INSTALLING UNSTRUCTURED FOR LANGCHAIN DOCUMENT LOADERS
Create a new environment:
use a python version less than 3.11!! 
I am using miniconda for virtual environments
conda create -n unstructured python=3.10

activate the environment
conda activate unstructured

Upgrade pip and setuptools:
- pip install --upgrade setuptools
- python.exe -m pip install --upgrade pip 
(if you change your python version these  tools might be downgraded)

if getting a permission error with pip upgrade then try this:
python.exe -m pip install --upgrade pip --user



pip install langchain
pip install openai

Read the installation instruction of these two documents carefully before begining:

https://langchain.readthedocs.io/en/latest/modules/document_loaders/examples/unstructured_file.html (gonna run into errors because GIT, TORCH)

https://github.com/Unstructured-IO/unstructured/blob/main/docs/source/installing.rst

Install Git
https://git-scm.com/



Next step will take a while:

pip install unstructured[local-inference]


if getting numpy errors then:

--- UPGRADE numpy for detectron git install: 

pip install numpy --upgrade

pip install cython

pip3 install torch torchvision torchaudio  (This install only uses the CPU, look at torch's getting started for installing wih GPU support)

very useful link for installing Detectron2:

https://haroonshakeel.medium.com/detectron2-setup-on-windows-10-and-linux-407e5382df1

git clone https://github.com/facebookresearch/detectron2.git
cd detectron2 (go to the detercton2 folder)
pip install -e .   (wih the dot included)
cd.. (go back to the main working directory)
pip install opencv-python


pip install layoutparser[layoutmodels,tesseract]


Install other dependencies:
https://github.com/Unstructured-IO/unstructured/blob/main/docs/source/installing.rst

pip install python-magic (first)

pip install python-magic-bin (second)

Downlaod Poppler:
download 7-zip:
https://www.7-zip.org/
unzip poppler and place it in your working directory
https://blog.alivate.com.au/poppler-windows/
add poppler BIN TO PATH


download and install tesseract:
https://github.com/UB-Mannheim/tesseract/wiki
TAKE NOTE OF THE PATH and add it to your environment path:
C:\Program Files\Tesseract-OCR
pip install pytesseract 


SKIPPED libreoffice, libxml2, libxslt
https://ask.libreoffice.org/t/install-python-package-for-libre-office/66934
https://pypi.org/project/unotools/

Run the following to install NLTK dependencies. unstructured will handle this automatically soon.
python -c "import nltk; nltk.download('punkt')"
python -c "import nltk; nltk.download('averaged_perceptron_tagger')"

restart your VS code and make sure your Unstructured env is active in VS Code by checking python interpreter by clicking ctrl + shift + p