[Python] How do extract text from PDFs including all the spaces

I'm being trying to create some code to extract text from PDFs and putting them into a database, I've been using extract_text() for this. However, for some reason some of the spaces between words are disappearing, (e.g. Australian Shiraz Challenge --> AustralianShirazChallenge) How do I deal with this or what alternative method/in built function should I use instead.

Additionally, I need to be able to adjust my code to be able to extract specific "columns" with too much additional difficulty.


import math
import pdfplumber
import fitz

def find_word_coordinates(pdf_path, search_word):
"""
Find the exact coordinates of a specific word in a PDF file.

aram pdf_path: Path to the PDF file
aram search_word: The word to search for
:return: A dictionary with page numbers as keys and lists of coordinates as values
"""

coordinates = {}

# Open the PDF file
with fitz.open(pdf_path) as doc:
for page_number in range(len(doc)):
page = doc[page_number]

# Search for the word in the page text
text_instances = page.search_for(search_word)

if text_instances:
# Store the coordinates for the page
coordinates = [
inst.x0
for inst in text_instances
]

return coordinates

# Function to extract blurb using keywords from the extracted text
def extract_column(pdf_path,x0,y0,x1,y1):
with pdfplumber.open(pdf_path) as pdf:
for page in pdf.pages:
# Get the dimensions of the page
width, height = page.width, page.height

# Define the crop box for the right-hand side (e.g., the right half of the page)
right_side = (x0, y0, x1, y1) # Coordinates: (x0, y0, x1, y1)


# Crop the page to focus on the right side
cropped_page = page.within_bbox(right_side)
print(page.within_bbox(right_side))

# Extract text from the cropped area
text = cropped_page.extract_text()

return text

def extract_full_text(pdf_path):
with pdfplumber.open(pdf_path) as pdf:
for page in pdf.pages:
width, height = page.width, page.height
name=extract_column(pdf_path,width/4,0,width,height/6);
top=extract_column(pdf_path,width/4,0,width,height/4.8);
column1=extract_column(pdf_path,0,height/4.8,width/5,height);

M=find_word_coordinates(pdf_path, "Winemaking");
N=min(M, key=lambda x:abs(x-width/2));
column2=extract_column(pdf_path,width/5,height/4.8,N-2,height);

P=find_word_coordinates(pdf_path, "Winemaker");
Q=min(P, key=lambda x:abs(x-width*.77));
column3=extract_column(pdf_path,N-2,height/4.8,Q-2,height);



column4=extract_column(pdf_path,Q-2,height/4.8,width,height);

text=[name,top,column1,column2,column3,column4];
return text

Continue reading...