Pipfile and docsLoader

.gitignore

@@ -1 +1,2 @@
-.env
+.env
+/langchain-docs

.vscode/settings.json

@@ -0,0 +1,26 @@
+{
+    "workbench.colorCustomizations": {
+        "activityBar.activeBackground": "#fa1b49",
+        "activityBar.background": "#fa1b49",
+        "activityBar.foreground": "#e7e7e7",
+        "activityBar.inactiveForeground": "#e7e7e799",
+        "activityBarBadge.background": "#155e02",
+        "activityBarBadge.foreground": "#e7e7e7",
+        "commandCenter.border": "#e7e7e799",
+        "sash.hoverBorder": "#fa1b49",
+        "statusBar.background": "#dd0531",
+        "statusBar.foreground": "#e7e7e7",
+        "statusBarItem.hoverBackground": "#fa1b49",
+        "statusBarItem.remoteBackground": "#dd0531",
+        "statusBarItem.remoteForeground": "#e7e7e7",
+        "titleBar.activeBackground": "#dd0531",
+        "titleBar.activeForeground": "#e7e7e7",
+        "titleBar.inactiveBackground": "#dd053199",
+        "titleBar.inactiveForeground": "#e7e7e799"
+    },
+    "peacock.remoteColor": "#dd0531",
+    "[python]": {
+        "editor.defaultFormatter": "ms-python.autopep8"
+    },
+    "python.formatting.provider": "none"
+}

Pipfile

@@ -0,0 +1,24 @@
+[[source]]
+url = "https://pypi.org/simple"
+verify_ssl = true
+name = "pypi"
+
+[packages]
+langchain = "*"
+beautifulsoup4 = "*"
+tiktoken = "*"
+openai = "*"
+pinecone-client = "*"
+unstructured = "*"
+nltk = "*"
+fastapi = "*"
+jinja2 = "*"
+uvicorn = "*"
+streamlit = "*"
+streamlit-chat = "*"
+tqdm = "*"
+
+[dev-packages]
+
+[requires]
+python_version = "3.10"

ingestion.py

@@ -0,0 +1,17 @@
+"""
+This file es responsible for the ingestion of the data (langchain documentation).
+It embedds the data into vectors, and stores it in the pinecone vectorstore.
+"""
+import os
+from langchain.document_loaders import ReadTheDocsLoader
+
+
+def ingest_docs() -> None:
+    # The ReadTheDocsLoader is a class that is in charge of taking the dump of some data
+    # fetching process and loading it into the vectorstore.
+    loader = ReadTheDocsLoader("langchain-docs-chatbot/langchain-docs")
+    raw_documents = loader.load()
+
+
+if __name__ == '__main__':
+    print('Hello world!')