_documents_for_insert(read_documents, content.id, metadata=content.metadata)

agno-2.4.3/agno/knowledge/reader/csv_reader.py

@@ -0,0 +1,215 @@
+import asyncio
+import csv
+import io
+from pathlib import Path
+from typing import IO, Any, List, Optional, Union
+from uuid import uuid4
+
+try:
+    import aiofiles
+except ImportError:
+    raise ImportError("`aiofiles` not installed. Please install it with `pip install aiofiles`")
+
+from agno.knowledge.chunking.row import RowChunking
+from agno.knowledge.chunking.strategy import ChunkingStrategy, ChunkingStrategyType
+from agno.knowledge.document.base import Document
+from agno.knowledge.reader.base import Reader
+from agno.knowledge.reader.utils import stringify_cell_value
+from agno.knowledge.types import ContentType
+from agno.utils.log import log_debug, log_error
+
+
+class CSVReader(Reader):
+    """Reader for CSV files.
+
+    Converts CSV files to documents with optional chunking support.
+    For Excel files (.xlsx, .xls), use ExcelReader instead.
+
+    Args:
+        chunking_strategy: Strategy for chunking documents. Default is RowChunking.
+        **kwargs: Additional arguments passed to base Reader.
+
+    Example:
+        ```python
+        from agno.knowledge.reader.csv_reader import CSVReader
+
+        reader = CSVReader()
+        docs = reader.read("data.csv")
+
+        # Custom delimiter
+        docs = reader.read("data.tsv", delimiter="\\t")
+        ```
+    """
+
+    def __init__(self, chunking_strategy: Optional[ChunkingStrategy] = RowChunking(), **kwargs):
+        super().__init__(chunking_strategy=chunking_strategy, **kwargs)
+
+    @classmethod
+    def get_supported_chunking_strategies(cls) -> List[ChunkingStrategyType]:
+        """Get the list of supported chunking strategies for CSV readers."""
+        return [
+            ChunkingStrategyType.ROW_CHUNKER,
+            ChunkingStrategyType.CODE_CHUNKER,
+            ChunkingStrategyType.FIXED_SIZE_CHUNKER,
+            ChunkingStrategyType.AGENTIC_CHUNKER,
+            ChunkingStrategyType.DOCUMENT_CHUNKER,
+            ChunkingStrategyType.RECURSIVE_CHUNKER,
+        ]
+
+    @classmethod
+    def get_supported_content_types(cls) -> List[ContentType]:
+        """Get the list of supported content types."""
+        return [ContentType.CSV]
+
+    def read(
+        self, file: Union[Path, IO[Any]], delimiter: str = ",", quotechar: str = '"', name: Optional[str] = None
+    ) -> List[Document]:
+        """Read a CSV file and return a list of documents.
+
+        Args:
+            file: Path to CSV file or file-like object.
+            delimiter: CSV field delimiter. Default is comma.
+            quotechar: CSV quote character. Default is double quote.
+            name: Optional name override for the document.
+
+        Returns:
+            List of Document objects.
+
+        Raises:
+            FileNotFoundError: If the file path doesn't exist.
+        """
+        try:
+            if isinstance(file, Path):
+                if not file.exists():
+                    raise FileNotFoundError(f"Could not find file: {file}")
+                log_debug(f"Reading: {file}")
+                csv_name = name or file.stem
+                file_content: Union[io.TextIOWrapper, io.StringIO] = file.open(
+                    newline="", mode="r", encoding=self.encoding or "utf-8"
+                )
+            else:
+                log_debug(f"Reading retrieved file: {getattr(file, 'name', 'BytesIO')}")
+                csv_name = name or getattr(file, "name", "csv_file").split(".")[0]
+                file.seek(0)
+                file_content = io.StringIO(file.read().decode(self.encoding or "utf-8"))
+
+            csv_lines: List[str] = []
+            with file_content as csvfile:
+                csv_reader = csv.reader(csvfile, delimiter=delimiter, quotechar=quotechar)
+                for row in csv_reader:
+                    # Normalize line endings in CSV cells to preserve row integrity
+                    csv_lines.append(", ".join(stringify_cell_value(cell) for cell in row))
+
+            documents = [
+                Document(
+                    name=csv_name,
+                    id=str(uuid4()),
+                    content="\n".join(csv_lines),
+                )
+            ]
+            if self.chunk:
+                chunked_documents = []
+                for document in documents:
+                    chunked_documents.extend(self.chunk_document(document))
+                return chunked_documents
+            return documents
+        except FileNotFoundError:
+            raise
+        except UnicodeDecodeError as e:
+            file_desc = getattr(file, "name", str(file)) if isinstance(file, IO) else file
+            log_error(f"Encoding error reading {file_desc}: {e}. Try specifying a different encoding.")
+            return []
+        except Exception as e:
+            file_desc = getattr(file, "name", str(file)) if isinstance(file, IO) else file
+            log_error(f"Error reading {file_desc}: {e}")
+            return []
+
+    async def async_read(
+        self,
+        file: Union[Path, IO[Any]],
+        delimiter: str = ",",
+        quotechar: str = '"',
+        page_size: int = 1000,
+        name: Optional[str] = None,
+    ) -> List[Document]:
+        """Read a CSV file asynchronously, processing batches of rows concurrently.
+
+        Args:
+            file: Path to CSV file or file-like object.
+            delimiter: CSV field delimiter. Default is comma.
+            quotechar: CSV quote character. Default is double quote.
+            page_size: Number of rows per page for large files.
+            name: Optional name override for the document.
+
+        Returns:
+            List of Document objects.
+
+        Raises:
+            FileNotFoundError: If the file path doesn't exist.
+        """
+        try:
+            if isinstance(file, Path):
+                if not file.exists():
+                    raise FileNotFoundError(f"Could not find file: {file}")
+                log_debug(f"Reading async: {file}")
+                async with aiofiles.open(file, mode="r", encoding=self.encoding or "utf-8", newline="") as file_content:
+                    content = await file_content.read()
+                    file_content_io = io.StringIO(content)
+                csv_name = name or file.stem
+            else:
+                log_debug(f"Reading retrieved file async: {getattr(file, 'name', 'BytesIO')}")
+                file.seek(0)
+                file_content_io = io.StringIO(file.read().decode(self.encoding or "utf-8"))
+                csv_name = name or getattr(file, "name", "csv_file").split(".")[0]
+
+            file_content_io.seek(0)
+            csv_reader = csv.reader(file_content_io, delimiter=delimiter, quotechar=quotechar)
+            rows = list(csv_reader)
+            total_rows = len(rows)
+
+            if total_rows <= 10:
+                # Small files: single document
+                csv_content = " ".join(", ".join(stringify_cell_value(cell) for cell in row) for row in rows)
+                documents = [
+                    Document(
+                        name=csv_name,
+                        id=str(uuid4()),
+                        content=csv_content,
+                    )
+                ]
+            else:
+                # Large files: paginate and process in parallel
+                pages = []
+                for i in range(0, total_rows, page_size):
+                    pages.append(rows[i : i + page_size])
+
+                async def _process_page(page_number: int, page_rows: List[List[str]]) -> Document:
+                    """Process a page of rows into a document."""
+                    start_row = (page_number - 1) * page_size + 1
+                    page_content = " ".join(", ".join(stringify_cell_value(cell) for cell in row) for row in page_rows)
+
+                    return Document(
+                        name=csv_name,
+                        id=str(uuid4()),
+                        meta_data={"page": page_number, "start_row": start_row, "rows": len(page_rows)},
+                        content=page_content,
+                    )
+
+                documents = await asyncio.gather(
+                    *[_process_page(page_number, page) for page_number, page in enumerate(pages, start=1)]
+                )
+
+            if self.chunk:
+                documents = await self.chunk_documents_async(documents)
+
+            return documents
+        except FileNotFoundError:
+            raise
+        except UnicodeDecodeError as e:
+            file_desc = getattr(file, "name", str(file)) if isinstance(file, IO) else file
+            log_error(f"Encoding error reading {file_desc}: {e}. Try specifying a different encoding.")
+            return []
+        except Exception as e:
+            file_desc = getattr(file, "name", str(file)) if isinstance(file, IO) else file
+            log_error(f"Error reading {file_desc}: {e}")
+            return []