Merge branch 'bagatur/community' of github.com:langchain-ai/langchain…

… into bagatur/community

.scripts/community_split/libs/community/langchain_community/agent_toolkits/__init__.py

@@ -0,0 +1,123 @@
+"""Agent toolkits contain integrations with various resources and services.
+
+LangChain has a large ecosystem of integrations with various external resources
+like local and remote file systems, APIs and databases.
+
+These integrations allow developers to create versatile applications that combine the
+power of LLMs with the ability to access, interact with and manipulate external
+resources.
+
+When developing an application, developers should inspect the capabilities and
+permissions of the tools that underlie the given agent toolkit, and determine
+whether permissions of the given toolkit are appropriate for the application.
+
+See [Security](https://python.langchain.com/docs/security) for more information.
+"""
+from pathlib import Path
+from typing import Any
+
+from langchain_core._api.path import as_import_path
+
+from langchain_community.agent_toolkits.ainetwork.toolkit import AINetworkToolkit
+from langchain_community.agent_toolkits.amadeus.toolkit import AmadeusToolkit
+from langchain_community.agent_toolkits.azure_cognitive_services import (
+    AzureCognitiveServicesToolkit,
+)
+from langchain_community.agent_toolkits.conversational_retrieval.openai_functions import (  # noqa: E501
+    create_conversational_retrieval_agent,
+)
+from langchain_community.agent_toolkits.file_management.toolkit import (
+    FileManagementToolkit,
+)
+from langchain_community.agent_toolkits.gmail.toolkit import GmailToolkit
+from langchain_community.agent_toolkits.jira.toolkit import JiraToolkit
+from langchain_community.agent_toolkits.json.base import create_json_agent
+from langchain_community.agent_toolkits.json.toolkit import JsonToolkit
+from langchain_community.agent_toolkits.multion.toolkit import MultionToolkit
+from langchain_community.agent_toolkits.nasa.toolkit import NasaToolkit
+from langchain_community.agent_toolkits.nla.toolkit import NLAToolkit
+from langchain_community.agent_toolkits.office365.toolkit import O365Toolkit
+from langchain_community.agent_toolkits.openapi.base import create_openapi_agent
+from langchain_community.agent_toolkits.openapi.toolkit import OpenAPIToolkit
+from langchain_community.agent_toolkits.playwright.toolkit import (
+    PlayWrightBrowserToolkit,
+)
+from langchain_community.agent_toolkits.powerbi.base import create_pbi_agent
+from langchain_community.agent_toolkits.powerbi.chat_base import create_pbi_chat_agent
+from langchain_community.agent_toolkits.powerbi.toolkit import PowerBIToolkit
+from langchain_community.agent_toolkits.slack.toolkit import SlackToolkit
+from langchain_community.agent_toolkits.spark_sql.base import create_spark_sql_agent
+from langchain_community.agent_toolkits.spark_sql.toolkit import SparkSQLToolkit
+from langchain_community.agent_toolkits.sql.base import create_sql_agent
+from langchain_community.agent_toolkits.sql.toolkit import SQLDatabaseToolkit
+from langchain_community.agent_toolkits.steam.toolkit import SteamToolkit
+from langchain_community.agent_toolkits.vectorstore.base import (
+    create_vectorstore_agent,
+    create_vectorstore_router_agent,
+)
+from langchain_community.agent_toolkits.vectorstore.toolkit import (
+    VectorStoreInfo,
+    VectorStoreRouterToolkit,
+    VectorStoreToolkit,
+)
+from langchain_community.agent_toolkits.zapier.toolkit import ZapierToolkit
+from langchain_community.tools.retriever import create_retriever_tool
+
+DEPRECATED_AGENTS = [
+    "create_csv_agent",
+    "create_pandas_dataframe_agent",
+    "create_xorbits_agent",
+    "create_python_agent",
+    "create_spark_dataframe_agent",
+]
+
+
+def __getattr__(name: str) -> Any:
+    """Get attr name."""
+    if name in DEPRECATED_AGENTS:
+        relative_path = as_import_path(Path(__file__).parent, suffix=name)
+        old_path = "langchain." + relative_path
+        new_path = "langchain_experimental." + relative_path
+        raise ImportError(
+            f"{name} has been moved to langchain experimental. "
+            "See https://github.com/langchain-ai/langchain/discussions/11680"
+            "for more information.\n"
+            f"Please update your import statement from: `{old_path}` to `{new_path}`."
+        )
+    raise AttributeError(f"{name} does not exist")
+
+
+__all__ = [
+    "AINetworkToolkit",
+    "AmadeusToolkit",
+    "AzureCognitiveServicesToolkit",
+    "FileManagementToolkit",
+    "GmailToolkit",
+    "JiraToolkit",
+    "JsonToolkit",
+    "MultionToolkit",
+    "NasaToolkit",
+    "NLAToolkit",
+    "O365Toolkit",
+    "OpenAPIToolkit",
+    "PlayWrightBrowserToolkit",
+    "PowerBIToolkit",
+    "SlackToolkit",
+    "SteamToolkit",
+    "SQLDatabaseToolkit",
+    "SparkSQLToolkit",
+    "VectorStoreInfo",
+    "VectorStoreRouterToolkit",
+    "VectorStoreToolkit",
+    "ZapierToolkit",
+    "create_json_agent",
+    "create_openapi_agent",
+    "create_pbi_agent",
+    "create_pbi_chat_agent",
+    "create_spark_sql_agent",
+    "create_sql_agent",
+    "create_vectorstore_agent",
+    "create_vectorstore_router_agent",
+    "create_conversational_retrieval_agent",
+    "create_retriever_tool",
+]

.scripts/community_split/libs/community/langchain_community/document_loaders/blob_loaders/file_system.py

@@ -0,0 +1,147 @@
+"""Use to load blobs from the local file system."""
+from pathlib import Path
+from typing import Callable, Iterable, Iterator, Optional, Sequence, TypeVar, Union
+
+from langchain_community.document_loaders.blob_loaders.schema import Blob, BlobLoader
+
+T = TypeVar("T")
+
+
+def _make_iterator(
+    length_func: Callable[[], int], show_progress: bool = False
+) -> Callable[[Iterable[T]], Iterator[T]]:
+    """Create a function that optionally wraps an iterable in tqdm."""
+    if show_progress:
+        try:
+            from tqdm.auto import tqdm
+        except ImportError:
+            raise ImportError(
+                "You must install tqdm to use show_progress=True."
+                "You can install tqdm with `pip install tqdm`."
+            )
+
+        # Make sure to provide `total` here so that tqdm can show
+        # a progress bar that takes into account the total number of files.
+        def _with_tqdm(iterable: Iterable[T]) -> Iterator[T]:
+            """Wrap an iterable in a tqdm progress bar."""
+            return tqdm(iterable, total=length_func())
+
+        iterator = _with_tqdm
+    else:
+        iterator = iter  # type: ignore
+
+    return iterator
+
+
+# PUBLIC API
+
+
+class FileSystemBlobLoader(BlobLoader):
+    """Load blobs in the local file system.
+
+    Example:
+
+    .. code-block:: python
+
+        from langchain_community.document_loaders.blob_loaders import FileSystemBlobLoader
+        loader = FileSystemBlobLoader("/path/to/directory")
+        for blob in loader.yield_blobs():
+            print(blob)
+    """  # noqa: E501
+
+    def __init__(
+        self,
+        path: Union[str, Path],
+        *,
+        glob: str = "**/[!.]*",
+        exclude: Sequence[str] = (),
+        suffixes: Optional[Sequence[str]] = None,
+        show_progress: bool = False,
+    ) -> None:
+        """Initialize with a path to directory and how to glob over it.
+
+        Args:
+            path: Path to directory to load from or path to file to load.
+                  If a path to a file is provided, glob/exclude/suffixes are ignored.
+            glob: Glob pattern relative to the specified path
+                  by default set to pick up all non-hidden files
+            exclude: patterns to exclude from results, use glob syntax
+            suffixes: Provide to keep only files with these suffixes
+                      Useful when wanting to keep files with different suffixes
+                      Suffixes must include the dot, e.g. ".txt"
+            show_progress: If true, will show a progress bar as the files are loaded.
+                           This forces an iteration through all matching files
+                           to count them prior to loading them.
+
+        Examples:
+
+            .. code-block:: python
+                from langchain_community.document_loaders.blob_loaders import FileSystemBlobLoader
+
+                # Load a single file.
+                loader = FileSystemBlobLoader("/path/to/file.txt")
+
+                # Recursively load all text files in a directory.
+                loader = FileSystemBlobLoader("/path/to/directory", glob="**/*.txt")
+
+                # Recursively load all non-hidden files in a directory.
+                loader = FileSystemBlobLoader("/path/to/directory", glob="**/[!.]*")
+
+                # Load all files in a directory without recursion.
+                loader = FileSystemBlobLoader("/path/to/directory", glob="*")
+
+                # Recursively load all files in a directory, except for py or pyc files.
+                loader = FileSystemBlobLoader(
+                    "/path/to/directory",
+                    glob="**/*.txt",
+                    exclude=["**/*.py", "**/*.pyc"]
+                )
+        """  # noqa: E501
+        if isinstance(path, Path):
+            _path = path
+        elif isinstance(path, str):
+            _path = Path(path)
+        else:
+            raise TypeError(f"Expected str or Path, got {type(path)}")
+
+        self.path = _path.expanduser()  # Expand user to handle ~
+        self.glob = glob
+        self.suffixes = set(suffixes or [])
+        self.show_progress = show_progress
+        self.exclude = exclude
+
+    def yield_blobs(
+        self,
+    ) -> Iterable[Blob]:
+        """Yield blobs that match the requested pattern."""
+        iterator = _make_iterator(
+            length_func=self.count_matching_files, show_progress=self.show_progress
+        )
+
+        for path in iterator(self._yield_paths()):
+            yield Blob.from_path(path)
+
+    def _yield_paths(self) -> Iterable[Path]:
+        """Yield paths that match the requested pattern."""
+        if self.path.is_file():
+            yield self.path
+            return
+
+        paths = self.path.glob(self.glob)
+        for path in paths:
+            if self.exclude:
+                if any(path.match(glob) for glob in self.exclude):
+                    continue
+            if path.is_file():
+                if self.suffixes and path.suffix not in self.suffixes:
+                    continue
+                yield path
+
+    def count_matching_files(self) -> int:
+        """Count files that match the pattern without loading them."""
+        # Carry out a full iteration to count the files without
+        # materializing anything expensive in memory.
+        num = 0
+        for _ in self._yield_paths():
+            num += 1
+        return num

.scripts/community_split/libs/community/langchain_community/document_loaders/generic.py

@@ -1,8 +1,16 @@
 from __future__ import annotations
 
 from pathlib import Path
-from typing import Any, Iterator, List, Literal, Optional, Sequence, Union, \
-    TYPE_CHECKING
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Iterator,
+    List,
+    Literal,
+    Optional,
+    Sequence,
+    Union,
+)
 
 from langchain_core.documents import Document
 
@@ -84,7 +92,7 @@ class GenericLoader(BaseLoader):
             parser=PyPDFParser()
         )
 
-    """
+    """  # noqa: E501
 
     def __init__(
         self,

.scripts/community_split/libs/community/langchain_community/document_loaders/parsers/generic.py

@@ -0,0 +1,70 @@
+"""Code for generic / auxiliary parsers.
+
+This module contains some logic to help assemble more sophisticated parsers.
+"""
+from typing import Iterator, Mapping, Optional
+
+from langchain_core.documents import Document
+
+from langchain_community.document_loaders.base import BaseBlobParser
+from langchain_community.document_loaders.blob_loaders.schema import Blob
+
+
+class MimeTypeBasedParser(BaseBlobParser):
+    """Parser that uses `mime`-types to parse a blob.
+
+    This parser is useful for simple pipelines where the mime-type is sufficient
+    to determine how to parse a blob.
+
+    To use, configure handlers based on mime-types and pass them to the initializer.
+
+    Example:
+
+        .. code-block:: python
+
+        from langchain_community.document_loaders.parsers.generic import MimeTypeBasedParser
+
+        parser = MimeTypeBasedParser(
+            handlers={
+                "application/pdf": ...,
+            },
+            fallback_parser=...,
+        )
+    """  # noqa: E501
+
+    def __init__(
+        self,
+        handlers: Mapping[str, BaseBlobParser],
+        *,
+        fallback_parser: Optional[BaseBlobParser] = None,
+    ) -> None:
+        """Define a parser that uses mime-types to determine how to parse a blob.
+
+        Args:
+            handlers: A mapping from mime-types to functions that take a blob, parse it
+                      and return a document.
+            fallback_parser: A fallback_parser parser to use if the mime-type is not
+                             found in the handlers. If provided, this parser will be
+                             used to parse blobs with all mime-types not found in
+                             the handlers.
+                             If not provided, a ValueError will be raised if the
+                             mime-type is not found in the handlers.
+        """
+        self.handlers = handlers
+        self.fallback_parser = fallback_parser
+
+    def lazy_parse(self, blob: Blob) -> Iterator[Document]:
+        """Load documents from a blob."""
+        mimetype = blob.mimetype
+
+        if mimetype is None:
+            raise ValueError(f"{blob} does not have a mimetype.")
+
+        if mimetype in self.handlers:
+            handler = self.handlers[mimetype]
+            yield from handler.lazy_parse(blob)
+        else:
+            if self.fallback_parser is not None:
+                yield from self.fallback_parser.lazy_parse(blob)
+            else:
+                raise ValueError(f"Unsupported mime type: {mimetype}")