fix: Dependency update for unstructured component (#3552)

* FIX: Dependency update for unstructured component

* [autofix.ci] apply automated fixes

* Add nltk download script

* chore(pyproject.toml): move nltk_setup script definition to [tool.poetry.scripts] section for better organization and readability

* [autofix.ci] apply automated fixes

* Make nltk resource downloading part of run

* [autofix.ci] apply automated fixes

* [autofix.ci] apply automated fixes

* Use logger for nltk messages

* Move the download resources function

* [autofix.ci] apply automated fixes

* Move nltk resource download

* [autofix.ci] apply automated fixes

* Update main.py

* Update poetry lock

* Update main.py

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Gabriel Luiz Freitas Almeida <gabriel@langflow.org>

pyproject.toml

@@ -31,7 +31,7 @@ langflow-base = { path = "./src/backend/base", develop = true }
 beautifulsoup4 = "^4.12.2"
 google-search-results = "^2.4.1"
 google-api-python-client = "^2.130.0"
-huggingface-hub = { version = "^0.22.0", extras = ["inference"] }
+huggingface-hub = { version = "^0.23.2", extras = ["inference"] }
 llama-cpp-python = { version = "~0.2.0", optional = true }
 networkx = "^3.1"
 fake-useragent = "^1.5.0"
@@ -87,7 +87,7 @@ langchain-chroma = "^0.1.1"
 upstash-vector = "^0.4.0"
 gitpython = "^3.1.43"
 cassio = { extras = ["cassio"], version = "^0.1.7", optional = true }
-unstructured = {extras = ["docx", "md", "pptx"], version = "^0.14.4"}
+unstructured = {extras = ["docx", "md", "pptx", "pdf"], version = "^0.14.4"}
 langchain-aws = "^0.1.16"
 langchain-mongodb = "^0.1.6"
 kubernetes = "^30.1.0"
@@ -105,6 +105,7 @@ wolframalpha = "^5.1.3"
 astra-assistants = "^2.1.0.10"
 composio-langchain = "^0.5.8"
 spider-client = "^0.0.27"
+nltk = "^3.9.1"
 
 
 [tool.poetry.group.dev.dependencies]
@@ -151,6 +152,7 @@ optional = true
 [tool.poetry.group.spelling.dependencies]
 codespell = "^2.3.0"
 
+
 [tool.codespell]
 skip = '.git,*.pdf,*.svg,*.pdf,*.yaml,*.ipynb,poetry.lock,*.min.js,*.css,package-lock.json,*.trig.,**/node_modules/**,./stuff/*,*.csv'
 # Ignore latin etc

src/backend/base/langflow/initial_setup/setup.py

@@ -3,6 +3,7 @@ import json
 import os
 import shutil
 import time
+import nltk
 from collections import defaultdict
 from copy import deepcopy
 from datetime import datetime, timezone
@@ -615,3 +616,19 @@ def initialize_super_user_if_needed():
         create_default_folder_if_it_doesnt_exist(session, super_user.id)
         session.commit()
         logger.info("Super user initialized")
+
+
+# Function to download NLTK packages if not already downloaded
+def download_nltk_resources():
+    nltk_resources = {
+        "corpora": ["wordnet"],
+        "taggers": ["averaged_perceptron_tagger"],
+        "tokenizers": ["punkt", "punkt_tab"],
+    }
+
+    for category, packages in nltk_resources.items():
+        for package in packages:
+            try:
+                nltk.data.find(f"{category}/{package}")
+            except LookupError:
+                nltk.download(package)

src/backend/base/langflow/main.py

@@ -25,6 +25,7 @@ from langflow.initial_setup.setup import (
     create_or_update_starter_projects,
     initialize_super_user_if_needed,
     load_flows_from_directory,
+    download_nltk_resources,
 )
 from langflow.interface.types import get_and_cache_all_types_dict
 from langflow.interface.utils import setup_llm_caching
@@ -182,6 +183,9 @@ def create_app():
 
     FastAPIInstrumentor.instrument_app(app)
 
+    # Get necessary NLTK packages
+    download_nltk_resources()
+
     return app