lfx-nightly 0.1.12.dev21__py3-none-any.whl → 0.1.12.dev22__py3-none-any.whl

lfx/base/data/base_file.py

@@ -8,6 +8,7 @@ from tempfile import TemporaryDirectory
 from typing import TYPE_CHECKING
 from zipfile import ZipFile, is_zipfile
 
+import anyio
 import pandas as pd
 
 from lfx.custom.custom_component.component import Component
@@ -15,6 +16,7 @@ from lfx.io import BoolInput, FileInput, HandleInput, Output, StrInput
 from lfx.schema.data import Data
 from lfx.schema.dataframe import DataFrame
 from lfx.schema.message import Message
+from lfx.utils.helpers import build_content_type_from_extension
 
 if TYPE_CHECKING:
     from collections.abc import Callable
@@ -242,25 +244,64 @@ class BaseFileComponent(Component, ABC):
             return [Data()]
         return data_list
 
-    def load_files_message(self) -> Message:
+    async def _extract_file_metadata(self, data_item) -> dict:
+        """Extract metadata from a data item with file_path."""
+        metadata = {}
+        if not hasattr(data_item, "file_path"):
+            return metadata
+
+        file_path = data_item.file_path
+        file_path_obj = anyio.Path(file_path)
+        file_size_stat = await file_path_obj.stat()
+        filename = file_path_obj.name
+
+        # Basic file metadata
+        metadata["filename"] = filename
+        metadata["file_size"] = file_size_stat.st_size
+
+        # Add MIME type from extension
+        extension = filename.split(".")[-1]
+        if extension:
+            metadata["mimetype"] = build_content_type_from_extension(extension)
+
+        # Copy additional metadata from data if available
+        if hasattr(data_item, "data") and isinstance(data_item.data, dict):
+            metadata_fields = ["mimetype", "file_size", "created_time", "modified_time"]
+            for field in metadata_fields:
+                if field in data_item.data:
+                    metadata[field] = data_item.data[field]
+
+        return metadata
+
+    def _extract_text(self, data_item) -> str:
+        """Extract text content from a data item."""
+        if isinstance(data_item.data, dict):
+            text = getattr(data_item, "get_text", lambda: None)() or data_item.data.get("text")
+            return text if text is not None else str(data_item)
+        return str(data_item)
+
+    async def load_files_message(self) -> Message:
         """Load files and return as Message.
 
         Returns:
-            Message: Message containing all file data
+          Message: Message containing all file data
         """
         data_list = self.load_files_core()
         if not data_list:
-            return Message()  # No data -> empty message
+            return Message()
 
         sep: str = getattr(self, "separator", "\n\n") or "\n\n"
-
         parts: list[str] = []
-        for d in data_list:
-            # Prefer explicit text if available, fall back to full dict, lastly str()
-            text = (getattr(d, "get_text", lambda: None)() or d.data.get("text")) if isinstance(d.data, dict) else None
-            parts.append(text if text is not None else str(d))
+        metadata = {}
+
+        for data_item in data_list:
+            parts.append(self._extract_text(data_item))
+
+            # Set metadata from first file only
+            if not metadata:
+                metadata = await self._extract_file_metadata(data_item)
 
-        return Message(text=sep.join(parts))
+        return Message(text=sep.join(parts), metadata=metadata)
 
     def load_files_path(self) -> Message:
         """Returns a Message containing file paths from loaded files.

lfx/utils/constants.py

@@ -203,3 +203,31 @@ MESSAGE_SENDER_AI = "Machine"
 MESSAGE_SENDER_USER = "User"
 MESSAGE_SENDER_NAME_AI = "AI"
 MESSAGE_SENDER_NAME_USER = "User"
+EXTENSION_TO_CONTENT_TYPE = {
+    "json": "application/json",
+    "txt": "text/plain",
+    "csv": "text/csv",
+    "html": "text/html",
+    "pdf": "application/pdf",
+    "png": "image/png",
+    "jpg": "image/jpeg",
+    "jpeg": "image/jpeg",
+    "gif": "image/gif",
+    "svg": "image/svg+xml",
+    "mp3": "audio/mpeg",
+    "wav": "audio/wav",
+    "mp4": "video/mp4",
+    "webm": "video/webm",
+    "zip": "application/zip",
+    "tar": "application/x-tar",
+    "gz": "application/gzip",
+    "doc": "application/msword",
+    "docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+    "xls": "application/vnd.ms-excel",
+    "xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+    "ppt": "application/vnd.ms-powerpoint",
+    "pptx": "application/vnd.openxmlformats-officedocument.presentationml.presentation",
+    "xml": "application/xml",
+    "yaml": "application/x-yaml",
+    "yml": "application/x-yaml",
+}

lfx/utils/helpers.py

@@ -5,6 +5,8 @@ from __future__ import annotations
 import mimetypes
 from typing import TYPE_CHECKING
 
+from lfx.utils.constants import EXTENSION_TO_CONTENT_TYPE
+
 if TYPE_CHECKING:
     from pathlib import Path
 
@@ -26,3 +28,7 @@ def get_mime_type(file_path: str | Path) -> str:
         msg = f"Could not determine MIME type for: {file_path}"
         raise ValueError(msg)
     return mime_type
+
+
+def build_content_type_from_extension(extension: str):
+    return EXTENSION_TO_CONTENT_TYPE.get(extension.lower(), "application/octet-stream")

{lfx_nightly-0.1.12.dev21.dist-info → lfx_nightly-0.1.12.dev22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: lfx-nightly
-Version: 0.1.12.dev21
+Version: 0.1.12.dev22
 Summary: Langflow Executor - A lightweight CLI tool for executing and serving Langflow AI flows
 Author-email: Gabriel Luiz Freitas Almeida <gabriel@langflow.org>
 Requires-Python: <3.14,>=3.10

{lfx_nightly-0.1.12.dev21.dist-info → lfx_nightly-0.1.12.dev22.dist-info}/RECORD

@@ -28,7 +28,7 @@ lfx/base/compressors/model.py,sha256=-FFBAPAy9bAgvklIo7x_uwShZR5NoMHakF6f_hNnLHg
 lfx/base/curl/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 lfx/base/curl/parse.py,sha256=Yw6mMbGg7e-ffrBItEUJeTiljneCXlNyt5afzEP9eUI,6094
 lfx/base/data/__init__.py,sha256=lQsYYMyAg_jA9ZF7oc-LNZsRE2uMGT6g16WzsUByHqs,81
-lfx/base/data/base_file.py,sha256=XFj3u9OGHcRbWfzslzvvxn-qpaCeX0uUQ0fStUCo65I,25495
+lfx/base/data/base_file.py,sha256=lHTwlupvcxm9eZLHBB_zIu1tL4VC0gqfJMfbHAMeTGc,26980
 lfx/base/data/docling_utils.py,sha256=gVDxOZghSJEo5n-UNkVGBQYqkvfNqkNkltBhAnoaJd4,13048
 lfx/base/data/utils.py,sha256=dGqEO4zE5s_V2Cs4j0EEeyLjYLX6Zex-EGzIOznK76o,5960
 lfx/base/document_transformers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -707,10 +707,10 @@ lfx/utils/async_helpers.py,sha256=py1koriS60Y0DAcX8FY0HLSWP7o7cWiYN3T0avermhs,13
 lfx/utils/component_utils.py,sha256=Zq2_HvXGd5V6ERMulY0slo-piKzKiXRK7QCOWeTnlqM,5734
 lfx/utils/concurrency.py,sha256=2k6hwDvGejH1Zr1yLylziG9LDePoQ18eIM2vkpyb6lo,1636
 lfx/utils/connection_string_parser.py,sha256=NmqhphFRNbhh7jvyNywDvUFgA4hPr8ikL-Sn11riizY,453
-lfx/utils/constants.py,sha256=1tMGRZVIHMhN33LdMw8YcEearIeO1xJKddJAjY6RIqA,6081
+lfx/utils/constants.py,sha256=4M8i93bROuQ7zmeKgfdNW85Znw7JFrK8KiagcDBpMRc,7036
 lfx/utils/data_structure.py,sha256=xU3JNa_4jcGOVa_ctfMxiImEj6dKQQPE_zZsTAyy2T4,6888
 lfx/utils/exceptions.py,sha256=RgIkI4uBssJsJUnuhluNGDSzdcuW5fnxPLhGfXYU9Uc,973
-lfx/utils/helpers.py,sha256=VcJ2oTKvQy0LFyeY-CvFjviA9RfQx4mnkg0zYI_OZf4,685
+lfx/utils/helpers.py,sha256=0LE0barnVp-8Y5cCoDRzhDzesvXqgiT7IXP6vtTSyGE,889
 lfx/utils/image.py,sha256=wMWBEI1gW3cFlQcio3mWgfHBaOw1uoAnqNmEacE_8xo,2133
 lfx/utils/lazy_load.py,sha256=UDtXi8N7NT9r-FRGxsLUfDtGU_X8yqt-RQqgpc9TqAw,394
 lfx/utils/request_utils.py,sha256=A6vmwpr7f3ZUxHg6Sz2-BdUUsyAwg84-7N_DNoPC8_Q,518
@@ -718,7 +718,7 @@ lfx/utils/schemas.py,sha256=NbOtVQBrn4d0BAu-0H_eCTZI2CXkKZlRY37XCSmuJwc,3865
 lfx/utils/util.py,sha256=xGR32XDRr_TtruhjnXfI7lEWmk-vgywHAy3kz5SBowc,15725
 lfx/utils/util_strings.py,sha256=nU_IcdphNaj6bAPbjeL-c1cInQPfTBit8mp5Y57lwQk,1686
 lfx/utils/version.py,sha256=cHpbO0OJD2JQAvVaTH_6ibYeFbHJV0QDHs_YXXZ-bT8,671
-lfx_nightly-0.1.12.dev21.dist-info/METADATA,sha256=FFH6xxBJeaIDtfggQ37m3IODz1Z15oVzPy5JHMrBjyA,8068
-lfx_nightly-0.1.12.dev21.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-lfx_nightly-0.1.12.dev21.dist-info/entry_points.txt,sha256=1724p3RHDQRT2CKx_QRzEIa7sFuSVO0Ux70YfXfoMT4,42
-lfx_nightly-0.1.12.dev21.dist-info/RECORD,,
+lfx_nightly-0.1.12.dev22.dist-info/METADATA,sha256=McnMcOwIfzbZLCZ5YxLdQ9m4_uih88lfaWx8j5uF4TQ,8068
+lfx_nightly-0.1.12.dev22.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+lfx_nightly-0.1.12.dev22.dist-info/entry_points.txt,sha256=1724p3RHDQRT2CKx_QRzEIa7sFuSVO0Ux70YfXfoMT4,42
+lfx_nightly-0.1.12.dev22.dist-info/RECORD,,