Create ingestion.py

src/ingestion.py

@@ -0,0 +1,43 @@
+import fitz  # PyMuPDF
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+
+def extract_text_from_pdf(file_path: str) -> str:
+    """
+    Extracts text from a PDF file using PyMuPDF.
+    
+    Args:
+        file_path (str): Path to the PDF file.
+    
+    Returns:
+        str: The extracted text from the PDF.
+    """
+    text = ""
+    # Open the PDF
+    with fitz.open(file_path) as pdf:
+        # Loop through each page
+        for page in pdf:
+            # Extract text from that page
+            text += page.get_text()
+    return text
+
+
+
+
+def chunk_text(text: str, chunk_size: int = 500, chunk_overlap: int = 50) -> list:
+    """
+    Splits extracted text into smaller overlapping chunks.
+
+    Args:
+        text (str): The full extracted text.
+        chunk_size (int): Max characters per chunk.
+        chunk_overlap (int): Overlap between chunks.
+
+    Returns:
+        list: List of text chunks.
+    """
+    splitter = RecursiveCharacterTextSplitter(
+        chunk_size=chunk_size,
+        chunk_overlap=chunk_overlap
+    )
+    return splitter.split_text(text)